Last week was pretty epic, and I remember writing that I considered myself “tired, but not exhausted”. This week there’s lots less to say, but somehow I feel absolutely shattered.

Some of that is because I’ve been in many more “pretending to be extrovert” scenarios, but also I suspect I’m slightly running out of mental swap space – there’s not quite enough autopilot yet to process some of the increasing numbers of things I’m working on.

I know it’ll come though. It always does. I try to remember this phase of any new role is like the arc of when we were new parents. Week four was always the worst, because you’d run out of mental and emotional reserves from before it all started – but weren’t yet seeing many tangible benefits. You just had to endure until week six or seven, when unexpectedly things would start to make a bit more sense.

The people, meanwhile, continue to be generally lovely. It’s just my brain that needs to catch up.

My week at GDS

• I had a delightful chat with another of the Deputy Directors, as I start to do my rounds of the organisation. We rather hit it off, and I hope we’re going to get the chance to work together soon. I’m seeing two more next week – and I really hope they go the same way.
• I had another of my open “meet Tom/Ask Me Anything” sessions for the Product Managers to find out a bit about me and how I work. There were lots of good questions from PMs ranging from senior to associate. It was particularly nice to be able to talk about my plans for how we can make career progression a bit more transparent. In the “vanilla” DDaT framework for product managers, it merely says that staff should be able to use product techniques with greater confidence – but doesn’t actually get into the specifics of what those are. So…one of my plans is that we should actually write down our detailed expectations at the different grades. When should you be focused just on usability? When might we expect you to be confident with A/B or landing page tests? When should you be able to create a prototype to test assumptions? How do you make the judgement on the risks around concierge tests? Etc etc. And that’s just product craft. We can do the same around how people think about value-for-money (and therefore the risks associated with any given bet). This all seemed to go down well, so now I just need to get on with bringing it together. The word “just” is doing a lot of heavy lifting there.
• I was part of a panel sifting a large number of CVs for a new role in our Digital Identity team, and the moderation session was a really good way to beter get to know some team members and see what they valued in Product Management in their particular context.
• The three of us got to collaborate on finessing the interview process a bit, bringing in some of the scenario-based techniques I introduced at DIT. The HoP did a fab job of coming up with something that I think is really going to give the best candidates a chance to shine – so 🎉.
• I also got to have a good in-depth chat with one of the Lead PMs from that programme, and it was super-useful to hear in detail about their background and current work, and start thinking about how I might best help them.
• And finally, I had my first “Heads of Product” get together – bringing together the HoPs from the three big directorates at GDS. We’re all so busy it took ages to find a diary slot, but I’m really glad we met. The four of us got to cover loads of really important stuff, in terms of what we’re doing for staff, how we can build a product leadership community, and how we can support each other. (We talked about early talent, mentoring, skills, L&D offer and loads more. Plus a bit of group therapy, of course.) I’m also picking up a few things they’ve had on their “I should do this for the greater good of the community, but dayjob argh” lists, which is a nice bit of help to be able to offer them already.

I’m trying to be ultra-ultra-Lean in how I approach many of these things. Do as little as possible to give people something to react to – I can fill in the detail later once I know things are ‘more true’. For example with the HoPs gathering I was pretty sure we’d end up with some sort of trello board, how our flow might work, and I had some ideas about what topics we should cover – but I knew that if I engineered flow too heavily around my reckons, I’ve wasted a big opportunity for everyone to tell me what they really need. (But yes, we do now have a trello board, and I’m very pleased that loads of the bigger things on it aren’t the priorities I first thought of.)

Similarly, we had a session about roadmapping with all the heads of product and delivery. I’ve joined a central team thinking about how we might improve things. I kicked off the session with a bit of a roundup of the thinking that had been going on – and on our hypothesis for improvements. But I deliberately didn’t over-engineer the Mural board we might use for testing the next stage work, or force anyone into an activity – because I wanted a chance to test the underlying value proposition of the work (always producting, me). We got a lot of interesting feedback to our approach, and it’s clearly going to need a bit of iteration, but we also got one team saying they’d be interested in doing a light-weight experiment with us to see how it plays out in reality, and another wanting to get involved in April. As ever, I’m always wondering about the least possible work we can do to get the next bit of insight we need.

It was also interesting when telling our story to a new group people that I spotted a slight creeping assumption. It would be incredibly convenient if it was true (which is why I’d not spotted it) but it definitely needs checking. So in some ways I’m actually glad we aren’t “go go go!” and have been gifted a little bit of time to validate. It’s funny how saying things out loud, to a completely new set of people, can reveal things like that.

An unexpected tour into meta-kanban

On Thursday evening I got to drop into a Zoom chat from one of the original masters of Kanban, Dave Burrows. Kanban isn’t a specialist area of mine, but I always like to stay curious, and Dave had come highly recommended by my fellow Head of Profession for software engineering.

