Better to finish an imperfect project than perfect an unfinished one

finished and imperfect

This is the last of twenty-one lessons for design students, gathered from the combined experience of being a student, and teaching students. I’ve published one lesson each weekday for the last month and now they’re done.

21. Better to finish an imperfect project than perfect an unfinished one

It is 2am in the morning. You are sitting in front of your computer screen, inspecting your SketchUp model with critical intensity. You measure each of the treads on your staircase to ensure they are even. You export your model once again to Artlantis and run a test render. It doesn’t look quite right. You alt-tab across to Photoshop and open your timber stair texture map. There are some inconsistencies in the grain that have been annoying you. You spend an hour zoomed in at 2000% rubber stamping them out of existence. Back to Artlantis again for another render. This time it’s perfect. It’s also 5am, so you fall into bed exhausted.

Does this feel familiar?

Computer programmer Tom Cargill said in the 1980s, “The first 90% of the work accounts for 90% of the time. The remaining 10% accounts for the other 90% of the time.”[1] If you are one of those rare students who arrives at the end of semester with time to spare, then you need not read on. Perhaps for you, a project that is both finished and perfect is within grasp. For the other 99% of the world, read on.

Like coal and uranium, time is not a renewable resource. All the more-so when semester lasts just 12 weeks.

Spending night after night as I’ve described above is a great way to never finish. The reason for this is simple: there is always something more that can be done. More adjustments to your design, more alternatives to explore, more lines to add to your plans, more renders to run, more work on your texture maps, more detail in your model. When I was a student and there was a week left of semester, I’d think I’d do anything for an extra day. Then when there was a day left, an extra hour. An hour left, just one more minute.

If you’re anything like me, you will find yourself drawn to perfection at every step. You’ll think to yourself that if you can just get this bit right, you can move onto the next stage. But what you can’t see, what nobody can see when they’re deep in the zone, is what’s important and what’s not. Instead of the forest, all you see is trees. This is as true in practice as it is in university, but I benefit from something you don’t: a cashflow incentive. I have no choice but to keep a lid on the time I spend on a project because otherwise my expenses outweigh my income at the end of the month and I don’t eat.

So you need to develop a strategy for moving forwards. I have a few ways you can think about this, hopefully you’ll find one that works for you:

  • Your entire manifesto for life need not find its way into this project. Hopefully, this is but one of many: you’ll have plenty of opportunities to test out other ideas in the future.
  • Forget about the million could-be projects and focus on the one will-be. Making the right design decisions can sometimes feel like a mountainous exercise, but this is the very nature of creation. The more you do it, the more confident you’ll be that you’re right.
  • Embrace the Japanese philosophy of wabi-sabi. It’s a worldview that accepts transience and imperfection, pursuing beauty that is “imperfect, impermanent and incomplete.”[1] In other words, imperfection is actually okay, don’t stress.
  • Test your project on your fellow students. Ask them to tell you what they think is important. You might be surprised to discover that your perfect staircase doesn’t even rate a mention.


  1. Tom Cargill quoted in Jon Bentley; Programming PearlsCommunications of the ACM; volume 28, issue 9; 1985. Cargill’s ninety-ninety rule is a riff on the Pareto principle, also known as the eighty-twenty rule i.e. that 80% of the effects come from 20% of the causes.
  2. Leonard Koren; Wabi-Sabi for Artists, Designers, Poets and Philosophers; Stone Bridge Press; Berkeley; 1994

Image source

  1. Better to finish an imperfect project than perfect an unfinished one, author’s own image.

Bring your A-game

bring your A game

This is the 20th of twenty-one lessons for design students, gathered from the combined experience of being a student, and teaching students. I will published one lesson each weekday until they’re done.

20. Bring your A-game

This may come as a surprise, but in design you are distinguished first by your commitment and only second by your capacity.

