Like many other product development team around the world, the global pandemic has meant we’ve found ourselves as an all-remote team since March. In the UK, there is much talk around “return to the office”. But we’re not rushing to return to the status quo.

I wrote this note to set out a vision of how we want to return to the office, and what we’ve learnt from the enforced remote period. It’s likely to resonate more widely, and I’m interested to hear from other product development teams about their plans.

Note: The proposals here do not necessarily fit within current (September 2020) rules and regulation. There is no timeline to these, but they set out a future ambition.

Background & what lockdown has taught us

Product development is a combination of highly collaborative work and deep focused solo work. The environment teams work in and the tools they have can make a significant impact on team performance.

We have always, and will always, support remote work. However, prior to lockdown we have struggled to make the remote experience as good as the in office experience.

During lockdown we have successfully operated as fully remote teams. Aspects of this have been a significant upgrade, especially for our primarily-remote team members. For example:

  • online meetings work better when everyone is dialled in individually. It was the people in the physical meeting room that was the problem!
  • we have invested in tooling that makes remote collaboration better (Zoom, Miro - on top of our existing tooling: AzureDevOps, Slack)
  • for some of the team, a private office has been an upgrade

There are times when it’s easiest to have everyone physically colocated. We already knew this, which is why our Belfast based team members have periodically flown over to London.

5+ months is too long without in-person interaction. The main drawback of remote work has been the human need for social interaction. This is what makes work fun.

Not everyone has had a great, or long term sustainable, home working environment. Many people have optimised their lives around working in a central London office.

Trade-offs: all-in vs all-remote vs mixed

Both all-in-person and all-remote regimes provide an optimal maxima, but with different trade-offs:

  • All-in-person teams have high conversation bandwidth, are great for helping each other out but often lead to poor documentation. It is difficult for people not-in-the-room to stay in the loop
  • All-remote teams naturally encourage better documentation and make it easier for everyone to stay in the loop. It is more difficult to help people out and harder to have ad-hoc conversations.

In both cases, we put significant effort into a) optimising for the regime we’re working in; b) overcoming the weaknesses / inherent biases in that regime.

In normal circumstances, all-remote teams periodically meet in person (for example, the almost hidden reference to “a few times a year all our employees fly in for a company-wide gathering” in the book REMOTE: Office Not Required by Basecamp)

Mixed regime teams (and their experiences pre-lockdown) have, to some extent, the worst of both worlds. Which is why they often co-located periodically.

Our experiences in lockdown have shown that we can do better than our prior mixed regime experience. By embracing remote-first, with periodic intensive in-person co-location, we can create a new optimum.

So how could this work?

In terms of working practice, we will be remote-first.

We will co-locate as teams around particular stages in the product delivery lifecycle:

  • sprint start/end (maybe not every sprint? 1 or 2 days)
  • quarterly product roadmap planning (5 days?)

Expect that in-person time means less at-keyboard focussed time.

We do need to provide “desk as a service” for those who want it… but the home of the team is online and we optimise for that. We’ll need some house rules to ensure we stick to remote first principles. For example, individually dialling in to meetings, even if some of you are in the same office. Unsure how to make that work in practice, so we’ll need to try.

We will continue to invest in tooling to make remote great.