10 things I got wrong in 2020

2020 was the year to forget but it was also the year I started Byrd. Starting the new year I wanted to take some time to look back at how 2020 went. Inspired by BrewDog’s James Watt I’m focussing on the top ten mistakes I made (since I made many more than ten!). I’ve tried to give credit to all the people who bailed me out through the year too. 

Talking about mistakes I’ve made at Byrd when it’s so early in the company’s existence makes me feel pretty uncomfortable. One of the reasons though that I started Byrd was because I wanted to go about doing business in a different way. Sharing our collective experiences, be they good or bad, is important. I hope it sparks an idea with someone else or leads to someone pointing out how we could do something better.

1. Not sharing enough of what we’re doing

The last time I posted about Byrd on LinkedIn was in July. It wasn’t even really an update, it was a job offer that gave very little away about what we’re doing. We’ve accidentally been working in stealth mode since we started.

I find it terrifying sharing early-stage design work. As designers we’ve all had the experience of miscommunicating concepts, or companies changing directions, and suddenly that fragile chrysalis of an idea has gotten squashed. Without making public what we’re doing, and explaining the philosophy and emotions that underpin the company it makes it impossible for anyone outside of our immediate network to get involved. 

Thankfully Nina Davies dragged me kicking out of my comfort zone towards the end of the year putting in place a strategy. There’s now a cadence of what we’re talking about, when we’re talking about it and why we’re talking about it. I’ve gone rogue writing this article - it’s nowhere on the plan - but hopefully I’ll be forgiven. If you’re interested in what Byrd is doing it’d be great to connect here, or check out @byrdrunclub.

2. Getting too quickly to the solution

At the start of 2020 Byrd was pretty much perfectly formed in my mind. There were lumps in the information architecture and the exact ways a runner would interact with the product but - broadly - it was done. That was mostly because I’d skipped any deep design thinking and simply created a product for a user-base of one: me. Possibly some close friends, who run in a very similar way to me, but mostly it would work just for me. That isn’t a great foundation on which to build a business. Thankfully some residue of professionalism remained and I started to test against real data and talk beyond my immediate network. 

The first lockdown helped. It forced me, and everyone else involved with Byrd, to slow down. All the running races got cancelled and it started to become clear that running was about more than just bagging personal bests. Even for those of us who ostensibly run for ‘performance’ it became clear that running meant something more than a bib number on our shirts. It’s about the headspace, emotional well-being and joy that running gives us everyday. 

But I was mostly saved by Erik Aremyr, with cameo guest spots for Joe Robinson and Tom Pritchard, who all brought different expertise from different disciplines within design. Stepping backwards, to properly explore the different problems, to talk with close to 100 runners, let us see that it wasn’t about getting a fast time, it was about the sense of satisfaction and self-worth that a fast time represents. It’s the buzz from going out on a daily adventure. They helped me to see how critical it was to weave in that complexity, and plurality, of running into the initial release.

3. Being overly optimistic

Coupled very closely to getting to the solution too quickly I was horrifically optimistic about timelines. It was going to be straightforward with an alpha release in April, a beta in August, and release in November. It was born out of the fear of the company being wound up before we’d actually released anything. Being worried about the company’s runway is a rational fear but isn’t a good reason to avoid being realistic about how long the work would have taken, especially given our cumulative experience of delivering in-house and client products.

The optimism bias led to us launching closed beta - a release to about 50 runners - at the end of August. In hindsight it was inevitable that it would be a mess. One of the key mistakes was that a number of the runners had been involved in earlier usability studies so had pre-existing user accounts that weren’t migratable to the new data model. We released late on a Thursday afternoon and by Friday everything had mostly blown up. We’re lucky we have awesome early beta users who have a lot of patience. It’s also fair to say that we probably learnt more from everything exploding than if it had gone smoothly. As with the first lockdown the friction of being slowed down forced us as a team to come up with better solutions.

