There's a moment on many projects where somebody asks whether the glazing could look a bit more alive.

It's a reasonable request. We've always thought of the windows as the eyes of a building, and glass in a CGI can come out flat and grey, because that's what glass does on an overcast day. Dead eyes. Nobody is commissioning a picture of an overcast day.

So we lift the reflections a little. Put something in the sky worth reflecting. The building starts to look like the building everyone has in their head.

Is that cheating?

Windows: the eyes of your building ©Blink Image Limited

What Actually Happens

Quite a lot, if we're honest about it.

Materials get tweaked. Surfaces become slightly more reflective, or slightly less. The sun is chosen for the image rather than inherited from the day the photograph was taken. Bins disappear. The tarmac is cleaner than tarmac ever is (and potholes are obviously removed!). Trees are the size they'll be in fifteen years, not the size they'll be on handover.

None of that is unusual and none of it is secret. It's the ordinary craft of making an image that communicates something, and most of it has a defensible reason behind it.

The industry even has a name for the output, and the clue is sitting in it. Artist's impression. An impression, by an artist.

The trouble is that nobody reads it that way. They read it as a photograph of the future.

Make It Flat

Years ago, on a project I won't identify, we put a slight undulation into the reflections across a glazed elevation. Real glass is never perfectly flat, so the reflections bend and ripple a little, and we had taken the trouble to show it.

The draft went out. The client came back and asked why the glass looked bumpy and distorted.

We explained that this is simply what glass does. The reply was that we were being too clever, and could we make it flat.

So we made it flat.

Not the project referred to above, but another one with complicated reflections. Initially, even the architects themselves weren't totally convinced that the reflections would actually look like this when we created the CGI.

The same client, on another view of the same scheme, asked us to take out the swans we had put on the river. Too much of a cliché, apparently. So the swans went as well.

I still think about those two notes, because they pull in opposite directions. One image was too accurate. The other was too picturesque. Nobody was trying to mislead anyone in either case.

Taken together, they're a fair description of the job. Not what's true, and not what's pretty. What reads as true.

The Easy Half

For planning work this is largely settled, and I've written about it here before. An AVR is a technical document rather than a creative one. It's produced so that somebody who would quite like to prove you wrong can check the camera position, the lens and the geometry, and if it can't be checked it isn't doing its job. (They've been a core part of what we do for the best part of thirty years, which I mention mostly so you know where I'm standing.)

The awkward half is everything else. The marketing image, the consultation board, the investor deck, the shot that ends up ten feet wide on a hoarding. No survey data, no agreed standard, no definition of too far. Just judgement.

Images move. One made to sell a scheme has a habit of turning up later on a consultation board, because it's the best one anybody has. At that point an illustration of intent is doing the work of evidence, in front of people who will stand on that spot and compare. That's usually how trust goes, and almost never because somebody set out to mislead. The image may have been honest in its original context. The problem starts when it's used in a different one.

Where I Draw It

I should be straight about this, because a list of things we would never do would be a bit precious.

We get asked for plenty. Put a north-facing elevation in full sun. Choose the one viewpoint where the service yard disappears. Ease the neighbouring block back a little so the scheme reads more generously.

We push back on those, and that conversation is a normal part of the week.

But the line isn't composition, because composition is the work. Choosing the hour, the weather, the season, the activity, what sits in the foreground and where the eye goes first: a photographer makes every one of those decisions before the shutter opens, and nobody accuses them of lying for waiting until six in the evening.

The line is a claim. A north-facing elevation in full sun isn't a composition choice, it's a statement about the building that isn't true, and the person standing there in June will work that out. Losing a storey, shrinking a mass, deleting a neighbour's house from a view they live with: same category.

Everything else is interpretation, and interpretation is the job.

The Test

A 3D model doesn't produce an image. It produces geometry. Everything between that and something a person believes is judgement: light, weather, distance, people, attention and restraint.

The question I find useful isn't whether an image flatters. They all flatter. Nobody has ever commissioned an unflattering one.

It's this. When the building is finished, and somebody is standing on the spot the image was taken from, would they feel misled?

If the answer is no, the reflections were probably fine.

Until next week.

Rich

PS: 25 years on, I still can't see a pair (or squadron!) of swans on a river without hearing that note. "You're such a cliché," I tell them. They have never once looked bothered.

Cliché. Be careful though - they can break your arm!

I'm writing this in a bakery.

The bakery is in Unit 68 of the Sandford Lane Industrial Estate in Kennington, just south of Oxford, tucked between a railway line and a field. It's called Proof Social Bakehouse, and it's a social enterprise that employs people in prison and prison leavers.

It's nine o'clock on a Friday morning. Around me there are families with toddlers, dog walkers, cyclists and a few people working on laptops. At weekends it's busier still.

Proof Social Bakehouse - Friday morning: young families, kids, toddlers, business-types, dogs, cyclists, walkers…and me!

Look at who else is here. A recording studio. An aviation company. Pharma engineering. A classic motorcycle dealer and a classic car specialist. A printer. Two garages. Engineers.

Not a chimney in sight.

Well. There is one, now I think about it. It's above the ovens behind me, and it smells of bread.

It Isn't a Showpiece

I should be honest about the place. It's tired - externally I mean. The tarmac has been patched more than once, the units have been here a long time, and nothing about it is going in anyone's brochure. It isn't the sort of place that usually gets treated as marketing material, unless somebody is planning a refurbishment.

And yet.

From where I'm sitting I can't see an empty unit. There are cars everywhere. Someone has put solar panels on one of the roofs. The bakery doesn't offer parking during the week, because the businesses here need the spaces.

Nobody ever produced a CGI of this estate. It has simply been quietly full for decades.

Drilling and Croissants

I can hear drilling from somewhere behind me. A compressor too. Someone is doing something noisy to a vehicle. Nobody in here seems to mind. They came for the coffee and the pastries, and they seem perfectly happy sitting in the middle of a working estate while it works. The radio in here is playing eighties hits. Some of the units look older than that.

A good number of them cycled in, along the river path from the city, or on one of the routes that follow a disused railway line.

That's a small piece of evidence for something developers spend chapters of a planning statement trying to prove. Active travel links, green routes, connections through to the next place along. Those sections attract a certain amount of eye-rolling, not least because they tend to be written for officers rather than for people. And yet here is a queue of people who have walked or cycled to an industrial estate on a weekday morning, because there's something here worth arriving at.

This is roughly what a lot of modern schemes are trying to engineer. Amenity space. Somewhere for staff to go. A reason to be on site that isn't purely functional. It costs real money to design in, and it's often the first thing to be value-engineered out.

Sandford Lane got there by accident, with one good tenant and a queue at the counter.

What Flexibility Actually Looks Like

Every industrial brochure I've ever read promises flexible accommodation suitable for a wide range of occupiers.

Sandford Lane is what that sentence looks like forty years later, after the market has had a proper go at it. Nobody planned for a sourdough bakery to end up next to an aviation business. The building didn't care. It had a roof, a door big enough, power and a yard, and the rest sorted itself out.

These units were built for one idea of what industry was, and they've been quietly reinterpreted ever since. At some point a landlord here said yes to a bakery, which can't have been an obvious decision at the time.

That's worth remembering when a scheme is designed around one named occupier, or marketed for a single use. The buildings that last are the ones that turn out to suit tenants nobody had thought of yet.

Sandford Lane Industrial Estate in Kennington

The Refurbishment Problem

This is where we do get involved with estates like this one, and it's a harder brief than it looks.

Refurbishment imagery has to do something new-build imagery never has to. It has to show what a place could become, while being honest about what it is now, and without quietly deleting the businesses currently working there. Show it too clean and nobody believes you. Show it as it stands and you haven't answered the question.

The useful version usually sits somewhere in between: the same estate, working better. The same yards, made clearer. The same buildings improved, without pretending the life already there has disappeared. Most of it happens in phases, around people who are still trading, which is a brief of its own.

Back to That Word

A couple of weeks ago I wrote something on LinkedIn about the word industrial, and how the picture it puts in a neighbour's head is chimneys and heavy plant.

If a planning notice went up on Sandford Lane tomorrow, that's roughly what people would imagine. Not a recording studio. Not aviation. Certainly not a bakery full of pushchairs on a Saturday morning.

The estate has been here for decades, proving the opposite every working day, and the word still hasn't caught up.

Maybe that's the missing piece. Not better images of the building being proposed, but better examples of places that already work. If what people fear is chimneys and heavy plant, showing them a bakery, a recording studio and an aviation business on an ordinary working estate might do more than another paragraph about flexible employment space.

I'm not sure what the better word is. But if you're trying to convince a community that a scheme can bring something useful to an area, a place like this may do more work than another aerial of an empty yard.

Until next week.

Rich

PS: If you do go, the pastries are the point. I've done the research on that too. (FYI They are a bit more expensive than your average, but so much more than average in terms of quality/taste/ingredients!)

A chimney that smells of croissants. On reflection, it actually looks a bit like a croissant. Intentional?

This morning I found myself photographing sausages in my local Tesco.

Also pizzas, ibuprofen, yoghurt and grated cheese. Always in pairs: an entry-level version on one side, and what I'd call the premium one on the other. I got a few looks. I'd like to think they were admiring the composition.

It was the prices I was after, and they're quite something.

Same Pills, Different Box

Tesco's own-brand ibuprofen was 35p. Nurofen, on the shelf next to it, was £1.90.

Same active ingredient. Same dose. More than five times the price.

And plenty of us pick up the box on the right anyway. Not because anyone thinks Tesco has found a cheaper kind of ibuprofen, but because the name and the familiar packaging make it feel more trustworthy. Whole industries exist because that works.

There's a footnote, though. About ten years ago Nurofen sold Back Pain, Migraine and Period Pain versions in Australia, with exactly the same ibuprofen in every box. It cost them A$6 million. Packaging that claims a difference that isn't there tends to get caught in the end.

The Sausage Is Different

Sausages are another story. The entry-level pack was £1.79. Porky White were £3.80, and Heck were £5.

Here the premium is usually real. More meat, better meat, fewer things you'd rather not know about. Same with the pizza: £1.85 for the basic cheese one, £6.25 for Crosta & Mollica. Eat them side by side and you'd know which was which with your eyes shut.

But you can't eat them in the aisle. On the shelf, all you've got to go on is the box. So the premium brand's packaging isn't inventing a difference. It's making sure you notice one.

That's the version that matters for property.

The Other Way Round

Nobody opens a packet of ibuprofen before they buy it. An occupier or an investor absolutely opens the box. There are site visits, surveys, agents and months of due diligence. You can't sell a building on packaging alone, and nobody sensible would try.

In property, the more common problem is the opposite. A well-specified scheme, with better cladding, a more generous yard or landscaping someone has thought hard about, shown to the market through a perfectly adequate image straight from the architect's model. Nothing wrong with it. It just doesn't tell anyone what they're looking at.

A premium sausage in an own-brand wrapper.

Last week I wrote about finding what really makes a scheme different. This is the other half of it: once you know, telling people.

Now With Added Protein

The cheese aisle was my favourite. Creamfields grated mature cheddar was £1. Cathedral City was £1.89, and Cathedral City's high-protein shredded cheddar was £2.55.

Cheddar has always been one of the most protein-rich things in the fridge. It just didn't used to say so on the front. Now protein is what people are scanning the shelves for, so there it is, in large letters.

Property has its own protein. At the moment it might be an EPC A rating, net zero, serious power capacity, or a building designed around the wellbeing of the people working in it. Whatever occupiers are looking for this year, if a scheme genuinely has it, it belongs on the front of the pack, not on page fourteen of the specification.

And if it hasn't, see the Nurofen footnote.

Shredded?

Even Tesco Does Both

The yoghurt is the bit I keep coming back to. Creamfields Greek-style yoghurt was 95p. Tesco Finest was £2.50.

It turns out both are Tesco's. Creamfields is one of its budget brands, sold only in Tesco, with a name that sounds rather more like a farm than a supermarket. Finest is the top of the range. Same shop, two completely different identities, and Tesco knows exactly which pot each product belongs in.

That's a fair model for a development pipeline. Not every scheme needs the premium wrapper, and the own-brand approach is the right answer for plenty of projects. The aim isn't to make everything look premium. It's for the packaging to match what's inside. Plain is fine when the product is plain.

The expensive mistake is the good scheme that ends up in the plain tub by default.

So, what sort of sausage is your development?