I’ve chosen my words carefully here. I’m not saying that to produce a great project you need commitment more than you do capacity. Nor that commitment is more important than capacity. Rather, I’m saying that whether you like it or not, your design studio is a competition. You are competing with your fellow students for my time, my attention and my trust.

In an ideal world, or perhaps one run by intelligent robots, I and my fifteen students arrive in class at the allotted time. I divide our three hour session into fifteen equal periods and spend exactly that time with each of you. My advice is dispassionate, objective and constructive. I am perfectly fresh and alert for the entire studio, and you each receive precisely the same input from me.

It goes without saying that this is not an ideal world.

I regularly run over time with some students, usually with the best and the worst of you. I must confess that I sometimes even miss out on getting to all of you, and have to make up for it the following session. I am more constructive with some students (usually the best) and more critical with others (usually the worst). I am without doubt less fresh and alert with my last student at 9pm than I am at 6pm with my first.

I am merely human. Which means our teacher / student relationship is as complex and imperfect as all other relationships.

Fortunately, there are ways you can manipulate me.

If you produce plenty of engaging work, and respond to what I and your peers have to say about it, then you earn my time. If you are committed to the studio culture, help others and are prepared to voice an opinion, then you earn my attention. And if your design work is consistently good, evolving steadily from week to week, then you earn my trust. Basically, if you are committed to the design studio, I will be committed to you.

If you can do this, then by the time you arrive at the end of semester you will have had more minutes from me, more enthusiasm, more constructive feedback and are more likely to gain the benefit of the doubt. An uncommitted student has to prove herself in her final examination, she has to fill in all the blanks to demonstrate that she’s addressed all the assessment criteria. In contrast, I fill in the blanks on the behalf of a committed student, and advocate on her behalf when a guest critic fires up.

So, do good design work and don’t do it half-assed.

Image source

  1. Bring your A-game, author’s own image.

A great project succeeds at every stage

success at every stage

This is the 19th of twenty-one lessons for design students, gathered from the combined experience of being a student, and teaching students. I will published one lesson each weekday until they’re done.

19. A great project succeeds at every stage

In my fourth year of study, I produced a project of which I was immensely proud. Unfortunately, my studio leader disagreed. In a tense conversation about the disappointing grade that followed, he commented that my design hadn’t hit the mark on a number of assessment criteria. I was incredulous: couldn’t he see what I was trying to do with the project!? Shouldn’t I be assessed according to the ambitions I set for myself at the start of semester!? Didn’t he realise that I was a prodigal genius lightyears beyond his limited capacity to understand!?

I never ended up challenging my grade, but I’m glad now I didn’t.

What my younger self failed to grasp was that all great projects succeed against all the assessment criteria. That’s how it is in life, that’s how it is at university.

Part of the reason is practical: my ambitions for the project were important to me, but ultimately it had to be judged against all the other students’ projects. Standardised assessment criteria are really the only way to do this. They are also good at acknowledging the breadth of a semester’s worth of work. Sitting on the other side of the fence these days, this is precisely how I establish my criteria. I look at your research, your siting strategy, your form making, your attention to craft, your communication technique. To achieve a great project, you need to succeed across the board.

The other part of the reason is aspirational: a design is greater than the sum of its parts. As alluring as they can be, a pile of sexy renders are only part of the story. Like the human body, you can’t point to any one element and say that’s its best bit. Both the body and a design project are holistic exercises where all parts balance and compliment each other: legs, building massing, arms, windows, eyes and stairs all do their own job within the greater whole.

Image source

  1. A great project succeeds at every stage, author’s own image.

Show it don’t say it

show it don't say it

This is the 18th of twenty-one lessons for design students, gathered from the combined experience of being a student, and teaching students. I will published one lesson each weekday until they’re done.