(An aside on curiosity: I remember signing up for a quite expensive workshop on the “Accelerate” book about DevOps with Nicole Forsgren and Jez Humble, thinking it would be an interesting perspective on my work with PaaS – and having SOOOOOO many new perspectives and insights and tangents and books from that day, lots of which I still use. Metrics hierarchies? Corporate culture mapping? How to measure anything? Storytelling with data? All from that day. As a result, I’m always prepared to take a chance on something new. Plus, one of my favourite books is Steven Johnson’s “Where Good Ideas Come From”, which highlights the importance of cross-pollination between disciplines to foster sustainable innovation. He calls this exaptation, and I’m a fan. I’ve rarely regretted going to a talk about something a bit leftfield.)

Mike’s talk was actually wonderful. I sadly had to join it late because I was in the first “Heads of Product” meeting mentioned above, but there was a ton of great things to think about – mainly textural rather than anything to do with what’s going on now, but a few highlights:

• “Watermelon projects” – Green on the outside, red on the inside. This term caused quite a virtual cackle from all of us on zoom. We’ve all seen them, some of us have worked on them, and now they have a name.
• A disaffection with the increasing process-centricity around agile. I’ve always worried about the lack of upward flows of insight within frameworks like SAFe, but Mike put it really well: “we’re in danger of reinventing waterfall, and losing all the benefits of this way of working.”
• He also talked about a list of “dysfunctions” – not in the Patrick Lencioni sense – but he would occasionally use throwaway lists like “dysfunctions of power, of levels of detail, of…” that I didn’t have time to write down. And I didn’t want to ask about because it may have been in the bit at the start that I missed. So I’m going to wait for the recording, and if it’s not covered I’m going to ask him on LinkedIn! (It may be I have to sign up for his course to find out, but hey – that’s quite tempting already.) The term “dysfunction” might be a bit ‘strong’, but there’s definitely something in it.
• One example he gave of this was in a context I’d simply never considered before, and was a total eye-opener. It’s what the diagram above represents. He talked about work dependencies or flows between teams – all happily working in a kanban way.
  
  But there’s also a layer of kanban *between* the teams.
  
  Handoffs in multi-team delivery can sometimes end up being “we’ve done our bit, it’s now on them”. (Guilty as charged here!) But he calls this a dysfunction of power: the first team feels they can push work into the next team’s queue before they’re ready for it, breaking the kanban pull flow of the team that inherits the work.
  
  Another potential way to look at it is: why was that team working on something before it was going to be needed by the team inheriting the work? Perhaps that work should remain “not fully delivered” within the first team’s board until it’s ready to be worked on by the new team? Would that create better visibility around bottlenecks and flow to management, making it easier to spot the overall tensions in the delivery/learning system?
  
  I need to do a lot more thinking about this, but that meta-kanban approach is something pretty new to me. (It might be buried in “Lean Enterprise” of course, but I didn’t spot it 5-6 years ago because I didn’t have enough experience/context!)

We were also in the presence of Dave “Cynefin” Snowden – who clearly has a “resting unimpressed face”. Perhaps he was also thinking “5:30 on a Thursday evening, what kind of a time for a talk is this?”

Elsewhere…

• I got to enjoy my first tentative visit to Rough Trade East as “the local record shop”. I’m going in gently – I picked up an online order of the new Orbital album – but I can see this is going to be a very fruitful (if financially ruinous) relationship.
• Managed to get out for a run on Tuesday morning before work, and also did Parkrun today – so last week’s excuse-making has been seen off, and I’m managing to hold onto the exercise habits I built up during January.
• It was a really unfocused week of piano practice this week – I felt I needed to play more for pleasure than to sort out any specific problems. So I don’t know how the lesson will go in fifteen minutes or so.
• Which reminds me: someone at work asked me “when do you find the time to write this”. In general I’ll write a list of key topics on Friday night – either in Notes on the phone, or maybe in wordpress if I’ve got the laptop to hand. But I simply don’t have the energy or perspective to try and synthesise anything meaningful from it all at that point. That generally comes out when I do the bulk of the writing – which is during Milo’s piano lesson, just before mine. Sometimes I realise I’ve got a lot more to say than I thought – or I spot there are bits of “thinking about thinking” that other people might find useful. Frameworks I realise I’m using, or implicit knowledge I can try to make more explicit. This is how last week’s turned into a bit of an epic, that needed to be finished off on Sunday!
• SXSW has kicked off. I’m slightly missing not being there – but not $10K-levels-of-missing. I’ve signed up for the online streaming/on-demand version, so there’s now an absolute ton of stuff I’ve now got queued up to watch. Obviously I don’t get to see Self Esteem or Baby Queen or Nova Twins from three feet away, nor go to premieres of Sandra Bullock/King Crimson films – but I also don’t get horribly, horribly homesick.