If you think you've got a Heck in an own-brand wrapper, I'm always happy to take a look.

Until next week.

Rich

PS: The crème caramels didn't make the cut. Tesco's were £1.50 and Bonne Maman's were £3.20. I bought both, purely for research. The research is ongoing…

Apple launched the iPhone 18 Pro this week.

I've got an iPhone 15 Pro Max, or possibly it's a 14. I genuinely need to check, which probably tells you most of what you need to know.

It's more than good enough for everything I do with it. The 18 is faster, the camera's better and the battery lasts longer, and I don't doubt any of that. But put the two side by side and I'm not sure I could tell you which was which. Nothing about it makes me want to switch.

What Apple also launched was the iPhone Duo, their first foldable. It's very expensive and it's genuinely different, and it's the one I'll probably end up buying. Not because my current phone has stopped working. Because it is a different thing.

I've been thinking all week about how much that sounds like an occupier looking at a new scheme.

https://www.youtube.com/live/39BalPDuTo0?si=49i2oYINwi2EUTn2

Why Would Anyone Move?

The occupier you're trying to attract already has a building. It's probably fine. It works, the staff know where it is, and moving is expensive and disruptive in ways that make changing a phone look trivial.

Doing nothing is always an option, and it's usually the comfortable one. All the cost, disruption and risk sit on one side of the decision. Whatever sits on the other side has to be specific enough to be worth it.

A box that's slightly taller, slightly newer and slightly better specified is the iPhone 18. Better in every measurable way, and still not quite a reason to go.

Which is a problem, because most sheds are, at heart, the same. Similar spans, similar eaves, similar yards, similar cladding. That is not a criticism. It's what a well-designed logistics building looks like, and it's why they work.

But it does mean the building itself is rarely what wins the argument.

Take the Logos Off

Here's a test worth trying. Take the hero images from five competing schemes in the same region, cover the logos, and lay them side by side.

Could an occupier tell which was which?

In my experience, usually not. The same camera height. The same corner of the yard. The same late-afternoon sun. The same three HGVs reversing onto the same docks.

I should say straight away that we make a lot of these, and I'm not about to argue against them. Every scheme needs one. A good hero image is the iPhone 18 of property marketing: expected, well made, and the thing the market assumes you'll have.

The problem is not that it exists. It's when it's the only thing. Because whatever every scheme has can't be the thing that makes yours different.

Hero images do a critical job. Here's an example created for Tritax. ©Blink Image Limited

Then the Giraffes Arrived

A few weeks ago I wrote about a lion in a warehouse. Since then I've started seeing giraffes.

Mostly on LinkedIn, added into warehouse interiors with AI, presumably to make an advert stand out. I rather like them. It's exactly the kind of ambition I was arguing for.

But the lion worked for two reasons, and it's worth being honest about both. In the 1990s it was hard to do, so nobody else had one. And it said something about the building: scale, power, presence.

A giraffe now takes about as long as making a coffee. Everyone can have one. Which means it risks becoming the next hero shot. And it has to answer the same question the lion did. What is it saying about your scheme?

Cover the logo again. You'll remember the giraffe. Will you remember whose warehouse it was in?

What's Your Duo?

Every scheme has one or two things that set it apart, and they're rarely the building.

It might be a motorway junction two minutes from the gate. It might be a power supply the scheme down the road can't get for years. It might be a yard deep enough for the trailers an operator actually runs, a bus route that gets shift workers in for six in the morning, or a site that sits comfortably above the flood plain. Usually dull. Almost always the real reason a deal gets done.

And almost never in the imagery, because they're hard to picture. How do you photograph a power supply?

That's where the effort and the invention should go.

Point the Model at It

Every scheme has one or two things that set it apart, and they're rarely the building.

It might be a motorway junction two minutes from the gate. It might be a power supply the scheme down the road can't get for years. It might be a yard deep enough for the trailers an operator actually runs, a bus route that gets shift workers in for six in the morning, or a site that sits comfortably above the flood plain. Often overlooked. Almost always the real reason a deal gets done.

And almost never in the imagery, because they're hard to picture. How do you make a power supply look like a reason to move?

That's where it gets interesting.

Currently a construction site, prospective occupiers wanted to know what the view from their desk would be like when complete. ©Blink Image Limited

Outside the Box

The 3D model built for the hero image can do far more than produce one view, as I wrote a few weeks ago. It can just as easily look beyond the building.

A film that follows a trailer from the slip road to the dock and lets the clock run. An aerial that deliberately puts the substation in frame rather than cropping it out. The gate at shift change, with the bus pulling in. None of these replace the hero image. They sit alongside it and show what it can't.

It does take both sides, though. We can't find a scheme's Duo on our own. The developer and the agents know what makes it different, usually far better than we do. What a visualisation team brings is the lateral thinking: spotting which of those overlooked things could help sell the scheme, and working out how to tease them out.

Better, or Different?

Most schemes will always need their iPhone 18, and it should be a very good one.

But better on its own rarely moves anyone. Different does. Somewhere in most schemes there's a Duo, sitting in a spec sheet or in an agent's head. Whether anyone looking at the marketing could spot it is another matter.

If you know what makes a scheme different and the imagery isn't showing it, that's a conversation I'd enjoy having.

Until next week.

Rich

PS: I'll let you know how I get on with the Duo. Assuming I can justify the price to myself first.

Tesla has just launched the Cybercab in Austin. Two seats, no steering wheel, no pedals.

The self-driving isn't the news. Driverless cars have been circling Austin for the best part of a year. What's new is that Tesla has removed the ability to intervene. Not switched it off. Actually removed it, physically, from the cabin.

I've been thinking about that empty footwell all week, because the same removal is happening on property projects right now. It just doesn't come with a launch event in Texas.

No wheel. No pedals. Screenshot of recording from ‘Brighter with Herbert’ YouTube channel.

Your Projects Already Have One

Every scheme has an override in it somewhere, and it almost never looks like one.

It is the draft round where somebody says the brick reads too warm, and they are right, and the material gets changed before anyone outside the project ever sees it. It is the person who asks whether that is honestly the view a planning officer will stand in, or just the one that composed nicely. It is the quiet half hour before something goes into a pack, when a person who knows the scheme looks at it properly.

None of that is glamorous. None of it appears on a programme. It's the least interesting part of the process to describe and the most expensive part to lose.

And it's always the first thing to go. Not through any decision to remove it. It goes because the deadline moved, or because the images arrived so quickly that checking them felt like an insult to the software.

Tesla Took Theirs Out Last

Here's the part I think is genuinely instructive, and it holds regardless of what you make of the company.

Tesla didn't start with an empty cabin. The sequence, publicly at least, ran through years of supervised driving in live traffic before anything was taken away. The wheel came out late, at the end of the evidence rather than at the beginning of the ambition.

That order is doing all the work. There's a large difference between removing a safeguard because you have established you no longer need it, and removing it because it was slowing you down.

Most adoption of AI in property is currently running that sequence backwards. I don't say that as much of a criticism. It's what happens when a technology arrives faster than anyone's ability to test it. But it's worth noticing when you're doing it, because the two look identical from the outside and only one of them is earned.

Where I Want No Override At All

I should be clear that plenty of our own work should have the wheel taken out, and I'd take it out tomorrow.

Camera matching is the obvious one. Lining a virtual camera up with a photograph, so that a building sits in a real street at the right scale and the right angle, is slow and fiddly and it matters enormously. Get it wrong and nothing else in the image can rescue it.

But there's a right answer, and you can see whether you've hit it.

That's the useful distinction, and it isn't about importance. Cleaning up a model that arrived with a hundred thousand duplicate faces. Upscaling. Removing noise from a process that used to take hours. None of that is trivial work, and all of it can be checked against something.

The tasks that are safe to automate aren't the unimportant ones. They're the ones where you can tell if it has gone wrong.

So this isn't an argument for keeping a hand on the wheel at all times. Quite a lot of the driving should be automated, and I'd rather spend the hours that frees up on the decisions that actually determine whether an image works.

Why Images Are the Awkward Case

The difficulty is that a CGI is not a record. It is a claim.

That matters because people make decisions from these images. Planning decisions, funding decisions, leasing decisions, and the reaction of a room full of residents. If an image makes a claim about material, massing, visibility or impact, somebody needs to be able to say where that claim came from and who checked it.

An image of a building that exists can be checked. Someone can stand in the street and compare. An image of a building that doesn't exist can't be checked against anything, because the thing it depicts hasn't happened yet. The only check available is a person who knows what was actually specified.

Which is exactly the check that gets removed.

So the failure mode here isn't ugly pictures. Ugly pictures are obvious and somebody fixes them. The failure mode is a claim nobody made on purpose. A material that reads warmer than the one specified. A massing that sits more politely against its neighbour than it will. A view chosen because it composed well rather than because it was the honest one.

None of those look like errors. They look like good images. The real risk was never that AI makes an image wrong. It's that it makes an image plausible enough that nobody thinks to ask who checked it.

George Street in Oxford: a verified view for planning on the left, a marketing image on the right. ©Blink Image Limited

The Question Worth Asking

I don't think the useful question is whether to use AI. Everybody is, ourselves included, and the people who say otherwise have usually stopped noticing the tools they already rely on.

The better question is the Cybercab question. Where is the override, and have we earned the right to take it out?

That's worth asking beyond imagery too. The planning note drafted quickly. The appraisal pulled from a model. The visual pack that arrived suspiciously fast and suspiciously cheaply. In each case something was automated, and in each case the question is the same: can anyone still say what the check used to be, and who was doing it?

Because if you can't name what the override was, you haven't automated a decision. You have quietly stopped making one.

If the imagery on a scheme needs to hold up to more than a first glance, it's worth knowing where the checks sit before they disappear. That's usually a short conversation and a useful one.

That's where my head was at this week.

Rich

PS: Can you imagine self driving cars in the UK? I still struggle with the idea, but I would love to see one in the flesh.

WARNING: this newsletter seems to have ended up a bit long. Sorry about that. But stick with - hopefully it’s useful!

The most common question we get asked is how long something will take.

The second most common is really the same question, just arriving slightly differently:

“We have a planning submission next week and need images.”

Or:

“We are in front of agents a week on Tuesday.”

We can usually help. That is not the interesting part.

The interesting part is what gets lost when the process is compressed, because it is rarely the thing people expect.

It is not always the quality of the final render.

It is the chance to change your mind.

The Honest Timing

A set of CGIs typically takes around four weeks.

That is not four weeks of rendering. It is four weeks that allows for the model to be built, views to be agreed, drafts to be issued, comments to come back, revisions to be made, and everyone who needs to review the images to do so properly.

We have turned images around in a week.

Sometimes less.

It is possible.

Aldgate Central House: 1 of 3 CGIs created in a week. Nearly killed us! ©Blink Image Limited

Animation and film usually need longer, because there are more decisions to agree: storyboard, pacing, camera movement, message, sequence, voiceover, music and review.

But the principle is the same.

The less time you leave, the fewer chances there are to improve the work before it has to be final.

Faster is often available.

It just costs you something other than money.

Drafts Are Where the Work Improves

Two rounds of drafts are our usual baseline, and they are not a courtesy.

They are where the project gets better.

Example Tritax project showing 3 drafts: some subtle and some not-so-subtle improvements as the project develops. ©Blink Image Limited

The first draft is the point at which everyone involved sees the same thing at the same time.

• The developer.

• The architect.

• The agent.

• The marketing team.

• The landscape architect.

• Us.

Sometimes the draft confirms the brief.

Quite often, it changes it.

• A material does not read as expected.

• The chosen view is not quite doing the job.

• The building feels heavier than anticipated.

• The foreground is distracting.

• The landscape hides something important.

• The image answers the wrong question.

None of that is failure.

It is the normal business of turning drawings, assumptions and intent into something visible.

But when the programme is compressed, drafts are usually the first thing to suffer.

You still get the image.

You just get something much closer to the version everyone guessed at in the first meeting.

Nobody Knows What It Looks Like Until It Exists

This is the part worth sitting with.

None of us really knows what a scheme will look like until the work is done.

• Not the architect.

• Not the developer.

• Not the marketing agency.

• Not us. (Yep, even little ole’ us!)

Case in point: 61 Oxford Street. I don't think anybody quite understood what this would look like until the image was created. ©Blink Image Limited

Everyone has a strong idea of it. Often a very informed one.

But it is still an idea.

It is only when the building is modelled, lit, viewed and placed in context that the specific issues surface.

