Thoughts on a four day work week

Sun, Sep 17, 2023 11-minute read

Summer of 2022 in Guernsey was pretty epic. It was gloriously sunny, hot-but-not-too-hot and just an all-round jolly pleasant place to be. It’s one of the perks of living here but it’s also probably one of the most undervalued perks of living here.

– When a beautiful sunny beach is always on your doorstep, where’s the fun in going to the beach?


OK, maybe we don’t exactly think that precisely, but there is a tendency to overlook just how nice it is and prioritise other things like… ugh, being in the office. Working. Yuck.

Meanwhile, Marc and I had been mulling ways of making life at Cortex as amazing as possible. From day 1, we’ve always supported flexible working, ample holidays, lots of free beer and toys but could there be new ways to reward the Cortexians for all their hard work?

The idea of a 4-day week is hardly new, but we mulled whether this could be one. How would people feel about more time out of work to focus on the important things - partners/family/kids, chores, relaxing, geeking out, socialising etc?

So we decided to just bloody go for it and run an experimental 4-day work week for the 5-week period from the start of August through to the 1st week of September1.

How and why and huh?

So this was somewhat predicated on the selfish (?) desire to have more time to enjoy a Guernsey summer, but it was accompanied by some hypotheses to see what went down:

  • What would happen to productivity? Would it stay the same? Or go up? Or… go down?
  • How would people feel and behave at work? Mad boost to enthusiasm or a slide down to being a slave to the keyboard?
  • How would people feel about and use the extra time?

One of the key decisions we made about implementing this was to not simply make it an extra free (5) day(s) of holiday, as nice as that might have been.

In other words, the working week was the same length (approx 35h) but we still had to squeeze that same amount of working time in. This ostensibly meant the usual 7h day became more of a 8.5h/9h day. It wasn’t measured across a week, though - i.e., we were simply looking that over the duration of the trial that an average of 35h weeks were achieved.

So that meant we could (in theory, at least) pick and choose how and when we got the hours done, and work around the practicalities of normal life.

Of course, this was made slightly more complicated by August already being a popular holiday month and so for those peeps that did have actual going-away leave booked it impacted the trial a bit.

And as nice as it would have been to just stick two fingers up to the outside world, slam shut the office doors on a Thursday evening and say “screw you guys, we’ll see you Monday” that actually isn’t really feasible.

(We do actually occasionally behave like a real, grown-up company, believe it or not.)

So that meant not only did we have to set some expectations with our clients that we wouldn’t have meetings etc. on Fridays and we also tend to aim for Thursdays for releases (to avoid the ‘breaking stuff on the weekend scenario’) so this tended to mean we pushed back release day to Wednesdays, but we also needed an on-call rota to carry out the various checks / stuff on fire triage - so some members of the crew inevitably had to at least be conscious of work on the Friday, monitor support emails and so on.

And hey, we’re all (fairly) grown-up… if something really needed looking at on that Friday (or over the weekend), then obviously it would get looked at.

How’d it go, yo?

We dropped the news to the team with very little warning (about a week, maybe less!), which in itself was probably sub-ideal as it meant a bit of a scramble to then rethink and prioritise some of the stuff that was already coming up. Everyone was well up for it though especially after we emphasised that it was very much a trial / experiment, and we were quite open about the motivation(s) for it.

Having crunched some numbers after the fact, they tend to suggest that productivity… drum roll please…. didn’t really change, or was marginally down.

This is not clear cut though since there are numerous ways to measure productivity. For instance - is it lines of code written? New Bragi features shipped? Releases? Billable hours and client revenue? The team is small and spread in varying ratios across all of the various metrics and so any variance against the norm could just as easily be anomalous as they could be statistically significant.

So it’s more of a gut-feeling - it seemed like everybody got their work done to moreorless their normal sorts of levels… but not more.

If there was a dip in output, this could be attributed to one of the recurrent themes that came out from speaking with the team after the trial: the longer days / compressed week meant people felt more tired. The typical approach was to try and start the day earlier, with some getting in to the office before 7am - but in most cases people were in earlier and working a bit later to cover it.

And the practicalities of this are not immune to normal life - if it’s your kid’s birthday on a Tuesday or your boiler explodes on a Wednesday, then these things need dealing with, and so the elongation of the other days becomes even more dramatic.

In addition to that, there was a noticeable wane in energy and output towards the end of the day. It’s not surprising that at the end of a long day, particularly cumulatively across a week (and weeks) that people’s ability to focus to a consistent level decreases. And if you were really enjoying the 3-day weekend, then you’re unlikely to be coming in on Monday morning completely refreshed and raring to go, so the fatigue would inevitably have built up across the duration of the trial.

And as a kicker, Guernsey Summer 2023 tended to be mostly completely rubbish with only the odd spot of the usual glorious sunshine. There’s nothing worse than having a day off only for it to piss down. Ye olde black fly in your Chardonnay situ2. So that put a bit of a metaphorical/literal dampener on the whole kaboodle.