We implemented a few changes to make sure it’ll never happen again. Dave Brealey and Toby Hammond were instrumental in making sure that there was a better release process, and that there was a sensible, systemic approach to deploying both server-side and client-side changes.

4. Not being a good enough story teller

The moment I hated most at any client pitch was the initial round-table where everyone introduces themselves. ‘Hi, I’m Edd …’ would come out easily enough but then would normally be followed by something resembling English words but not necessarily in any sensible order. It was fine as soon as I could talk about design, a product, or really anything that didn’t relate directly to me. This led to a na?ve assumption that it would be easy to reel off the story about Byrd. This proved to not be the case. Running, the importance of running, and how Byrd could get more people enjoying running, in my mind, was all too closely bound to me. There are a number of phone calls and pitch opportunities that I won’t get back where I mumbled quickly through what the ‘why’ before rushing quickly into talking about the product.

Nina bailed me out again with her exceptional storytelling skills and recommendations of books to read to get better at it. We also took the decision to start working with ISQ, which has been incredibly helpful at putting lots of disparate pieces together and pulling together a coherent sense of what ‘Byrd’ is.

5. Being too much of a player-manager

For parts of the year I had the feeling we were just a few lines of code away from being ready. In all the fear of starting a new company there was something comforting about having a mobile emulator running and being able to make it do what I wanted to. There’s no clear path of how best to run your own company and the busy work of writing lots of for loops was a place I could convince myself I was adding value. 

Because there was a sense that we were only ever a few lines of code away from being done I took “pragmatic” decisions that led to some fairly unattractive files that were very difficult for others to maintain. One day when we’ve safely released version 1 we’ll all look back and laugh at the 6,000 line monolith of a file that powered several screens in the UI, made requests to and from the database as well as running various algorithms to compute the data being inputted. It “worked” but it shouldn’t have done and my time would have been much better spent writing down the thinking that was underpinning the goals section I was coding.

Dave and Toby again had the patience and confidence to step in and enforce coherence through the code and force me to properly learn how to use Dart rather than hacking my way around it.

6. Not doing enough to reduce bus factor

In the world of product bus factor is the risk of all the knowledge, or most of the knowledge, of the product sitting with one person. If that person gets run over by a bus the product is done. I didn’t do enough in 2020 to reduce that bus factor. This was obviously wrapped up in my other mistakes. If we’re a few lines away from release why bother to share the secret sauce? But we weren’t a few lines away from release.

With Byrd the goal early on from the team was that we’d have a high-autonomy / high-alignment dynamic. That is, that we’d know what we were trying to accomplish and, depending on the scenario, collectively or individually solve it. There’s a classic Henrik Kniberg drawn version here if you’re into that sort of thing.

Because I hadn’t taken the time to move things from my head on to paper though there were instructions flying all over the place. Instructions are a bad way to build resilient teams. Even if used with good faith (which I hope I did) they can get dangerously close to micro-managing and gives no space for the person being instructed to potentially present a better approach. Without the underlying biases, opinions and values of Byrd being properly written down it was often difficult for anyone else to interrogate and challenge the underlying design.

Part of the problem was that I found remote work difficult. I think everyone on the team did to a greater or lesser extent. Even when I’ve been working with clients in other countries there’s always been a physical space where we’re working and it’s easier to share thoughts. Our first day back in 2021 we’ve tried to do better by creating that ‘war room’ space using Miro boards with the aim of having structured workshop sessions alongside deliberately informal coffee calls, even if that makes us feel a bit daft.

7. Not enjoying running enough

This year, for the first time in a long while, I only hit my C goal for the amount I ran. I ran badly too, with lots of repetitive routes and similar distances. It was partly caused by sticking firmly to the initial lockdown rules meaning I had to make the most of my hour outside, which then framed much of my running over the summer and partly because I wasn’t able to run with other people to push me into a new routine. In August I got injured for the first time in years with a problem in my hip. For someone building a business around the importance of runners and running this was not a good place to be.