A colour behaves differently at scale than it did on a sample board.

A façade that looked balanced in elevation feels overbearing from the one angle that matters.

A planting strategy that works beautifully in plan quietly hides the thing the image was commissioned to show.

A view that seemed obvious turns out to be adequate, while the better one sits a few metres to the left.

That is why review time matters.

Because if you brief at the last minute, you are betting that the design, the view and the brief were all right first time.

They rarely are.

Too Early Can Cost You Too

The obvious conclusion is to start earlier.

Sometimes that is absolutely right.

If there is a board meeting, agent presentation, consultation, planning submission or design decision coming up, early visualisation can be extremely useful. The imagery has a job to do, and the review time has value.

But “earlier” is not always better.

If you bring a visualisation team in while the design is still likely to change substantially, and there is no decision or meeting in the meantime that needs images, you may simply be paying to build something before it is ready.

We can absorb a certain amount of change. We build a reasonable allowance for it into what we quote.

But revisions are not free, and we would rather be upfront about that.

A radical design change, a wholesale shift of viewpoints, or a change of mind about what the images are for can mean genuine rebuilding.

So the useful distinction is this:

Review cycles improve the work.

Revision cycles rebuild work that was based on information that changed.

Both have their place.

Only one reliably makes the project better.

The Useful Test

The question is not:

“How early can we start?”

The better question is:

“Is there something happening between now and the deadline that these images need to serve?”

If there is, then early may be genuinely useful.

• A board meeting.

• A consultation.

• An agent presentation.

• A planning submission.

• A material decision.

• A design option that needs testing.

In those cases, visualisation can help people make better decisions while there is still time to act on them.

If the honest answer is “no, we just wanted the images in hand”, it may be worth waiting until the design has settled.

Otherwise you may be buying revision cycles rather than review cycles.

And those are not the same thing.

A Different Kind of Timing

We have been brought onto projects at every conceivable stage.

At one end, everything is locked and coordinated, ready for a detailed planning submission.

At the other, there is barely a design at all.

Both can be right.

They are just answering different questions.

We once worked on a rooftop extension off one of London’s main streets where the client did not initially need an image for marketing or planning.

They needed to know the precise volume within which the architect could design without the extension becoming visible from street level.

Can you see it? Providing data that ensures your design is invisible.

So we helped calculate that envelope, and the architect designed inside it.

That was visualisation being used as a design constraint rather than a marketing output.

It only worked because we were involved early enough for the answer to be useful.

That is the point.

The question is not really:

“When should we commission visualisation?”

It is:

“What do we need visualisation to answer, and when does that answer need to exist?”

Some Things Have Their Own Clock

A few parts of the process do not care about your deadline.

Views and compositions are one of them.

On a plan, a view can look perfectly reasonable. It is only when alternatives are tested that the better option often appears. Rush that stage and you may end up with something adequate, which is a quiet and expensive kind of failure.

We rushed to get the photography (left) due to an apparent deadline but it was worth retaking to improve the image (right). ©Blink Image Limited

Photography has its own clock too.

For anything involving real context, photography is doing at least half the work. Given time, we shoot it properly: the right position, the right light, the right conditions.

Given no time, we work with whatever exists.

We can improve existing photography.

We cannot change the weather it was shot in.

Seasons matter as well.

Bare trees in February are not the same as full canopy in May. If greenery matters to the scheme, that is a fixed date in the calendar, not a preference.

And then there is the campaign itself.

A few weeks ago, in something like thirty degrees, we were putting together proposals for clients’ Christmas presentations.

It felt slightly absurd.

It was also correct.

Three Questions Worth Asking

So, rather than simply asking when to commission visualisation, it is worth asking three more useful questions.

First: what does the imagery need to achieve?

Planning, marketing, investor confidence, agent conversations, public consultation and design testing all have different requirements. The right timing depends on the job.

Second: when does that answer need to exist?

Not when would it be nice to have the image, but when does someone need to make a decision based on it?

Third: how much is still likely to change?

If the design is still moving significantly, starting too early may mean paying to rebuild. If the design is mostly settled but the deadline is close, starting too late may remove the chance to improve the work properly.

Too early can cost money.

Too late can cost options.

The useful moment is somewhere between the two.

Our advice is not “commission earlier”.

That is the sort of advice a supplier gives, and it is only true some of the time.

The better advice is to work backwards from the point at which the imagery needs to do something useful.

Add time for review.

Add time for the things with their own clock: photography, seasons, view testing, campaign dates.

Then commission.

That will sometimes land earlier than expected.

Occasionally it may land later than expected, because the design was not ready and there was nothing in the diary that genuinely needed the images yet.

Both can be right.

Nearly thirty years of doing this means we can usually see some of the problems before they arrive, and help steer around them.

But it does not create time that has already gone.

If you have a scheme coming up and are not sure when the visualisation should start, that is usually a short conversation and a useful one.

Happy to talk it through.

richard.birket@blinkimage.com

Every so often a client asks how long an image will take, and there is a pause when we answer.
Not because the number is outrageous.
Because somewhere in the back of most people’s minds sits the idea that the model goes in, a button gets pressed, and the picture comes out.
There is no render button.
Or rather, there is.
But pressing it is much closer to the end of the process than the beginning. And what comes out of it is not the finished image you eventually receive.
That is not a complaint. Clients do not need to understand every technical step, in the same way they do not need to understand a concrete pour to commission a building.
But there is a practical reason to understand the bones of the process.
People who understand what visualisation can actually do tend to brief it better.
And better briefs usually produce better images.

We Build the Building First
The most useful way to think about an architectural visualisation team is not as a photographer.
It is closer to a contractor.
Before anyone can “take a photograph” of your scheme, someone has to build it.
Not the real building, of course, but a detailed three-dimensional version of it: walls, glazing, cladding, junctions, landscape, context, materials and lighting.
That is where much of the work sits.
It is construction of a different kind.
And, like construction, it happens in a sequence.
That also explains something that often surprises people.
When the design changes, we do not simply adjust the image.
We go back into the model, rebuild the part that has changed, update the materials or geometry, and then produce the image again.
A revision that sounds small on a drawing - a changed cladding module, a shifted door, a reworked entrance canopy - can mean genuine reconstruction before anything can be re-rendered.
That is not a problem.
It is just worth knowing when you are planning a programme.

A Design Model Is Not Always a Visualisation Model
Architects build excellent models.
But they are usually built for design, coordination and documentation.
They are not necessarily built to be photographed.
That distinction matters.
A design model is often asking:
Does this work?
A visualisation model is asking something slightly different:
What does this look like from here, in this light, at this time of year, to someone who has never seen the drawings?
Those are not the same question.
We can often start from the architect’s model.
We can rarely finish with it.
To create a convincing image, the model usually needs cleaning, optimising, refining and rebuilding in places. Materials need to behave properly. Edges need to resolve. Context needs to be added. Landscape needs to feel grown rather than placed.
If the model is poor, everything downstream inherits the problem.
Lighting can help.
Post-production can help.
A good camera angle can help.
But none of them can fully rescue a model that was never properly built for the job.

The Image Is a Decision, Not a Default
Once the model exists, you can point a camera almost anywhere.
That sounds like freedom.
In practice, it is one of the most important decisions in the whole process.
Who is looking at this?
And what do they need to understand?
A planning officer, an investor, a prospective occupier and a local resident may all be looking at the same building, but they are not asking the same question.
A planning image may need to show scale, context and impact.
A marketing image may need to show ambition, quality and appeal.
An occupier image may need to explain access, operation and suitability.
A community image may need to make change understandable and less abstract.
The building has not changed.
The audience has.
That is why the brief matters more than people often expect.
Not just the technical brief.
The intent.

Once the Model Exists, Use It Properly
This is the part that is still under-used.
Once the model has been built, the constraints of real photography no longer apply in quite the same way.
You are not waiting for the right weather.
You are not limited to the current season.
You do not need the building to be finished.
You can test different times of day.
You can show the landscape at maturity.
You can make the interior visible through glazing.
You can add activity, people, vehicles and occupation.
You can compare different materials or tenant scenarios.
You can create a film, animation, cutaway or interactive model from the same underlying work.
None of this is trickery.
It is the ordinary consequence of having built the thing digitally rather than found it in the real world.
And it means the model can do more than produce a single still image.
It can help explain how a building works, not just what it looks like.

Three Things Worth Thinking About
So, before you brief your next set of images, it is worth asking three questions.
First: are we treating the CGI as a picture, or as a model we can use?
If you only need one image, that is fine. Sometimes that is exactly the right answer.
But if the model is already being built, there may be more value in it than a single view: alternative materials, extra angles, animation, social edits, cutaways, occupier-specific versions, or future marketing assets.
Thinking about that early usually gives you more options.
Second: who does this image need to convince?
Not everyone needs the same thing.
A planning officer, investor, agent, occupier and local resident may all need to understand the same development in different ways. If you brief one image to satisfy all of them, it may end up doing none of those jobs particularly well.
The clearer the audience, the better the output.
Third: what will we wish we had asked for later?
This is where the best teams get ahead.
They do not just commission the minimum image needed for the immediate deadline. They think about the life of the project.
Planning.
Marketing.
Investor conversations.
Occupier interest.
Public consultation.
Social content.
Launch material.
The schemes that make best use of visualisation tend to think about those needs before the first render starts, not after the model has already been built and everyone is trying to retrofit extra value into it.

The Point
The idea of a render button is misleading.
Not because it undersells the work.
Because it makes visualisation sound like an output, when really it is a sequence of decisions.
What needs to be built?
Who is it for?
What does the viewer need to understand?
What else could the model help us communicate?
If you know a model is being built, you may brief it differently.
If you know a change means reconstruction, you may plan the programme differently.
And if you know what becomes possible once the model exists, you may ask for something more useful than a standard external view.
That is where the opportunity sits.
Not just in getting a good image, but in using the process intelligently enough to get more value from it than the next scheme does.
If you have a project where you are not sure what the imagery actually needs to achieve, that is usually a conversation worth having early rather than late.
Happy to talk it through.

richard.birket@blinkimage.com

Detail from one of our recent images - basically Bridgerton levels of intrigue here

Put figures into a scene in our studio and within about a minute the conversation has left the architecture entirely. Who in the image is saying what to whom. Who's arguing. Who's flirting. Which two are in a love triangle with the bloke by the bike rack.

It's a joke, and it's the same joke every time. But it only works because every one of us is doing the reading involuntarily. Posture, distance, where someone's weight is sitting - we can't not interpret it. Which makes it a useful reminder, because the person opening your brochure or working through your planning submission is doing precisely the same thing, with no less conviction and considerably less humour.

Which raises the question that actually costs money. A story is being told either way. Is it the one you commissioned?

Film has never needed convincing on this point. Where the camera sits, who's in shot, what the light is doing, what's been left just outside the frame - all of it is understood to be a decision that carries meaning. Nobody shoots a scene and calls it finished on the grounds that it looked convincing.

Visualisations get treated differently, and I've never really understood why. It’s a single frame doing the work that film spreads across a whole sequence. That's harder, not easier. And it's exactly where the line falls between a pretty picture and a visualisation: one is a model that has been rendered accurately. The other has been directed.

Which brings me to the thing our industry gets wrong most often.

People aren't set dressing

Figures aren't in an image to prove the space gets used. Buildings are made for people - they're a framework for human activity - so what the people are doing is the closest thing you have to showing what the building is actually for.

Which means the demographic matters. What they're wearing matters. Where they gather, and where they deliberately don't, matters.

A hospitality shot with nobody in it says nothing about the buzz of the place at seven on a Friday. A plaza shot with people distributed evenly across it says “Weekly fire drill.” Crowds milling through a space designed for calm quietly undo the thing you're selling. Get it right and the same figures say: relaxed, well used, a business thriving in this location.

People do other work too - drawing the eye to the entrance, marking out the amenity space, showing the new route where somebody will walk their dog.

Proposals for Clarendon Square, Central Oxford

Which is really about direction

If you haven't decided where you want the viewer to look, they won't look anywhere in particular. No hierarchy, no engagement. They get bored - or worse, they engage hard with the wrong thing and leave with a message nobody intended.

Some direction is heavy-handed, and should be. We produced a plaza shot with a full Christmas market: reindeer, braziers, snow, the lot. Cinematic, unashamedly.