What do you do with your weekend?

The overwhelming feedback from the team was positive though - everyone appreciated, for a range of different reasons, having the extra time. Whether it’s the placebo effect of knowing you’ve got 3 straight days ahead of you or whether it’s actually using that time to good effect, it was on the whole a good thing.

Interestingly, with the odd exception, most people tended to treat the additional day as a direct extension of the weekend.

Well, duh.

In other words, rather than saying

"ooh yay! I've got a whole new day to do something completely kajunga, the oyster is my world... before the weekend!" 

people tended to think

"Woohoo, these are all the things I normally do over the weekend, I can now spread them out across 3 days!"

So this tended to make weekends a little less of a chaotic rollercoaster, since there was more time to just get stuff done - whether that was the usual house chores, those jobs that always need doing, running the kids around to clubs or even just chilling out and enjoying your time.

But in only the odd case did anyone think

"I'm going to use that extra time to work on this thing that I want to work on / learn this new thing that I want to learn..."

We could deduce a few things from this, but generally it seemed like people were knackered from the compressed week, and so were quite happy to just get on with a regular but more relaxed weekend.

And there is totally something to be said for giving yourself a break from the relentless train of trying to do all the things. A geek’s work is never finished so cut yourself some slack.

It’s not your fault, it’s not your fault 3

We have a room full of hardcore geeks whose work is usually fairly well compartmentalised and usally fairly predictable. There’s plenty of it, but in most cases, we know that x things need to be built in y time with z dependencies, so it can be planned and estimated. And this means control over output is a little easier. If you’re in the middle of a big feature, then nailing an 11 hour day to get more (of your 35 hours) done is doable (assuming it’s actually doable, of course.)

My role is a little different. I still write some code, but I tend to do pretty much (but not always) all the other things that are involved in running a business. Whether it’s sales/leads, accounting, contracts/HR or just making sure we have teabags in the fridge and beer in the cupboard… as well as thinking about the things we’re actually doing. These sorts of things don’t neatly fit in to a 4-day week.

In fact, they don’t really fit neatly in to a 5-day week. Maybe it’s founder mentality, I’m not sure, but it’s pretty much a 24/7 or more gig already, so adjusting ‘actual work hours’ isn’t really a thing. My weeks are never 35 hours, so nominally saying the hours need to occur on a Mon-Thurs didn’t really work out.

(And I’d bet my colleagues would argue that my ability to keep to any sort of an orthodox schedule is already fucking awful not the best.)

A loooot of my to-do list items are short tasks with a high degree of dependency on others (especially those outside the business) and so to maintain the same level of output/efficiency there were plenty of things that simply couldn’t wait until Monday.

(Or at least that’s what I told myself. And one of the Fridays fell on the 1st day of the month, which is billing/invoicing day for us (me), so that wrote off most of that particular Friday morning. And even if someone else was on call, I was still pretty much on call anyway.)

I started the trial with the very best of intentions of trying to put in pseudo-normal hours; and the first week or so went reasonably well. But I soon drifted back to my usual ways. Start later, end later which meant that it seriously ate in to the evening times at home. And the weekends were just a slightly elongated normal weekend, and I’m not certain I checked my phone any less than normal.

This probably says as much as about me as it does about the trial. Being able to more effectively switch off is likely to be dependent to be more effective during the time you are actually working.

So. Takeaways? Would we do it again?

In concept it’s definitely a good thing, and although there was possibly a hit on our productivity, on the whole it was a positive thing for everyone.

(Although I do get the sense that were all reasonably keen to get back ’to normal’ - read in to that what you will.)

I did enjoy the prospect of the longer weekend, but did a pretty crap job of sticking to it - whether that was doing too much work stuff over the weekend as normal and/or as a result of not doing enough during the 4-days. It’s hard to tell, but I don’t think that I will ever escape that commitment to be available.

It is my (our) business, after all.

For the future, there are a few options we are mulling over. We could make it an annual occurrence, in the same guise, although part of me is curious to find a way of doing it so that it doesn’t mean compressing the usual amount of work, since this did seem to have a bit of negative impact. We also didn’t find a way to maximise output whilst decreasing the amount of effort. Is there another way of doing it - e.g., a 9 day fortnight? And if the goal is to encourage the crew to invest in their own learning (hopefully guilt-free and beyond the reaches of their other familial commitments) is there another way to do this?

There is plenty to think about.

(You can also check out what Marc thought and Danny thought.)

  1. When the kids went back to school. ↩︎

  2. This doesn’t need explaining, right? ↩︎

  3. This started as a therapy-session ‘So, Matt… how did that make you feel??’ kind of a gag, and so obviously it just got changed in to a Good Will Hunting reference. Obvs↩︎