18. Show it don’t say it

This is another subject I’ve discussed previously. It first popped up almost three years ago, arriving 21st in a series of lessons to my younger self, a series that now numbers 47.[1] I wrote it after attending an atrocious lecture given by Spanish architect, Francisco Mangado, who disobeyed the only two rules of this lesson:

  1. Use less words
  2. Use more pictures

As I observed at the time, Mangado in fact had the two confused. He spoke for many minutes without advancing slides, and rushed through the few slides he did have as though they were insignificant. It was such a blatantly poor presentation, I wondered if he was doing it on purpose.

So there’s a lesson hidden inside this lesson: if you despair of gathering the courage to speak publicly, if you mangle every opportunity, if you get so nervous you pee a little, know that an architect as widely respected as Mangado still hasn’t worked it out.

In contrast, Bjarke Ingels has this facet of his craft entirely under control. His diagrams are legendary: he is the king of using less words and more pictures. I saw him present a number of years ago at the World Architecture Festival in Barcelona. He was merely famous then, and I tuned in with polite interest only to be entirely engrossed by his proposal for a future of driverless cars. His project was a thinly veiled marketing campaign for his project partner, Audi, but I was hooked.

What Ingels understands is that words are insubstantial puffs of air. They’re pretend, they could have been made up a moment ago. Pictures committed to the page are proof, they’re evidence that you’ve worked all this out before.

Not only are they convincing, pictures force you to achieve resolution. To draw a good diagram, you need to have distilled your ideas to their simplest forms. To assemble a good plan, you need to have considered all expects of your programming and spatial arrangement. To render a good image, you need to have thought through space, volume, light and materials. Words leave too much room for speculation, pictures tell the whole story.

Pictures have one other bonus: for those of you bad with words, they are a great crutch. I do want to hear what you have to say, but I’m content for most of the talking to be done by your slides. An IT friend of mine recently mentioned pacing his presentations at 2 minutes per slide. For an architect, this is glacial. I aim for 20 seconds a slide, and often move through them more quickly if they link together into a sequence.


  1. I was a late bloomer.

Image source

  1. Tell a story, author’s own image.

Tell a story

tell a story

This is the 17th of twenty-one lessons for design students, gathered from the combined experience of being a student, and teaching students. I will published one lesson each weekday until they’re done.

17. Tell a story

I believe that while good communication isn’t as important as good ideas, it will get the most out of them. It is necessary because your ideas can’t speak for themselves. I need you to act as their ambassador, to explain the agenda of your project and demonstrate its effectiveness.

The way this plays out is different if you’re presenting via screen or via panels. The former is a linear sequence that limits your ability to jump around your project, but focusses your narrative. The latter is more free-flowing, but makes it harder to direct your audience. I have a preference for screen-based presentations because I like the structure of narrative. They also do better with larger groups of people, and provide the opportunity for video content or even just a spot of rudimentary stop-motion animation.

Either way, design stories are told with both words and pictures.

Comic book artists are much better at this form of storytelling than we are. They understand that the story is shared by what’s drawn and what’s written, and capitalise on the strengths of each medium. They also know that their art is a form of caricature: to show a hero running for example, the most effective picture is with her legs at full extension, not limping along at half-stride.

So to tell your story with pictures, you need to pick the right ones. Your hero images should show your design in all its glory, money shots that both explain and sell it. Make sure each picture does more than one job: it could be a view of one room from another, or show both the built form and its context. Less but better is the goal here. You should focus your audience on your key ideas, not bamboozle them with endless renders.

If your project is a retrospective architectural biography for Julian Assange, then a good place to start would be an illustration of the phased evolution of Wikileaks. It’s not enough to design the spaces to suit Assange’s early hacker days and later more mainstream organisation, you need to pick the images that convey these relationships.

Image source

  1. Tell a story, author’s own image.

Choose the right scale

the right scale

This is the 16th of twenty-one lessons for design students, gathered from the combined experience of being a student, and teaching students. I will published one lesson each weekday until they’re done.

16. Choose the right scale

After years of being around architectural drawings, I have developed a feel for how big things should be when they’re drawn. I imagine this is the case with all architects. When I look at a drawing, the first thing I do is check its scale. This gives me the key to understanding it. I instantly know how big things should be, and often pick up when things are not the right size: a door that’s too narrow or a bench seat that’s too high.