I was helped by my wife, Elena Massucco, who was getting back into running, and whose enthusiasm was infectious. For our respective birthdays - both of which are in September - we strapped the kids into running buggies and set off. Those micro-adventures rekindled my own love of the sport and reinforced the importance of what we were doing as a team at Byrd.

8. Not stepping away from the computer enough

I took very little time off in 2020. This is normally bragged about by founders but is a stupid thing to do and probably my biggest mistake of the year. Intuitively it seems sensible that if you want to get something done you have to work all the hours of the day to get it done. What tends to happen is that you end up with blinkers on and fixated on a single solution. I was too scared of being accused of slacking to take the time off.

 Worse, everytime I stepped away from the computer I solved a problem of one sort or the other. Even things that could be considered truly frivolous, like watching Netflix shows, unlocked something. For a long time in the autumn we were trying to work out how to avoid swapping things around in a runner’s future because we saw it was confusing for the user and very brittle programmatically. I was completely stuck on it because all I was totally fixated on lists and ever more efficient ways to move elements between them. Watching Queen’s Gambit though I was reminded of game theory, which reminded me of the Nash equilibrium and decision trees. It led to a solution that completely got rid of the lists and is much more robust as a consequence.

 The message finally sunk in just before Christmas, which is why this has arrived on the second working day of 2021, since I took the full Christmas break to watch Trolls eight times with my daughter, start seeing my son crawl and eat too much chocolate with my wife.

9. Not asking for support enough

Looking back on 2020 I’m not sure how I managed to get anybody to join me on Byrd. As an extension of my mistake not sharing enough I wasn’t good enough at asking for support. I was worried people would say ‘no’, or not think it was a ‘serious’ project, or worse they’d echo back all the doubts I had running around my own head. This was an idiot move as the vast majority of people I got in touch with were incredibly helpful. Yes, there were some people who said unkind things or simply didn’t engage but even that’s a positive. Understanding that the product isn’t resonating with some people means we’re doing something right. I’ve no interest in working on something that appeals to all the people all the time.

I was lucky that some people gave me a lot more time and energy than I deserved. Russell Beard was particularly helpful at pushing me out of my comfort zone and getting me to talk to people outside my obvious network. Stephen King was instrumental in helping me recognise that a business and a product aren’t the same thing and that if either were to survive we needed to work out how to put together a financial model and business plan that wasn’t steeped in optimism. And Peter Fullagar was, as always, excellent at pointing out when I was fixating on a micro design problem rather than looking at the wider context.

10. Not dancing enough

It seems daft to talk about dancing on a professional network but 2020 was not a good year for dancing. This is for the entirely obvious reason that there’s been a global pandemic so parties shouldn’t be happening and because our son, and second child, was born in March. Even if there had been parties I probably wouldn’t have been at them. 

Dancing is like running, it’s about rhythm, cadence and being lost in something outside yourself. It’s that idea of meditation in movement. Like running I find I can clarify a lot of disparate thoughts dancing. Having my brain focussed on a different task allows the mind to bring . It’s particularly useful for those ideas that span across different domains. They’d seem daft, and get dismissed immediately, if you thought about them directly. But if they’re in the shadows whilst you’re doing something else they can form safely. 

It doesn’t even have to be proper dancing, just moving to music on my own is good enough. Byrd could have used some of that thinking during the year. I’m fairly confident problems, or solutions, that took weeks to become clear by brute-forcing them head-on would have gotten sorted more easily. But this year has been awful and stressful and the idea of dancing felt frivolous and small. It was my daughter who stepped in to help me with this. She doesn’t see dancing as frivolous or small. She’s hit that age where music’s her whole world and is unwilling to accept ‘no’ as an answer.

----

Thanks for reading. I’ve been enjoying reading other people’s experiences of 2020 and look forward to seeing more of them in the coming days.