Grantham Designer Outlet. You can feel the Christmas chill

Most of it isn't. Cloud shadow falling so the new building sits in light. A little more shine on the new fabric than on the existing around it, so you know without being told which is the new thing. Vehicles arranged so the circulation reads. The surrounding development dropped back a touch in intensity so the eye goes where it should.

Camera height does more than any of it. High up and you're in overview mode - detached, diagrammatic, thinking about function. Drop to eye level and it becomes immersive. Put a single figure at the centre and the viewer feels the space through them.

The readings you didn't ask for

The same wiring works against you.

A scheme shown at dusk, lit up, full of activity can read to local residents as a workplace that never stops - an objection handed over for free. Too many cars on the approach and the brain files it as congestion, so connectivity becomes the problem you were trying to argue away. Leaves across the paving and planting spilling over the edges, and a potential occupier concludes - without ever consciously forming the thought - that the landlord isn't on top of things.

That's why I'm exacting about landscape looking considered and well maintained. It isn't tidiness for its own sake. It's a statement about how the building will be run.

A picture of a diligent landlord

Why the industry is behind on this

For most of the history of archviz, the battle was making things look real. It took a disproportionate share of the time and it was genuinely hard. That's largely solved now - and the habit has outlived the problem. Photorealism is still treated as the goal when it was only ever the entry ticket, and the storytelling layers get skipped.

Meanwhile, the part of a viewer's brain that reads people, light, tidiness and hierarchy isn't going to give anyone a break for having spent the budget elsewhere.

The awkward part

When all of this is working, nobody notices. It's processed below the surface and the response is simply that it's a good image.

Which is why we point it out. Otherwise the assumption is that we rendered the architect's model and it all came together on its own. It didn't. It came together because we worked out what story needed telling - and then built every decision in the frame around it.

Somewhere in the distant corners of my memory is a story about a lion in a warehouse…

(Or possibly a panther.)

Lion Tamer, 1930s archive photograph

The details have become slightly unreliable with age, which is annoying, because I would love to know whether any photographs still exist.

The story, as I remember it, came from a conversation with the head of marketing at what was then Slough Estates. This would have been around the time we first started working with them in 1998, although the campaign itself may well have been earlier.

At some point in the 1990s, they had apparently commissioned a photo shoot inside one of their warehouses involving a large live animal - complete, one assumes, with a trainer and a fairly nervous risk assessment.

The idea was simple enough.

A standard photograph of a large empty warehouse can only say so much. A big cat standing inside one says something else entirely.

• Scale.

• Power.

• Presence.

You may think that sounds unusual.

It was.

But it was also rather effective thinking for its time.

And clearly memorable, because nearly thirty years later I am still talking about it.

“It’s Just a Shed”

A VERY old image from the Blink archives. Limited tools back then, but interesting to see how even then we tried to add dynamism and drama to a logistics concept design. Michael Sparks Associates. ©Blink Image Limited

I mention this because we have been talking to industrial property clients for nearly thirty years about the possibilities of their marketing, and in that time one phrase has come up more than almost any other.

You have probably heard it yourself.

You may even have said it.

“It’s just a shed.”

That phrase does a lot of damage.

Not because it is entirely wrong. A warehouse is, among other things, a large building designed to store and move things.

But because of what it implies about ambition.

If the product is just a shed, then the marketing can be just a brochure.

1. A floor plan.

2. A render.

3. An aerial photograph.

4. A brief description of available square footage.

Job done.

The trouble is that nobody really buys that way.

Or, at least, nobody pays attention that way.

Dull Products Still Need Attention

GoCompare’s Gio Compario character, played by Wynne Evans, who features in its ad campaigns.

Think about the industries that have had to work hardest to make people care about something spectacularly dull.

• Car insurance.

• Utility bill comparison.

• Broadband switching.

These are not naturally thrilling categories. They have no obvious visual appeal, no emotional pull, and no particular reason for anyone to stop what they are doing and pay attention.

And yet Compare the Market built a cultural phenomenon around animated meerkats.

Go Compare found an opera singer with an enormous moustache and turned him into someone so irritating that he became impossible to ignore. (No offence Wynne.)

Neither campaign was really about the product.

They were about attention.

Industrial property is not consumer advertising, of course.

Nobody is suggesting that a logistics park needs a mascot, a jingle or a small fictional animal with a questionable accent.

But the basic challenge is not completely different.

• People are busy.

• Attention is limited.

• Facts alone rarely make something memorable.

And that matters, because industrial property has much more of a story than it often allows itself to tell.

More Than a Metal Box

Haynes-styled cutaway CGI of a modern industrial ‘shed’ - © Blink Image Limited

If you believe you are marketing a shed, you will market it like a shed: factually, efficiently and without much imagination.

But a logistics facility serving a major e-commerce network is not a shed in any meaningful sense.

It is infrastructure.

It is a node in a supply chain that gets products to people’s doors before they have quite finished expecting them.

It supports retailers, manufacturers, supermarkets, construction projects and entire distribution networks.

• It has operational intensity.

• It has scale.

• It has movement.

• It has purpose.

The question is whether the marketing reflects any of that.

Too often, it does not.

The building may be commercially important, technically sophisticated and strategically located, but the marketing still says:

Here is the square footage.

Here is the clear height.

Here is the yard depth.

Here is a CGI of the front.

All useful.

But not always enough.

A Few Signs of Change

There are signs that the sector is beginning to stretch a little.

We have seen a whale placed inside a warehouse on LinkedIn to demonstrate scale.

A lovely piece of work posted on LinkedIn recently.

We have seen dinosaurs used to make clear height instantly understandable.

We have seen more developers using short films, teaser campaigns, motion graphics, drone footage and operational storytelling to explain not just what a building looks like, but why it matters.

These ideas still stand out because they are not yet the norm.

That is the opportunity.

The point is not that every warehouse needs a gimmick.

It is that every scheme needs an idea.

A live lion in a warehouse was one way of making people look twice.

In 2026, we have far more tools available.

• Cinematic films.

• Coming-soon teasers.

• Operational cutaways.

• Animated scale comparisons.

• Social-first edits.

• Interactive presentations.

• CGIs that show not just the box, but the life, movement and value around it.

Used well, these tools can do more than fill a space on a website.

1. They can stop the scroll.

2. Make scale instantly understandable.

3. Create anticipation.

4. Add a sense of drama.

5. Make the functional feel commercially significant.

They do not require a live animal and a trainer.

Which is probably for the best.

The Ambition Gap

The lion - or panther, or whatever it actually was - has stuck with me because it represented something important.

Someone looked at a large empty warehouse and decided it needed more than a straightforward photograph.

They decided the marketing needed an idea.

That instinct still feels relevant.

Perhaps more relevant than ever.

Industrial buildings are larger, smarter, more automated and more important to everyday life than they were in the 1990s.

The tools available to communicate them are also considerably more powerful.

The question is whether the ambition has kept up.

Because if we keep describing these buildings as “just sheds”, we should not be surprised when the marketing behaves accordingly.

richard.birket@blinkimage.com

One of the things I enjoy about working in the logistics sector is that it never stands still.

We have spent nearly thirty years producing imagery for industrial and logistics developments. During that time, the buildings have become larger, the sites more complex, and the expectations placed upon them considerably higher.

But recently I have found myself wondering whether it is not just the buildings that are changing.

Perhaps it is the story around them too.

For decades, one of the strongest arguments in favour of a new logistics development has been employment.

Hundreds of jobs.
Sometimes thousands.

It was a compelling case because it was easy to understand and easy to communicate.

I am not convinced that will be enough on its own for much longer.

Not because warehouses are becoming less important.

Quite the opposite.

Look at Amazon’s increasingly automated fulfilment centres. Look at Ocado’s robotics. Look at the rapid growth of data centres.

(And who can ignore Optimus from Tesla?)

Nothing to worry about…

These are hugely valuable pieces of infrastructure, yet many employ far fewer people than equivalent developments might have done twenty years ago.

The developers reading this will know all of that already.

The more interesting question is what comes next.

What Does the Building Enable?

Nova Oxford, IM Properties - just one potential use case ©Blink Image Limited

If automation continues to reduce employment density, what becomes the dominant planning argument?

Perhaps the answer is that we are asking the wrong question.

Instead of asking only how many people work inside a building, perhaps we should also be asking what that building enables.

A modern distribution centre does not just store products.

• It helps keep supermarket shelves stocked.

• It supports manufacturing.

• It enables next-day delivery.

• It keeps construction projects supplied with materials.

• It connects producers, suppliers, retailers and customers.

Its value is not measured solely by the number of people working inside it.

It is measured by everything it allows everyone else to do.

That feels like a very different conversation.

Instead of asking:

“How many jobs will this create?”

Perhaps we are increasingly asking:

“What happens if we do not build it?”

That shifts the discussion away from employment alone and towards productivity, resilience and national capability.

These are infrastructure arguments rather than purely employment arguments.

And infrastructure arguments require a different kind of storytelling.

The Next Layer of Complexity

Pulse: what lies beneath? ©Blink Image Limited

One area where I have often wondered whether we will eventually see more change is multi-storey logistics.

Compared with countries such as Japan, Singapore and parts of mainland Europe, the UK has built remarkably few examples.

We have had the opportunity to visualise a couple of conceptual schemes over the years. They were fascinating projects and felt like a glimpse into a possible future, yet the market never quite seemed ready.

There are understandable reasons for that.

Historically, land has often been cheaper than building upwards. Multi-storey logistics introduces structural complexity, vehicle ramps, more demanding fire engineering and significantly higher construction costs.

Then there is power.

As HGV fleets gradually transition towards electrification, supplying enough electricity to charge large numbers of heavy vehicles becomes a significant challenge. Grid capacity is already delaying developments across the country, while charging infrastructure, battery storage and the associated safety considerations add another layer of complexity.

None of those feel like permanent barriers.

They feel like engineering challenges.

Whether the solutions arrive in five years or fifteen, they feel more like questions of timing than questions of possibility.

The Communication Problem

A sample of films communicating a bigger picture ©Blink Image Limited

All of this has made me think differently about visual communication.

Traditionally, architectural visualisation has focused on showing people what a development will look like.

That is still important.

But perhaps it is becoming only part of the job.

If the planning conversation is moving beyond headcount, then the way we communicate logistics schemes may need to evolve as well.

The appearance of a building is only one part of its story.

• There is also the operational efficiency.

• The automation.

• The energy infrastructure.

• The transport strategy.

• The biodiversity improvements.

• The role the building plays within a much larger supply chain.

Those things are difficult to appreciate from a single external CGI.

That does not mean the external CGI is no longer useful. It still has an important role, particularly in planning, context, scale and appearance.

But it may not be enough on its own.

Increasingly, films, operational visualisations, cutaways, interactive tools and other forms of visual storytelling can help explain not just what a development looks like, but what it does.

Perhaps making the invisible visible is becoming just as important as showing the building itself.

A Different Kind of Planning Story

None of this suggests that every logistics development should be approved.

Large buildings have genuine impacts on landscapes, transport networks and local communities. Those impacts deserve careful scrutiny.

Nor do I pretend to know exactly how planning policy will evolve over the next decade.

But it feels increasingly unlikely that the conversations we will be having in 2035 will be the same ones we are having today.

As automation continues to reshape the sector, perhaps the industry needs to become better at explaining the wider value these developments create.

Human workforce: thousands > hundreds >a handful

The narrative is changing.

The warehouse is no longer simply a place where lots of people go to work.

It is becoming part of the infrastructure that enables almost every other part of the economy to function.

If that is where the conversation is heading, perhaps the way we communicate these developments needs to evolve alongside it.

I would be interested to hear whether others in the industry are seeing the same shift.

richard.birket@blinkimage.com

“How much does it cost?” is the question we hear more than any other.

It is also a completely reasonable thing to ask.

For much of the visualisation industry, the answer has tended to be some version of: it depends, let us come back to you with a proposal.

That is understandable. Pricing genuinely does depend on the project. But it is not always very useful to the person asking.

So this is an attempt at a more honest answer.

Not just what CGI costs, but why quotes for apparently similar scopes can vary enormously - and what that gap usually represents.

The Market Is Wider Than Most People Realise

The visualisation industry covers a very broad range of operators.

At one end, you might have a skilled individual working alone, with a computer, some software and minimal overhead.

At the other, you have larger production studios with teams of specialists, project management, quality review processes, infrastructure, and the costs that come with all of that.

In between, there is everything from offshore studios benefiting from lower labour costs to small UK teams at various stages of development.

A couple of options for getting from point A to point B. Neither is wrong.

