The Images We Make for Ourselves

You know the estate agent photograph. Small bedroom, camera jammed into the furthest corner, widest lens in the bag.

The intention is obvious enough: make the room look generous. The effect is the opposite. The bed stretches away into the distance, the walls lean out, and the whole thing looks like a doll's house photographed through a fishbowl. Nobody comes away thinking the room is big. They come away thinking something is being hidden.

It's a convention that defeats itself, and it survives because everyone does it.

Our industry has its own conventions. Better executed, better intentioned, but occasionally doing something similar.

The Beautiful Dark One

There's a register of architectural image that is genuinely gorgeous. Low cloud, half the building lost in weather, one figure walking away, large parts of the frame given over to nothing much at all. Quite often it’s raining.

The obvious example is MIR, in Bergen, who have been doing this since 2000 and work with the likes of SOM, Snøhetta and Zaha Hadid. "MIR style" is a phrase people in our trade actually use, which tells you something. A good part of the industry studies their work, ourselves included, and if you've never looked at it properly, mir.no is worth ten minutes of an evening.

It's wonderful work and I'm not going to pretend otherwise. Stunning.

It also does exactly the job it was commissioned to do. That work is made for competitions, for concept stages, for one architect falling in love with an idea, and in those rooms atmosphere isn't an indulgence. It's the brief.

What interests me is what happens next. A visual language developed for one audience gets admired, then borrowed, and turns up somewhere it was never meant to be. The style is still beautiful. It can also be completely wrong for somebody trying to work out how a building will affect their street, or whether their operation fits inside it. In front of a planning committee, an image where you can't quite see the building starts to look like evasion.

Everything in October

Then there's the colour. At some point the whole industry agreed that schemes look best in late afternoon in early autumn, and we've mostly stayed there.

Warm light raking across the elevation. Trees turning. Long shadows. A sky doing something interesting in the top third.

I like all of that. We produce plenty of it. But it has become a default rather than a decision, and defaults are visible once you notice them.

Ask an AI tool to enhance an architectural image, or simply ask it how you ought to, and it will suggest golden hour almost every time. The convention is now so thoroughly established that the machines have learned it and hand it straight back, which is about as good a definition of a default as I can think of.

If every scheme in a borough arrives looking like the same golden Tuesday in October, the light stops saying anything about the particular building it's falling on.

The Square-On Elevation

This is the one I find most interesting, because it's the most beautiful and the least useful.

The almost-orthographic shot. Dead straight on, no perspective to speak of, the elevation presented like a drawing that happens to have materials. They can be stunning, and they are extremely popular inside design studios.

They should be. For judging proportion, rhythm, the relationship between solid and void, that view is exactly the right tool. It's how the elevation was designed, so it's how the design team wants to see it.

The trouble is that very few people will ever experience the building that way. You arrive at an angle, from below, with half of it behind a tree and a van parked in front of the rest. No resident, officer or occupier has ever stood where that camera stands.

So it's a true image that answers a question the audience wasn't asking.

Who Is It Really For

The thread running through all three is the same, and it's the same as the estate agent with the wide lens.

Each of those images is made with somebody in mind, and quite often that somebody is us. The design team. The portfolio. The awards entry. The people who will notice the restraint, the palette, the perfectly resolved parapet line.

There's nothing wrong with that, as long as it's deliberate. A concept image for an architect genuinely should be atmospheric. An elevational study genuinely should be square on. The problem is when one of those images goes out to do a job it was never built for.

And the audience often has a very different question from the one we're answering. What is it? How big? What's it like to stand next to? Will it ruin my view? Can my business work in it? None of those are answered by a beautiful parapet.

I suspect familiarity is a large part of it. We spend months looking at drawings and models and forget how much we're carrying around in our heads. We look at an atmospheric image and quietly fill in everything it doesn't show. Somebody seeing the scheme for the first time can't do that. They only have what's in front of them.

We're as guilty of this as anyone. There are images on our website chosen because we're proud of them rather than because they did the most work, and I suspect that's true of every studio you could name.

Some images exist to explore a design, set a direction, or simply be good. Others have to help somebody understand something and then decide. Occasionally one image manages both. I wouldn't assume it will.

A Way to Check

There's a simple test, and it costs nothing.

Show the image to somebody who hasn't sat in the design meetings. Don't explain the building first. Ask them what they think they're looking at.

Their answer will often be more useful than another round of comments from people who already know the scheme inside out.

Nobody ever bought a house because the photograph made the bedroom look like a doll's house.

Until next week.

Rich

PS: Apologies in advance. Once you have noticed the wide-angle trick you cannot unsee it, and by Sunday evening every bedroom on Rightmove will look to you like a small corridor.