The Startup drug

It’s 4.24am on a Tuesday morning. I’m in that half-sleep state between sleeping and waking where I’m starting to realise that my dream about being a doughnut farmer may not represent reality to the fullest extent, but I’m struggling to tell which bits are real and which are imagined. I feel like my heart is beating a little too fast and I’m filled with a strong sense of anxiety that something is wrong. I can’t quite fit words around it though.

What’s happened? Did I break the build?

No, that can’t be it. I mean, I didn’t push any code today so I couldn’t have, right? Or did I? Oh, but there was something that I was going to do that might break it. I might need to rethink that because I’m not dealing with the dependencies properly, really I should be pulling that package from there so it shows up on our doughnut feed, or the social media guys won’t get…

Hang on, doughnuts? No, it can’t be doughnuts. We pushed all the donuts with the last version of the app. The doughnuts for this release haven’t finished hatching.

Do doughnuts hatch? Aren’t they laid? Whatever. Actually, maybe we do have doughnuts in this version. I mean, the build is broken. Shit, I’m confused. Oh, and I need the bathroom…

It’s not until I’ve stumbled my out of the bedroom and stood blinking in the light of the living room for a few seconds that I realise that there are no doughnuts and there is no app and there is no build to break. At least not here, standing forlornly in my own home wearing nothing but a pair of pants and an odd sock.

So I head back to bed and get my head down again, but my mind is very much awake now and fifteen minutes later I’m still tossing and turning and mulling over some slightly more sensible design issues relating to a new chat messaging API that I’m supposed to be starting. I’ve been here before and it’s not long before I realise that sleep is hopeless and I’m now fully awake and that’s not likely to change. So to avoid waking up Clarissa (again) I decide that maybe it’s best if I just get up.

This has become a fairly regular occurrence. It also feels different from other bouts of insomnia I’ve had. Previously when I’ve been through patches of bad sleep the trouble has always been getting to sleep in the first place and the problem has generally been rooted in disillusionment with work or ‘poor lifestyle choices’ (one usually resulting from the other).

This time round though I have no problem drifting off to sleep but I’m becoming fast friends with 4.30am. Stress has certainly been a factor but a better way of describing it I think is ‘overstimulation’. It feels a bit like a drug.

Startup life

This all started about 4 months ago, when I moved from a relatively comfortable agency job to a new position at a startup – Movebubble. As a developer, previous jobs have been largely about producing quality code, but at Movebubble I would say that a bigger part of my job is decision-making.

Adapting to this new reality has been a frenetic but fun experience. For context, we’re at the bit of the company’s story which, if this were a Hollywood film, would probably be made into a montage. The basic foundations have been laid, the early hurdles overcome, and now we see lots of cut away scenes of haggard-looking workers engaged in late-night discussions, occasional shouting arguments and plenty of moody, soul-searching glares over hipster coffee cups, before they suddenly start to look healthy and happy and then Bam! They’re all millionaires.

Now, while that montage might well be the four most forgettable minutes of the film, in reality this journey is 90% of the success of any startup. A good idea and a solid foundation is important, but it still takes a skilled execution and that means engaging in continuous, structured learning exercises and reacting effectively to feedback. There is a roadmap of sorts, but it’s liable to change week on week and it is often hard to predict what we’ll be working on even a fortnight from now.

Making that work is a team effort. It’s not about individual brilliance or long hours – the office is largely empty by 6.30pm. Product meetings are discussions rather than dictations and product design ideas, while led by product designers, are collaboratively sourced from across the company. We’re constantly looking for ways that we can improve, too, with regular meetings where we can get together and discuss ways that we can improve the manner in which we work.

What all this means is that there is a lot for a feeble human mind to process – often too much to process in a single day. There have been afternoons where I felt like I was drowning in the sheer volume of concerns being thrown my way, many of which I have volunteered myself to handle.

In December, I started using a time tracker to try and get on top of this and come to some conclusions about my working behaviour. out what I was spending my time on and by the end of the first week I had 42 different line items across 6 different projects, not including breaks. Around a third of the line items were repetitions – there were two larger tasks which took 2-3 days to which I repeatedly kept coming back to – but many others were just 15 minute ‘can you fix this please’ or ‘do you have 10 minutes to talk about this’. On average, I switched from one task to another just over once an hour.

This may not seem like a lot, but when you’re engaged in project work the mental overhead of switching contexts is pretty massive. It can cause you to miss details or go over the same ground multiple times, or completely miss requirements as your mind skips over items on your mental checklist. It requires a lot of extra effort to keep yourself organised and people often underestimate just how much that effort costs them – and how much it can impact the quality of what they do. There have been days when I have gone home having tried to address 8 or 9 different things and barely resolved more than a couple. This can be, and has been, overwhelming at times.

Yet crucially, I don’t go home looking forward to a break; I go home regretting the fact that I’m too tired to do any more.

Clocking off

Now, your brain is a pretty powerful processor. It can do quite a lot of things all at the same time. And just because you’re physically doing one thing or in one places that doesn’t mean you’re not thinking about another. Equally, just because you’ve left the office for the day doesn’t mean that the problem you were dealing with earlier isn’t still ticking away in the back of your mind.

If you stack too many of these things on your subconscious problem-solving queue then it’s going to start demanding extra resources from you to cope with the load. You’re going to find yourself distracted by random thoughts and impeded from engaging in everyday activities, such as normal conversation with your wife that doesn’t revolve around the shortcomings of RavenDB.

Which is how I find myself heading to bed at a reasonable hour and passing out within about 10 minutes of my head hitting the pillow, only to find myself in a strange dreamland where technical and product questions from my day job get mixed up with surreal subconscious imagery in ways that simply don’t make sense, and before I know it it’s 4am and I’m convinced I just had a conversation with Judy Dench about sexing up our microservices.

Now, when you’re going through one of these periods it can be extremely damaging to your self-confidence; you feel that you’re cracking under the strain of something that you should be able to cope with. But when I started to open up about it I discovered that several of my colleagues were experiencing similar issues.

The trigger for this was when a colleague of mine messaged me to ask how a particular bug had managed to get into production and remain undetected for the best part of a month. It was code that had been checked in during my second month at the company when sleep deprivation had been at its worst.

My initial reaction on receiving the message was one of embarrassment and resentment (I was feeling especially weary that morning and fatigue breeds unkind thoughts) but after typing out a couple of more combative replies that I then deleted I simply went with: “how this happened: fatigue, carelessness and mental overload”.

This ultimately led to a very helpful coffee meeting that helped me to get some things off my chest. When I later talked to the product manager on our team, he described how he’d been going through broadly the same thing.

The even more striking thing was – neither of us had any regrets. We both agreed that the job was proving a lot harder than we had anticipated and we were really enjoying the challenge.

Which ultimately is why I liken the experience to a drug – a great feeling with potentially problematic side effects. It’s surely something that can be managed, but the standard advice – more exercise, less booze, find time to switch off – has all helped me to feel more energetic and less fatigued but hasn’t necessarily improved the quality of my sleep. I’ve tried reading, watching films, playing board games and even just meditating to music before bed in an effort to switch off, but there are nights when I find it impossible to let go of an idea.

At the very least, I’m starting to understand why so many silicon valley magnates become so obsessed with health and fitness, as the limitations of their own body start to affect their capacity to get things done. Perhaps I just haven’t taken the healthy living exercise far enough.

Or perhaps what I need is simply to start managing my concerns better. At the end of the day, it is ideas that are keeping me awake, and ideas are just unrealised potential.



Leave a Reply

Your email address will not be published. Required fields are marked *