The car market is a useful analogy.

A fourth-hand hatchback and a well-specified executive saloon will both get you from A to B.

They are not the same thing.

And the difference in price reflects a difference in what you are getting - not simply a difference in whether you arrive.

The question is not which one is better in the abstract.

It is which one fits the journey.

Lower Cost Does Not Mean No Skill

It is worth saying this clearly.

There is genuine skill at the lower end of the market.

A solo operator with low overheads may be able to produce good work at a price a larger studio simply cannot match. For a simple project with a fixed design, a clear brief, a forgiving timeline and a limited number of outputs, that may be exactly the right fit.

Offshore studios can offer a similar efficiency from a different angle: lower labour costs in markets where skilled 3D artists are available at rates that are not viable in the UK.

Again, for the right scope, the output can be entirely adequate.

What can be harder to replicate is local context.

Clarendon Centre, Oxford: a more traditional impression of a very traditional city © Blink Image Limited

The visual cues, signage, road markings, vehicles, street furniture and general texture that make a UK street, business park or development site feel genuinely familiar.

That is not a criticism of offshore work in general.

It is a practical observation about where that model works well, and where some projects need more local knowledge.

What You Are Paying For

When faced with current reality, a CGI does far more than simply show what something will look like. It helps potential tenants see into the future and make decisions based on that. © Blink Image Limited

There is a useful distinction between paying for an output and paying for an outcome.

A CGI is an output.

• Helping a planning officer understand the impact of a proposal is an outcome.

• Helping an occupier believe in a space is an outcome.

• Helping an investor feel confident in a scheme is an outcome.

• Helping a development team communicate something complex clearly is an outcome.

The image serves the outcome, but they are not the same thing.

That gap between output and outcome is where process lives.

Studios at the higher end of the market tend to charge more not because the render button is intrinsically more expensive to press, but because the process around the image is more developed.

• Structured draft stages.

• Clear feedback management.

• Project coordination.

• Specialist knowledge.

• Quality control.

• Photography.

• Verified views.

• Film.

• Interactive work.

• The ability to handle evolving design information without the whole process falling apart.

These things cost money to maintain.

They also exist because real projects are usually messier than the original brief suggests.

When Projects Start to Move

This project wasn't so much about changing designs but about an incredibly rapid turnaround. Three images created in seven days. © Blink Image Limited

Design change is one of the most common reasons visualisation budgets shift.

A scheme may start with a clear brief and a fixed design, but that does not always last.

• Planning consultants comment.

• Architects refine the façade.

• Landscape information changes.

• The client wants to test a different material.

• An occupier requirement appears.

• The programme moves.

• Someone new joins the review process.

None of this is unusual.

It is just how development projects work.

Lower-cost models often work best when the brief is stable. When the project starts changing, the commercial model can come under pressure quickly - either because additional work has to be charged separately, or because there simply is not much slack in the system.

The third image in the set from above © Blink Image Limited

A more developed studio process is partly about absorbing that reality.

Not making every change free.

Not pretending endless revision has no cost.

But understanding that development projects evolve, and building a process that can manage that without the quality, communication or programme suffering unnecessarily.

The Right Fit Depends on the Job

The cheapest quote may be the right quote for some projects.

A simple scheme with a fixed design, a single image requirement, a clear deadline and a limited number of stakeholders may be perfectly well served by a freelancer or an offshore studio.

There is nothing wrong with that.

But a complex multi-stakeholder development with evolving design information, a planning submission deadline, verified view requirements, marketing outputs, film, animation or interactive work probably needs something more robust.

In that situation, the cost is not just for the image.

It is for the people and process around it.

The questions worth asking are:

• What does this image need to do?

• Who is relying on it?

• How likely is the design to change?

• How many people need to review it?

• What happens if something goes wrong?

• Is this simply a picture, or is it supporting a serious decision?

The answers to those questions should shape the kind of supplier you choose.

Where Blink Image Fits

Blink is not set up to be the cheapest option in the market, and we do not try to be.

That also means we are not the right fit for every project.

If a scheme is simple, the design is fixed, the brief is clear, the deadline is forgiving and the output is a single straightforward image, there may be leaner options that make more sense commercially.

A freelancer or offshore studio may be exactly the right answer.

Where we tend to be more useful is on projects where the process matters as much as the output.

• Verified views.

• CGI.

• Film.

• Animation.

• Interactive work.

• Complex schemes with multiple stakeholders, evolving information, planning pressure, photography, local context, or outputs that need to support a serious decision.

That is what our pricing reflects.

Not just production time.

Not just the finished image.

But the people, process and judgement required to get to something that can be relied upon.

Lots more info at www.blinkimage.com

We have published our pricing guide at blinkimage.com, not because it is the lowest, but because the question deserves a direct answer.

The cheapest supplier is not automatically the wrong choice.

The most expensive supplier is not automatically the best one.

The right supplier is the one best matched to the job.

If you are weighing up options for a project, we are always happy to talk through what you actually need - even if the honest answer is that someone else may be a better fit.

That is a much more useful conversation than pretending every project needs the same level of process, cost or support.

richard.birket@blinkimage.com

Most CGIs are commissioned after the important decisions have already been made.

• The materials are chosen.

• The façade treatment is agreed.

• The colour palette is largely fixed.

The image is then produced to support planning, marketing, investment or sales.

All of that is valid. It is where visualisation is most commonly used, and for good reason. A good image can help a planning officer understand a proposal, help an occupier imagine a space, or help an investor see the direction of a scheme.

But there is another use for CGI that tends to get less attention.

Using it earlier.

Before material and colour decisions become expensive commitments.

A Different Use for Visualisation

A/B Design testing © Blink Image Limited

CGI is often treated as a way of communicating design decisions once they have been made.

But it can also help test those decisions while they are still open.

If I am honest, we have probably underplayed this ourselves. For nearly thirty years, most conversations have started with planning or marketing. That is where the immediate need usually sits.

But clients have been using CGIs to inform design decisions for years - testing materials, colours, façade treatments and lighting conditions before anything is ordered or installed.

We have been helping them do it without always framing it that way.

This is not a different kind of visualisation.

It is a different moment to use it.

The Problem With the Traditional Tools

A really useful diagram. But far from reality.

A manufacturer’s brochure is designed to sell a product.

The photography is carefully lit, the material is shown at its most flattering, and the scale is usually intimate.

A physical sample is useful, but it shows the material in isolation. It does not show what happens when that material covers several thousand square metres of façade, sits next to another finish, or faces a particular direction in a particular light.

A flat elevation in an architect’s design package is valuable too, but it is still a diagram. It can indicate colour, texture and setting out, but it is not trying to simulate reality.

None of these tools answer the question that often matters most at design stage:

What will this actually look like when it is built?

What Changes at Building Scale

Scale changes almost everything.

A RAL colour that looks clear and confident on a swatch can behave very differently across a large elevation.

A dark finish that feels rich and considered at sample size can feel heavy and imposing once applied to the whole building. It may absorb light rather than work with it.

Reflective and semi-reflective materials are even harder to judge in isolation.

Metal cladding can change dramatically between an overcast morning and a clear afternoon, between a north-facing elevation in winter and a south-facing elevation in summer.

What a manufacturer photographs under controlled conditions does not always tell you how the same panel will perform on a real scheme, in real context, under real light.

Adjacent materials complicate this further.

A brick that reads as warm next to timber can look entirely different next to white render, dark glazing or metal cladding.

The relationships between materials matter as much as the materials themselves.

And those relationships are difficult to judge properly until everything is seen together, at scale, in context.

What CGI Offers Instead

Snapshot of CGI showing fine detail of brick types and coursing © Blink Image Limited

CGI cannot promise absolute accuracy.

It is a simulation, and real-world conditions will always introduce variables: weather, workmanship, material batches, maintenance, ageing, reflections, and all the other things that make built reality slightly unpredictable.

But it can get considerably closer to reality than a brochure image, a sample board or a flat elevation can usually provide at design stage.

More importantly, it gives the design team something to test.

• Different cladding colours on the same building.

• The same brick under different lighting conditions.

• Two façade treatments compared side by side.

• A darker option against a lighter one.

• A warmer material palette against a cooler one.

• A scheme viewed in bright sun, soft overcast light, or at dusk.

These are decisions that often happen late in the process, sometimes under pressure, when procurement is approaching and the direction of travel already feels fixed.

Visualisation gives those conversations a better place to happen.

Earlier.

Before the cost of changing direction becomes much higher.

Changing a CGI Is Not Free

Testing different scenarios and fitouts © Blink Image Limited

It is worth being clear about this.

Changing materials, colours or façade treatments during image production still takes time.

• Models may need to be updated.

• Materials need to be rebuilt.

• Lighting may need to be adjusted.

• Images may need to be re-rendered and reviewed again.

So no, CGI does not make change free.

But it does move the decision to a point where change is usually far less expensive than it would be later.

Changing a cladding colour in a render has a cost.

Changing it after panels have been ordered, fabricated, delivered or installed is a very different conversation.

That is the value.

Not avoiding cost altogether, but reducing the risk of discovering a problem when the options have narrowed and the consequences are much larger.

Beyond the Static Image

The same thinking can extend beyond still imagery.

VR environments and interactive 3D models allow clients and design teams to experience a scheme before a single panel is ordered.

You can move around a building, understand how materials read from different angles, test the relationship between inside and outside, and see how decisions behave at the scale they will actually be encountered.

A static image can answer a lot.

An interactive model can answer different questions.

But the principle is the same:

See it before committing to it.

A Different Way of Thinking About the Brief

Hawley Lane Farnborough. 1 building or 3? © Blink Image Limited

The conventional brief for a CGI often starts when design decisions have largely been made.

• Materials are specified.

• Facades are fixed.

• The visualisation is there to communicate the outcome.

• To a planning committee.

• To the market.

• To investors.

• To occupiers.

But there is no reason the conversation has to start that late.

Bringing visualisation into the design process earlier can turn it from a communication tool into something more useful: a way of making better decisions with greater confidence before the real world makes those decisions final.

That does not mean every project needs endless options, or that every design decision should be tested visually.

It does mean that when a material, colour or façade treatment matters, CGI can give the team a clearer basis for discussion than samples, brochures and drawings alone.

The best time to discover that something does not quite work is not after procurement.

And definitely not after installation.

It is while changing direction is still relatively painless.

CGI is not just for showing a finished decision.

Used at the right moment, it can help make the decision in the first place.

If you are approaching material choices, colour decisions or façade options on a scheme - and want to understand how they might read before anything is specified or ordered - it is worth a conversation.

Until next week…
Rich

richard.birket@blinkimage.com

Bad lighting can make a good building look flat.

Good lighting can make a simple scene suddenly make sense.

That is true whether the image is a photograph taken on site at six in the morning, a CGI render produced in software, or a frame from a drone film shot on a clear October afternoon.

The medium changes. The principle does not.

The difference is control.

A photographer plans carefully and waits for the right conditions. A CGI artist creates those conditions from scratch. A filmmaker sequences them over time.

But in each case, the best images are the result of deliberate lighting decisions.

Nothing should be there by accident.

Here are a few of the things to look for when reviewing any piece of architectural imagery.

Direction, Angle and Time of Day

Whether you are briefing a photographer or a CGI team, the first lighting question is usually the same:

Where is the sun, and what time of day does this need to be?

A high midday sun can flatten facades and produce short, unhelpful shadows. A lower, more directional sun — earlier or later in the day, or lower in the sky through the winter months — creates longer shadows that help a building read as a solid, three-dimensional object rather than a flat graphic.

Shadows are not a problem to remove.

They are one of the main ways the viewer understands shape.

Used well, they direct the eye towards the right part of the image without anyone consciously noticing it is happening.

In photography, getting this right means planning around the calendar, the weather and the clock.

In CGI, it means choosing the lighting deliberately rather than accepting whatever the software gives you by default.

Either way, the question is not simply: is the building visible?

It is: is the light helping people understand it?

The Glazing Problem

Glazing is one of the most common places where otherwise strong imagery falls slightly flat.

A glazed elevation in direct sunshine often reads as a wall of mirrors. The exterior light is so much brighter than the interior that the glass reflects the outside world instead of revealing anything behind it.

Technically accurate.

Visually uninformative.

That may be fine if the aim is to show the exterior form. But if the inside of the building matters — office space, reception areas, activity, occupation, warmth — the lighting strategy has to make the inside visible.