Unfortunately, I don’t believe that scale is taught well in architecture schools. Perhaps it dropped off the curriculum at the same time as handwriting and life drawing classes. This is a pity, because knowing your way around architectural scales is an important foundation lesson that’s used a lot in practice.

Scale (by which I mean a proportion between two sets of dimensions) is a framework that facilitates visual communication. Just like heavy lineweights mean masonry and dashed lines refer to things that are hidden, scale allows you to show the things you need to show with the right level of detail.

As a general guide:

  • Site plans should be drawn at 1:1000 or 1:500
  • Floor plans should be drawn at 1:100 or 1:50
  • Elevations should be drawn at 1:100 or 1:50
  • Sections should be drawn at 1:50
  • Joinery should be drawn at 1:20
  • Details should be drawn at 1:10 or 1:5
  • Joinery details should be drawn at 1:5 or 1:2

You’ll note that I’ve only used scales at multiples of 1, 2 and 5. This is essential. My scale ruler (which I carry with me at all times) happens to only have these standard scales. If you use one that’s not on there, I can’t measure it. I often have students presenting drawings at 1:250 or at bizarre scales like 1:300. These might conveniently allow you to fit your drawing on an A3 piece of paper, but not only do they render my ruler useless, they dismantle all of my intuition. It’s like showing me a piece of prose without all the spaces between words. The letters are all still there, they just don’t make sense anymore.

Image source

  1. Choose the right scale, author’s own image.

Roll design and presentation into one

design and presentation

This is the 15th of twenty-one lessons for design students, gathered from the combined experience of being a student, and teaching students. I will published one lesson each weekday until they’re done.

15. Roll design and presentation into one

The design of your project and the presentation of your project are not separate exercises. They are fundamentally interconnected, and should be developed in conjunction with one another.

There are three good reasons for this:

First, you need to find a way of distilling an entire semester’s worth of work into a few drawings, a model and a quick monologue. Knowing what to emphasise and what to throw away is a hard task. Don’t be lulled into thinking that the project that’s in your head and the one on paper are identical: almost certainly there are facets to your design that you think are as clear as day but are in fact entirely missing.

Beginning the presentation phase of your project earlier allows you to uncover many of these missing pieces. I can only respond to the material you show me, so lay out what the end of your semester is going to look like and I will give you feedback. Just as importantly, the very act of explaining your evolving presentation material over and over again will do most of the work for you: practice makes perfect.

Second, presentation is the most strategic part of your semester. It’s where you decide what to show and what not to show, or what you need to spend time on and what you don’t. The scales at which you present your drawings will determine how much detail is visible. The angles from which you render your project will determine the parts you don’t need to model. The layout of your panels will determine how many drawings and renders you need in the first place.

The more you can leave out of your presentation, the more time you can spend on the bits you leave in. Reams of paper (or streams of pixels) filled with incomplete plans and low quality renders are almost worse than blank space: they are unfulfilled promises.

And third, do yourself a favour and begin work on your presentation material before you enter the twilight zone that descends at the end of semester. Good decisions are not made at 4am the morning of your due date.

I remember having to draw a section detail through one of my projects in second year. It was the night before the project was due and I was knackered. The next morning I was surprised to discover that I had drawn the top half of the detail at 1:5 and the bottom at 1:10.

More recently, and in contrast to my fatigue-induced mistake, one of my students a few years ago started bringing presentation-ready images to our studio sessions from as early as week 8. The rendering software she was using made this easy, but it was a stroke of genius to commit to the final images so soon. I was able to provide feedback early enough for her to make meaningful changes, she was able to resolve all her questions of light, shade and materiality, and by the time her final presentation rolled around, every single image on the screen had been iterated multiple times.

Image source

  1. Roll design and presentation into one, author’s own image.