Stugoo — Writing

The one I got wrong

Struggling isn't the failure. Staying quiet about it is.

6 min read

Topics

  • Career
  • Planning
  • Shipping
  • Management
  • Failure
TL;DR
  • In 2015 I was the sole developer on a highly creative site for a client, with about six weeks to build it.
  • I underestimated the scope and skipped the planning. I was confident, so I didn't check, and it went live in a state it shouldn't have.
  • What it taught me: understand what you're committing to, plan before you build, and only commit to what you can deliver.
  • As a manager, I'd rather hear "this isn't going to work" in week two than a day before launch.

I've worked on a lot of different projects. Many that I'm incredibly proud of, some that didn't quite go to plan, and a few that probably shouldn't have happened in the first place.

But there's one project I rarely talk about.

Not because it was particularly complicated or technically interesting, but because I failed to deliver it. And for a long time, I was genuinely ashamed and it's taken me over a decade to feel comfortable talking about it.

So here goes.

The brief

In 2015, a friend recommended me for a job at a small creative agency. They needed someone to build a website for a client, and I was brought in as the sole developer to make it happen.

The brief was to build a highly creative site on a blogging platform. Lots of moving shapes, animations, randomised layouts and interactions. The kind of site where the visual experience was the product, and almost everything had some sort of design flourish or stylistic touch.

By that point, I'd done creative work, and had spent the recent years developing my approach to front-end development. I cared about accessibility, clean markup, reusable patterns and making websites that worked properly and would scale.

This project challenged a lot of those principles.

I was given roughly six weeks to deliver. I knew it would be difficult, but I was confident I could do it. So I did what I thought I was good at.

I got stuck in.

Getting stuck

Looking back, that's where it went wrong. I underestimated the scope and skipped the planning, and everything else followed from those two things.

With such a tight deadline, I convinced myself the best approach was to start building immediately. I didn't break down the work, map out the interactions or build a style guide. I just started writing code.

Without a plan, things got messy fast. Fixing one thing would break something else, the platform was being pushed well beyond what it was designed to do, and compromises crept in on the things I cared about most. My answer was to work harder. Longer days, later nights, and instead of stopping to reassess, I kept pushing.

Beer O'Clock

I vividly remember the handover.

It was a Friday afternoon, and I had a train to catch. I'd arranged to go away for a stag weekend. The tickets were booked, and I'd already made it clear that I needed to leave at a particular time.

I wasn't going to cancel my plans.

But there I was, trying to hand over a website that wasn't ready, watching the clock because I needed to get to the station.

I made my last commit and left...

git add -A
git commit -m "beer o'clock"

Worse still, I'd accidentally committed my personal TODO file, containing a rather, eh, "colourful" list of issues.

It was not my finest moment.

The site went live, but it wasn't in the state it should have been, and it took more work to put right. I hadn't delivered it to the standard expected, and I hadn't called out that I was struggling.

A bit of a knock

It knocked my confidence. I'd always taken pride in being somebody who gets it done, and I'd failed at the thing I was supposed to be good at.

For a long time, I couldn't talk about this project at all.

Ten years on

In software, we always talk about failure as a learning experience. Feeling inspired to write about this, I went back through the emails from 2015 with a post-mortem lens. Whilst the spike of anxiety took me straight back, something struck me as I read. I already knew what had gone wrong. I'd underestimated the scope, skipped the planning and taken on too much for one developer in the time available.

Hubris and failure go hand in hand. I was confident, so I didn't check.

Three lessons have stuck with me ever since.

Understand what you're committing to

These days, if it's not written down, it's not happening. This forces you to get on the same page and understand the requirements. Not only does it reduce scope creep, it makes you think about the moving parts of the work. I don't start until I've seen the work. I review the designs, identify the difficult parts, challenge assumptions and break the work into manageable chunks.

A design isn't a delivery plan, and enthusiasm isn't an estimate.

Plan before you build

I skipped the planning because the deadline felt too tight for it. That was backwards. The tighter the deadline, the more the plan matters.

A plan tells you the order of the work and where the risk is. The hardest, least certain parts should come first, so if something isn't going to work, you find out in week one, not week five.

Would a style guide have saved the project? Probably not. But making one would have forced me to understand the work and ask the questions I'd skipped.

Only commit to what you can deliver

I said yes to six weeks before I knew what six weeks would buy. Once you've committed, the deadline becomes the plan, and every problem gets solved with more hours.

A commitment should come out of the planning, not go into it. If the scope doesn't fit the time, something has to give: the scope, the time or the approach. Working longer isn't one of the options.

Owning it

It's been over a decade since that project, and it's the version of me from 2015 who I worry about. Someone capable, keen and too proud to admit they're out of their depth.

It's why my introduction to every new report on my team asks them to highlight risks and tell me when I'm heading in the wrong direction. I'd much rather hear "this isn't going to work" in week two than a day before launch.

For a long time I was ashamed of this one. These days I chalk it up as the project that taught me the most.

Struggling isn't the failure. Staying quiet about it is.