The solution, in both photography and CGI, is to think about the relationship between interior and exterior light.

Placing that elevation in shade immediately makes the interior lighting relatively brighter, adding depth and the suggestion of activity behind the glass.

Dusk or dawn can go further still. At that point in the day, the balance between inside and outside light is much closer, so both can read at the same time.

The building has an exterior.

But it also starts to feel occupied.

That decision has to be made before the shutter opens or the render starts.

Soft Light and When It Wins

Not every image benefits from sunshine.

In fact, some of the best architectural images are made under soft, diffuse light.

Interior views often work better this way. The even quality of an overcast sky keeps harsh shadows from window frames, mullions and furniture out of the shot, allowing the viewer to focus on the space itself: its proportions, quality and character.

The same can be true of certain exterior views, particularly where the shape of the building is simple and the material quality, landscape or public realm is doing more of the work.

Bright sun can add drama.

Soft light can add clarity.

The question is which one the image actually needs.

In photography, that may mean waiting for the right sky, or scheduling a shoot when the facade will be evenly lit. In CGI, it means choosing a lighting model that serves the purpose of the image, rather than defaulting to blue sky and sunshine because it feels like the obvious option.

Sometimes sunshine helps.

Sometimes it gets in the way.

The Small Tools That Guide the Eye

Some lighting decisions are big and obvious.

Others are almost invisible.

A vignette, for example, sounds like something you add in Lightroom when you have run out of ideas. Used badly, it is. Used well, it quietly darkens the edges of the frame and helps guide the eye towards the centre of the image.

The viewer does not notice the vignette.

They just look where you want them to look.

The same effect can be created more naturally with shadow from off-camera trees, neighbouring buildings or architectural elements. It frames the scene without announcing itself as a compositional trick.

Reflections do something similar.

Flat metal cladding can be difficult to make visually interesting in any medium. Reflecting a bright surface — sky, ground, glass, or an adjacent building — into it adds variation and life that direct light alone often cannot provide.

People and vehicles need the same care.

Place them entirely in flat light and they can feel pasted on. Put them partly in light and partly in shadow, and they start to belong in the scene.

These are small decisions.

But they are often the difference between an image that feels arranged and one that feels observed.

Cloud Shadows in Aerial Work

Aerial images have their own version of this problem.

Whether the image is a CGI, a drone photograph, or a composite of both, there is a risk that a highly realistic aerial view blends a new development so seamlessly into its surroundings that nobody knows where to look.

That can be a strange kind of failure.

The image is realistic.

The context is accurate.

The development is there.

But the viewer has to work too hard to find the point.

Cloud shadows are one of the more elegant ways to solve this.

By casting shadow across parts of the surrounding context while leaving the subject development in clear light, the image draws the viewer’s eye to the right place without using an obvious visual device.

It reads as weather.

It is actually a decision.

That is often what good lighting does. It helps the image do its job without the viewer ever feeling pushed.

The medium changes.

The principle does not.

Whether you are reviewing a photographer’s portfolio, a CGI studio’s work, a drone film or an architectural animation, the useful question is the same:

Has the light been chosen, or has it simply happened?

The answer usually tells you how much thought has gone into the image.

Because good architectural imagery is rarely just about showing the building.

It is about helping people see what matters.

ed.griffiths@blinkimage.com
07777 146 495

Extract from the London Tube Map. TFL.

In 1933, Harry Beck produced something that looks, at first glance, like a simple diagram.

The London Underground map replaced a geographically accurate representation of the network with something deliberately abstracted: straightened lines, simplified routes, consistent spacing, and no real attempt to show where stations sat in relation to each other on the ground.

It was immediately understood.

It still is.

Beck’s insight was not simply to make the network simpler. It was to make it useful. He understood what passengers actually needed to know, stripped out what did not serve that need, and designed around the rest.

The map is not incomplete.

It is purposeful.

And that purposefulness is exactly what makes it work.

The same principle sits at the heart of good architectural visualisation.

The Complexity Behind a Scheme

The Complexity Within - not everybody needs to see this, but for some, it is extremely useful ©Blink Image Limited

A proposed development does not arrive as a neat, single idea.

It arrives as layers of information: architectural intent, structural constraints, landscape strategy, planning requirements, access, servicing, interior specification, commercial objectives, sustainability targets, occupier needs and budget realities.

All of it matters.

But very little of it is immediately accessible to the people who need to understand the scheme quickly.

• A planning officer may need to understand scale, context and impact.

• An investor may need to understand value, risk and market positioning.

• A prospective occupier may need to understand operation, access and suitability.

• A member of the public may simply need to understand what is changing and why.

The visualisation does not replace the complexity behind the project.

It translates it.

Into something a specific audience can understand, respond to and use.

Why the Brief Is Where It Goes Right or Wrong

This is where the Tube map analogy becomes useful in a practical sense.

Beck did not set out to draw every piece of information about the Underground network. He set out to help people navigate it.

The destination came first.

Everything else followed from that.

The same logic applies when commissioning visualisation, and it is where briefs most often fall short.

“We need some CGIs of the building” is a starting point, not a brief.

It describes the output, but not the purpose.

A fantastic overview of Symmetry Park, Ardley. But a local dog walker does not need to see this view. ©Blink Image Limited

The important questions come next.

• Who is it for?

• What do they need to understand?

• What decision is this helping them make?

The answer should shape every decision that follows: the viewpoint, lighting, time of day, level of context, people in the scene, what sits in the foreground, and what is left out entirely.

A planning image and a marketing image of the same building can look very different from each other.

Not because the building has changed.

Because the audience has.

And therefore the job the image needs to do has changed too.

Treating them as interchangeable is the visual equivalent of handing a passenger the engineer’s schematic instead of the Tube map.

Technically full of information.

Not especially useful for the person trying to get somewhere.

Snapshot of an AVR document - not much use for marketing a development, but perfect to determine visual impact for a planning meeting ©Blink Image Limited

The Brief Question Worth Asking

Before briefing any piece of visualisation work - an image, animation, film or interactive model - it is worth pausing on one question:

What does this need to make someone understand, feel, or decide?

That answer should shape the rest.

• Which views to commission.

• What format to use.

• How much detail is needed.

• How much detail becomes a distraction.

It is a conversation we have with clients at the start of most projects, and it consistently produces better outcomes than working backwards from a shot list.

Because a shot list tells you what to make.

A good brief tells you why it needs to exist.

Boring empty space vs exciting potential for an interested occupier ©Blink Image Limited

Choosing the Right Tool for the Level of Complexity

The same principle applies at every scale.

A single CGI can translate one carefully chosen idea: how a building sits in context, how a landscape strategy softens an edge, how a space might feel once occupied.

An animation can do more. It can move through a scheme, connect inside and outside, explain arrival sequences, show multiple buildings, and reveal relationships that a single still image cannot.

A film can go further again, shaping a clearer narrative around the project - who it is for, how it works, and why it matters.

And when the complexity increases again - multiple buildings, multiple phases, multiple sites, or a national portfolio - the challenge becomes something else entirely.

It is no longer just about producing an image or a film.

It is about making complex information navigable.

That is where we are currently doing a lot of thinking with Interact3D: a way of bringing project information, locations, visuals and data together in a format people can explore more intuitively.

Tantalising glimpse of Interact3D ©Blink Image Limited

But the underlying question is the same, whatever the tool.

• Who is the audience?

• What are they trying to understand?

• What complexity are we trying to remove?

Whether the answer is a CGI, an animation, a film, or something more interactive, the point is not to show everything.

It is to show the right things clearly.

We are now at the stage where we can demonstrate Interact3D properly, and we are setting up meetings with clients and project teams who may find it useful.

If that sounds relevant - or if you are working on a scheme where the information is becoming difficult to communicate clearly - get in touch and we will show you where we have got to.

That's where my head was at this week!
Rich

ed.griffiths@blinkimage.com
07777 146 495

Most problems in visualisation projects do not come from the rendering.

They come from timing, information, feedback, late changes, unclear decision-making, and assumptions nobody realised needed discussing.

Visualisation is one of those disciplines where the people commissioning the work and the people producing it can end up speaking slightly different languages. Clients often have questions they are not sure how to ask — and assumptions about process, timing and cost that do not always match the reality.

So this week is a practical one: the questions we get asked most often before a CGI, animation or architectural film begins.

What Do You Need From Us to Get Started?

CGI created from a box of highly detailed paper hand drawings - Quinlan Terry Architects ©Blink Image Limited

The short answer is: whatever you have.

We can begin from early-stage 2D drawings — plans, elevations and sections — and we increasingly work from architects’ Revit models, although these usually need optimising before they are ready for visualisation.

Landscape drawings are useful early too, because planting and soft landscaping take time to model properly.

The more complete the information, the more efficient the process. But it does not all need to be perfect before a conversation starts.

IQ Slough, SEGRO plc - images and animation created from a masterplan only ©Blink Image Limited

We have been doing this since 1998, which means we have learned to work from almost anything — occasionally including something not far removed from a sketch on the back of an envelope.

That level of flexibility is not universal, so it is worth asking any visualisation studio what they need before they can begin.

The Design Is Not Finalised Yet. Can You Still Start?

Planning for and shooting the photography early maximises potential for getting the lighting just right for your scheme ©Blink Image Limited

Yes, to a point.

We can begin 3D modelling from early-stage drawings, and photography can often be commissioned independently of where the design has reached.

That can be useful. Starting early buys time — particularly for photography, where the right weather, light and season cannot always be scheduled around a deadline.

The risk is that significant design changes after modelling has started may require parts of the model to be rebuilt.

A good visualisation process is less about avoiding change altogether and more about knowing when change becomes expensive.

It works best when the overall form and massing are reasonably settled, even if the detail is still evolving.

What Is the Process?

The process exists to catch decisions before they become expensive.

For still images, it broadly works like this.

We build a 3D model from the design information. Once it is sufficiently developed, we agree the views: the angles, context, camera height, time of day and what each image needs to communicate.

Photography of the site and surroundings is commissioned in parallel where required.

Once that is done, we camera-match the 3D model to the photographs, produce the first draft, and issue it for comment. A second draft — often the final — follows, and once that is signed off, the finished images are issued.

For animation and film, there is an additional stage before rendering begins.

We produce a storyboard first: a static sequence showing what the film will cover and in what order.

Once that is agreed, we create an animatic — the storyboard set to a rough timeline, with any voiceover or narration, so the pacing and structure can be checked before production starts.

Only once the animatic is approved does full rendering begin.

Those early stages exist for a reason. They are there to avoid expensive rework later.

When Can Changes Be Made?

Spot the difference. Designs evolve. We get that. Tritax Park Newark ©Blink Image Limited

Earlier is always better.

The view agreement stage is one of the most important decisions in the whole process. A lot of the work that goes into finishing an image is specific to that particular viewpoint.

If the camera position changes after that work has been done, some of it has to be redone. It can usually be accommodated, but it has a cost.

The useful analogy is construction.

Once we have design drawings, we build to them. If the design changes, we rebuild the parts that changed — and that has a knock-on effect on everything completed since.

The ideal is to have designs as resolved as possible before production starts, but we are used to working with evolving information.

It just helps to know what might change, and when.

How Long Does It Take?

As a rough guide, still images are often two to six weeks.

Animation and film are more commonly six to twelve weeks.

The range depends on the complexity of the scheme, the number of outputs, whether photography is required, how developed the design information is, and how quickly decisions are made.

Both assume a reasonably smooth approvals process.

Slow feedback rounds extend timelines on any production.

What If We Need It Faster Than That?

Aldgate Central House - one image from a set created in around a week from instruction. Intense and not ideal, but we managed it! ©Blink Image Limited

Compressed timelines are something we can often accommodate, but it is worth being clear about what that means.

Rushing the early stages — particularly view agreement and photography — tends to create problems later that take longer to resolve than the time saved.

A tight programme works best when decisions are made quickly and feedback comes back consolidated and promptly.

The timeline is rarely just about production capacity. It is about how the whole process runs on both sides.

If a hard deadline is fixed, tell us at the start. We can work backwards from it and be honest about what is achievable.

How Much Does It Cost?

This is the question we are most often asked and least able to answer without knowing more.

Cost is driven less by the word “CGI” and more by what the image needs to do.

A single planning image has a different brief from a hero marketing image for a launch campaign. A short teaser film has a different requirement from a full architectural film. A basic massing view is not the same thing as a fully populated internal visual with furniture, people, lighting, landscaping and surrounding context.

The main drivers are usually:

• number of views or length of film

• complexity of the architecture

• amount of modelling required

• photography and site context

• level of finish

• programme

• intended use

What we would always say is that it is worth having the conversation before the brief is written, not after.

How Do We Get the Best Out of Feedback?

I couldn’t resist an Ai generated image here.

The single thing that most affects the quality of a final image — beyond the technical work — is how feedback is structured.

When comments arrive from multiple stakeholders separately and across multiple rounds, the image can end up pulled in competing directions and satisfying nobody.

A single consolidated round of comments from one decision-maker produces a more coherent result and usually a faster one.

That does not mean only one person should review the work. Architects, landscape architects, clients, agents and consultants may all need to comment.

But it helps enormously if someone gathers those comments, resolves any contradictions, and issues one clear set of instructions.

It is worth agreeing internally who has final sign-off before the first draft lands.

If you are not sure how to structure that process, that is exactly the kind of thing we can help with. We would rather have that conversation early than unpick it later.

What About AI?

AI tools have changed parts of the production process, and we use them.

They can help with exploration, enhancement, testing, and certain types of workflow improvement.

But they have not changed the fundamentals: site photography, 3D modelling, design information, judgement, review, and experience.

They have not replaced the need to understand what a planning officer needs to see, what an investor is trying to understand, or what will help an occupier believe in a space.

A prompt is a starting point, not a process.

Most of the questions above are really about one thing: how to get the best possible result with the least avoidable friction.

That happens when a visualisation studio is treated as part of the project team, not simply handed a brief at the end and asked to make it look good.

The best work happens when we are brought in early, kept in the loop as a design evolves, and given the room to advise as well as produce.

If you are about to brief a CGI, animation or film and want to make sure the process starts in the right place, we would be glad to talk.

ed.griffiths@blinkimage.com
07777 146 495

Not every image needs the full kerb-and-mulch treatment.

Sometimes a quick render is exactly the right thing. For a feasibility study, an early-stage design review, or an internal presentation, something fast and functional can do the job perfectly well.

Not everything needs to be polished to within an inch of its life. Spending as though it does is a waste of everyone’s money.

But when the image actually matters — when it needs to persuade a planning committee, launch a pre-let campaign, or represent a scheme to investors — the difference between “fine” and genuinely convincing is often found in details nobody consciously notices.

Until they’re wrong.

Here are a few of the sad little things that tend to separate visualisation that works from visualisation that nearly works.

Kerbs

Crop focussing on kerbs and other finer details. © Blink Image Limited

It sounds absurd, but kerbs are one of those details that quietly reveals whether someone is really looking.

A kerb that is too perfect draws attention to itself. A kerb that is too worn draws attention to itself. The target is somewhere in the middle: a surface that reads as real without becoming the thing the viewer focuses on.

Enough imperfection to suggest it has been there for a while. Not so much that it looks like the council gave up in 1987.

Getting that right across different scales, materials and lighting conditions is more involved than it probably should be.

When it works, nobody notices.

When it doesn’t, something in the image feels slightly off and nobody can quite say why.

Road Markings

Road markings - they ain’t perfect. © Blink Image Limited

Next time you pull into a supermarket car park, actually look at the white lines.

Not a quick glance. Properly look.

They won’t be solid or uniform. There will be worn patches, blurred edges, gaps where the paint skipped, corners that are not quite square.

That is what a real painted line looks like.

Who’d have thought? © Blink Image Limited

Apply a clean, solid white line to a CGI and it can read as false immediately, even to someone who has never consciously thought about road markings in their life.

Which, let’s be honest, is most normal people.

The care applied to something as dull as a white line tells you a lot about the care applied everywhere else.

Mulch (IKR!)

Crop to show edging treatment. © Blink Image Limited

Landscape drawings might show a lawn with a hedge alongside it.

The quick approach is to create a grass surface and a hedge, join them at a neat edge, and move on.

The problem is that hedges do not grow directly from turf. There is usually a zone of soil, mulch or planting bed at the base. A good lawn rarely meets a hedge with a perfect, computer-generated edge.

In high-quality visualisation, this is modelled rather than implied — sometimes with actual 3D geometry for the mulch itself.

This sounds excessive until you see two images side by side.

In one, the landscaping feels real.

In the other, it looks like a material applied to a shape.

The difference is often at the edges.

Cars

Number plates, right hand drive vehicles. Worth getting right. © Blink Image Limited

Cars are useful in CGIs. They give scale, activity and context. They help a service yard feel operational, a car park feel occupied, and a building feel like it belongs to the real world.

They are also very good at ruining the illusion.

A left-hand drive car in a British scheme is a small error with an outsized effect on credibility. Once you have seen it, you can’t unsee it.

The same goes for number plates, vehicle specification, road position and traffic direction.

This is becoming especially relevant with AI-generated or AI-enhanced imagery. A lot of AI tools seem very happy to put vehicles on the right-hand side of the road, even when the image is supposed to be in the UK.

At first glance, everything can look impressive.

Then you notice the traffic.

And suddenly you are no longer looking at the building. You are wondering whether the scheme is in Birmingham or Belgium.

That is the problem with tiny errors. They do not stay tiny once the viewer spots them.

Why Materials Sometimes Look Like Plastic

A crop of a larger image to show materiality and character of surfaces. © Blink Image Limited

The technical term is specular highlights: the way light catches a surface and bounces back.

When this is calibrated properly, materials read as real. Glass looks like glass. Metal looks like metal. Concrete has the right quality of absorption and reflection.

When it is wrong, the same materials can start to look plasticky, flat or oddly synthetic — like something from a video game rather than a proposed development.

This is one of the more technical distinctions between studios, and one of the harder ones to assess from a portfolio.

But if the materials in an image look slightly fake, overly shiny, or strangely lifeless, this is often part of the reason.

Making Your Building the Subject

Tritax Park Newark, using lighting to direct focus © Blink Image Limited

A CGI that blends seamlessly into its surroundings is not always a success.

Sometimes, of course, restraint is exactly the point. A planning view or a landscape-led consultation image may need to show that a building sits quietly in its context.

But for many marketing images, the development needs to be clearly understood as the subject.

Good visualisation uses compositional techniques to guide the eye: differential brightness, selective focus, atmospheric depth, foreground framing, contrast, movement.

None of this needs to shout.

But it does need to direct attention.

If your scheme is hard to distinguish from the context around it, the image may be accurate, but it may not be doing the job you need it to do.

None of these things usually appear on a standard shot list.

And most of them should never announce themselves in the final image.

Nobody should look at a CGI and say, “excellent mulch”.

But these details are often the difference between an image that looks broadly fine and one that quietly convinces.

When a project matters enough for the image to do real work, those details are worth caring about.

That’s my thoughts for this week.
Rich.

If you want to know more, give Ed a call:
ed.griffiths@blinkimage.com
07777 146 495

Have you ever walked through a gallery and found one painting stops you in your tracks? Or scrolled through LinkedIn and paused on one image while everything else blurs past?

That pause is rarely accidental.

Composition, light, viewpoint, scale, human activity - all of it affects whether an image holds attention or quietly fails to convince.

Most developers and architects review CGIs instinctively. Something feels right, or it doesn’t. Something reads as convincing, or it seems slightly flat, slightly wrong, slightly off - but it isn’t always obvious why.

A CGI rarely fails because of one big mistake.

More often, it fails because of several small decisions that were almost right.

Understanding those decisions helps you have a better conversation with your visualisation partner before the work starts, rather than trying to diagnose the problem afterwards.

The Camera

Forced verticals and wide angle leads to looming structures (photo taken intentionally by me to demonstrate the issue)

The camera is doing more work than most people realise.

Keeping vertical lines vertical is a basic principle of architectural photography and CGI. Tilt the camera up and verticals converge - the building appears to lean backwards, the top narrows, and the whole composition can start to feel unstable.

The usual correction is to keep the camera level.

But that can be overcorrected.

A strictly level camera aimed at a tall building does not automatically produce a better image. Sometimes the top of the building looms heavily over the viewer, because changing the camera angle without reconsidering the viewpoint does not really solve the problem. It just changes the character of it.

Often, raising the viewpoint - sometimes to around a third of the building’s height - gives the composition a more natural balance.

Alternatively, the better answer may be not to show the full height at all. A considered view of the lower two-thirds of a building is often more interesting and more readable than a strained attempt to get everything into frame.

Lens choice matters too.

The temptation with a large building is to reach for the widest available lens. But ultra-wide lenses - 14mm, 16mm - can make objects recede too quickly, which often works against conveying scale rather than in its favour.

They also distort anything near the edge of frame. A circular column becomes oval. A parked car becomes something noticeably strange. The viewer might not know what is wrong, but they will feel that something is.

A longer lens, with the camera positioned further back, typically produces a more honest and more convincing result.

Camera height is another small decision with a large effect.

At true eye level, the ground plane often compresses into a narrow strip. Raise the camera a few metres and the space opens up: a courtyard becomes legible, a landscaped approach reads properly, a service yard can be understood as an operational space rather than a thin band of tarmac between viewer and building.

The balance is not to go so high that the image becomes an aerial view by accident. Drone photography is a separate and often very effective tool in its own right.

But a small lift in camera height can completely change what the viewer understands.

The Light

Symmetry Park Ardley for Tritax Big Box - Shot in dramatic evening sun | © Blink Image Limited

Good exterior lighting is how a building reads as three-dimensional rather than flat.

One elevation in shadow, with the adjacent facade in full sunshine, creates the contrast that makes form legible. When every surface is lit evenly, the eye has less to work with and the building can become harder to read.

Sometimes it is even worth putting the main elevation into shadow rather than full sun.

That may sound counterintuitive, but it can produce a more interesting image. Reflections in the glazing pick up bright surfaces beyond. Materials show more depth. The building starts to feel like it exists in a real environment rather than under a generic lighting setup.

For interiors, the instinct is often to show sunlight streaming dramatically through the windows.

Sometimes that works.

But just as often, it creates a busy, competing composition where the light becomes the subject rather than the space. Dramatic diagonal shadows across floors and walls can look impressive, but they can also make the room harder to understand.

Softer, more diffuse light often does the job better. It lets the viewer read the space, the proportions, the materials and the atmosphere without fighting the image.

And where the interior needs to be visible through glazing - a retail frontage, a reception, a fit-out that is part of the proposition - dusk or dawn renders can be worth considering.

At twilight, interior and exterior lighting are closer in balance. In full daylight, the brightness and reflectivity of the glass often win, and whatever is happening inside becomes secondary.

The People

Green Lane for Prologis - selling the lifestyle of the work space | © Blink Image Limited

People are often treated as finishing touches.

They are not.

Scale is the obvious issue. Figures that are slightly too tall can make a building look small and domestic. Figures that are too small can make the whole scene feel oddly disconnected.

But scale is only part of it.

The type of person matters as much as their height.

The figures who belong outside a logistics facility are not the same as those who belong outside a city centre retail unit, a school, an office entrance or a residential development.

Clothes, posture, pace, activity - all of it contributes to whether a scene feels inhabited or staged.

People placed without purpose create a particular kind of unease. Someone drifting across a service yard. A group standing near an entrance for no clear reason. A person looking in the wrong direction, doing nothing, belonging nowhere.

The viewer may not consciously identify the problem.

But they will feel it.

People need to look like they are doing something, and like they belong where they are.

This is not decoration. The human element in a CGI is part of the story the image is telling. Get it right and it confirms everything else. Get it wrong and it quietly undermines the whole thing.

The Purpose

Created for planning - The composition, the scheme interiors and the people used throughout the image are focused on telling the story of a scheme designed to fit perfectly with both low and high rise City neighbours and which will make a lively and positive contribution to the street scene © Blink Image Limited | © Blink Image Limited

None of these decisions exists in isolation.

The right camera, light and people depend on what the image is for.

A planning image, a leasing image and an investor image may all show the same scheme, but they should not necessarily use the same visual language.

A planning view may need restraint and accuracy. A marketing image may need atmosphere and aspiration. A tenant-focused image may need to explain operation, access, layout or fit-out.

That is why the question is not simply: does this image look good?

The better question is: does this image do the job it was commissioned to do?

Most of these decisions are made in the setup and briefing - before a single render is produced.

The point is not that clients need to direct the camera, choose the lens or place every figure.

The point is that these decisions shape whether the image convinces.

And the best time to talk about them is before the render begins.

ed.griffiths@blinkimage.com
07777 146 495

1998: Buying Your First Family Computer | The Learning Zone | BBC Archive

We don’t usually open with a personal anecdote, but bear with us on this one.

Last week I came across a BBC documentary from 1998 on YouTube, following two families buying their first home computer.

The excitement.
The confusion.
The huge monitors.
The complete uncertainty about what these machines were actually for, or what they might one day be able to do.

1998 is also the year Dan and I started Blink Image. (I know, right!?)

Watching it was strange, because it took me straight back to the way we were working at the time.

We took out a significant loan to buy a single workstation. One computer, shared between two people. The software couldn’t render reflections — the processing power simply wasn’t there.

To produce one image, we would set it rendering at the end of the day, go home, and hope it had finished by the following morning. Hope it hadn’t crashed. Hope we hadn’t missed something obvious.

For reference material — people, cars, trees — we were scanning magazine clippings (no Google Image Search), shooting slides on film cameras, and cutting figures out by hand to drop into scenes.

We even ran a small sideline selling CD-ROMs of cut-out people, plants and cars to architects who needed the same thing. (Some readers might still be using them!)

The craft existed.

The tools were brutal.

It is easy to treat that as nostalgia. But the more interesting point is how much the available toolkit has changed while many visual briefs have barely moved.

Our library of people cut-outs. ‘2 blokes gesturing’ was a firm favourite! ©Blink Image Limited

The Leaps That Changed Everything

Not every development over the past 27 years moved the needle equally.

Some things were incremental. A handful were genuinely transformative.

Global illumination was one.

Before it, CGI lighting could look technically competent but somehow flat. You could tell something was off, even if you couldn’t say exactly what.

Global illumination introduced the subtle, bounced quality of real light: the way it softens under overhangs, reflects off floors, quietly fills corners. Suddenly images started to feel less like illustrations and more like places.

Digital photography changed the workflow completely.

Our first professional digital camera cost well over £1,000. The memory card cost another £1,000 (for 1GB!). The resolution was somewhere between 2 and 4 megapixels, which at the time felt extraordinary. The batteries (double-A x 4) seemed to last about 30 minutes I seem to recall!

Our first digital camera - which we still have!

But the important thing was immediacy.

We could shoot reference material and have it straight away. No film. No developing. No scanning.

A 360-degree lens attachment opened up something else entirely: early virtual tours of existing buildings, stitched from photographs. Something previously impossible to offer was suddenly part of the toolkit.

GPU rendering changed the economics of production.

Before it, a single high-quality image could mean a full day’s processing — or networking several machines together to share the load, a setup we ran for a while that felt improbably complicated for the results it produced.

GPU rendering brought that down dramatically. What might once have run overnight could be produced in roughly an hour on a single machine.

That changed iteration. It changed ambition. It changed what could realistically be offered within a commercial project.

Animation, previously almost prohibitively slow at 25 frames per second, began to become achievable.

Not easy.

But achievable.

The internet changed almost everything else.

Dial-up gave way to always-on. Research became instant. Image libraries became accessible. Online forums meant a problem that might once have taken weeks to solve could be answered in hours.

And eventually, it made it possible to run a studio without a fixed base — something we now do every day through Zoom, Google Workspace and a fully remote workflow.

Yes, it’s a whale. A screenshot of an AR project ©Blink Image Limited

Where Things Are Now

The current toolkit would have been genuinely unimaginable in 1998.

Real-time rendering means views, materials and light can be tested far more quickly. AI assists with enhancements, style exploration and quality refinements — powerful, though still dependent on rigorous underlying work.

Gaussian Splatting is opening up new ways to build interactive 3D environments from photographic capture, allowing existing spaces to be explored on a phone or computer without specialist hardware.

Interactive Gaussian Splat model of York Minster ©Blink Image Limited

Animation is more accessible than it has ever been. VR has surfaced and retreated several times over the years, but wireless headsets are now genuinely useful in the right client presentations.

The point is not that every project needs all of this.

Most do not.

The point is that these tools now exist, they work, and they allow a development to be communicated in far more ways than a single static image.

A CGI can still be the right answer.

But it is no longer the only answer.

What This Means for You

This is the part that is actually worth pausing on.

The gap between what is technically possible and what many developers are currently commissioning is wider than it has ever been.

For a long time, the ceiling was low enough that a CGI for a brochure was roughly what the technology could deliver.

That ceiling has gone.

The budget conversation that once ended with a single static render can now open up a much wider range of options: an animated flythrough, an interactive capture of an existing building, a VR-ready presentation model, a teaser film, or a series of stills designed to work across planning, marketing, investment and consultation.

Not all of those are right for every scheme.

But they are available.

And the developers using them are often telling a fundamentally different story about their projects from those who are not.

The brief has not changed in 27 years:

Make someone believe in something that does not yet exist.

The tools available to do that have never been more powerful.

The question is no longer whether the technology can support a richer story.

It can.

The question is whether the brief asks enough of it.

Screenshots from our recent VIA animations - way beyond a CGI! ©Blink Image Limited

If you would like to talk through what is now possible for an upcoming scheme — before the brief is written — we would be glad to have that conversation.

ed.griffiths@blinkimage.com
07777 146 495

How do you get a perfect score?

The instinct when commissioning visuals is to want everything perfect.

The clearest sky.
The cleanest surfaces.
The most immaculate landscaping.

It makes sense. You are investing in the image, so naturally you want the scheme to look its best.

But there is a version of perfect that works against you.

An image that is too clean, too uniformly lit, too flawless in every detail can start to read not as aspirational, but as unconvincing. And an image that does not convince does not do the one job that really matters: helping someone believe in what is being proposed.

The aim is not perfection.

The aim is belief.

Digital music, airbrushed fashion, optically perfect camera lenses

We’ve Been Here Before

This is not unique to architectural visualisation. It is a pattern that appears whenever technology makes perfection too easy.

Glossy magazines spent decades airbrushing people to a standard that bore less and less resemblance to real human beings. Eventually, the backlash was not just ethical. It was aesthetic. The images began to feel alienating rather than aspirational.

Music has had its own version of this. Vinyl records are less convenient, noisier, and more prone to wear than digital formats, yet many people still prefer the warmth, grain and imperfection of the format. Not because it is technically cleaner, but because it feels more connected to something real.

Analogue photography has followed a similar path. Film grain, lens flares, slight colour shifts — details once treated as defects — are now often sought out because they carry the signature of a real moment, in real light.

The appetite for authenticity is not a niche preference. It is a broader cultural shift.

And it applies to development imagery too.

The Sky

Using the sky to help focus the image, the shadows create a natural vignette

A cloudless blue sky feels like the obvious choice for a drone shot or exterior render.

Perfect conditions. Perfect light. Nothing to distract from the building.

The problem is that nothing can start to feel like nothing.

A uniform blue sky gives the eye very little to work with. Everything beneath it is lit with the same flat, even light, and the result can look technically competent but oddly lifeless.

Cloud shadows add drama and focus.

A pool of light on the main elevation.
A slight shadow across the car park.
A sky with movement and character.

These are the details that guide the eye and make a scene feel like it exists at a particular moment, rather than in a sterile, permanent nowhere.

We sometimes introduce cloud shadows deliberately into otherwise clear-sky images, because the compositional work they do is worth it.

The Surfaces

HGVs don’t do donuts, but they do leave a subtle imprint

Apply a clean, uniform material to a car park or service yard and it looks like somewhere that has never been used.

And therefore, somehow, somewhere that does not quite exist.

The fix is not dramatic.

A few subtle tyre marks near a loading bay.
Slight tonal variation across a concrete apron.
Small areas of wear where wear would naturally occur. (That was a mouthful!)

None of this should be conspicuous. The goal is not a post-apocalyptic aesthetic.

It is just the quiet suggestion that this is a place where things actually happen.

That small amount of deliberate imperfection is often what separates a surface that reads as real from one that reads as a simple material sample.

The Landscape

Fake grass vs fake grass, but only one winner

Perfect grass is the AstroTurf problem.

Uniform colour. No variation. No unevenness. Nothing that suggests weather, growth, season or time.

It reads as a simulation of a lawn rather than a lawn.

Real landscapes have variation. Tone shifts. Edges soften. Planting meets hard surface in ways that are rarely perfectly crisp.

A landscape that looks as though it was installed on Tuesday and photographed on Wednesday is not always convincing. One that looks as though it has been there for a season usually is.

That distinction matters.

Especially when landscape is doing more than decoration — softening mass, reassuring a local audience, supporting a planning argument, or helping a scheme sit more comfortably in its context.

Why This Matters for Your Scheme

Planning committees, investors, agents and prospective occupiers have all spent years looking at development imagery.

They may not be able to articulate why a particular image does not quite convince them.

But they feel it.

The same instinct that makes an airbrushed photograph look slightly off, or a pristine digital image feel cold, applies when a CGI has been polished past the point of believability.

The brief for your visualisation partner should not simply be to make everything look as good as possible.

The better brief is to make it feel believable enough to trust.

That does not mean making a scheme look worse. It means understanding the difference between polish and persuasion.

A good image still needs to be attractive. It still needs to show the project at its best.

But it also needs to leave room for real life.

The goal is not a perfect image.

It is a persuasive one.

Understanding the difference is part of what good visualisation work actually involves.

If you are thinking about the visual strategy for an upcoming scheme — before the brief is written — we would be glad to have that conversation.

ed.griffiths@blinkimage.com
07777 146 495

The word flexibility appears in almost every industrial marketing campaign.

It is used to describe buildings that could accommodate logistics, or manufacturing, or mid-tech, or life sciences - often in the same sentence.

What it rarely does is appear in the visuals.

The exterior CGI shows a building. The building is well-presented, professionally lit, architecturally credible. It does not show a logistics operation running through it. It does not show racking, or manufacturing zones, or the kind of front-of-house environment a life sciences occupier needs to see before they can take a scheme seriously.

The specification claims flexibility. The visuals show a shell.

That gap is where occupiers get lost.

Helping manufacturers imagine the possibilities, Film for: Nova Oxford - IM Properties © Blink Image Limited

What Occupiers Are Trying To Understand

When a potential tenant looks at a building they have never visited, they are not just evaluating a specification.

They are trying to picture their operation inside it.

A logistics operator is tracing the vehicle flow in their head - the approach from the road, the yard depth, the dock door positions, the circulation between intake and despatch. They need to believe it works before they will spend time on it.

A manufacturer is thinking about something different: power provision, the flexibility of the floor, how machinery can be positioned, how a workflow moves through the space across a shift.

A life sciences business is assessing something different again - controlled environments, clean zones, the quality of the office provision, whether the building signals the right kind of credibility to the people they are trying to recruit.

All three might be a genuine fit for the same building.

None of them will assume that without help.

Peel Back Animation for Hillwood UK’s Crewe 335 to Reveal Interior Racking and Layout © Blink Image Limited

Showing the Same Building Differently

The response to this is not a more detailed brochure or a longer specification.

It is visuals that do the imaginative work for each occupier type - taking the same shell and showing it in the operational mode that is relevant to them.

A logistics-focused interior that demonstrates racking capacity, yard flow, and clear-height in use. A manufacturing scenario that shows the floor loaded, the servicing accessed, the space working hard. A life sciences interpretation that foregrounds the office provision, the controlled environment potential, the quality of finish that sector expects.

Nothing about the building changes between those images.

What changes is what the occupier can see themselves doing in it.

That shift - from could this work for us to I can see exactly how this would work - is the moment a conversation becomes a viewing, and a viewing becomes a negotiation.

Helping different tenant types picture the possibilities in a sought-after location, Project: Ferrier Street, Wandsworth - Columbia Threadneedle © Blink Image Limited

One Brief, a Wider Market

The practical implication is straightforward.

A scheme designed to appeal to more than one occupier type benefits from a visual strategy that reflects that. Not one generic interior, but a set of targeted ones - each built from the same model, each showing a credible version of how the space performs.

The cost of producing three fit-out scenarios from a single 3D model is considerably lower than most teams assume, particularly when set against the cost of a void that persists because the right occupier couldn't see past an empty shell.

Flexibility is only a marketing asset if it's visible.

A specification that claims it and imagery that doesn’t show it leaves occupiers doing imaginative work your visuals should be doing for them.

If you're marketing a scheme with a broad occupier brief and your current visuals only tell one version of the story, it's worth a conversation.

✉️ ed@blinkimage.com 📞 07777 146 495