Monday’s commute read: A field guide to dudes who ruin meetings — I actually first read this over the weekend but it generated so much discussion when I shared it on twitter that I re-read it on Monday.
Really productive conversation with a colleague in another department about their plans around open data, and their drivers for establishing better data sources for their service delivery. I find it so useful to have conversations like this, they pull me up out of the weeds and remind me what the challenges are when data is just one part of delivering an effective and efficient service. Love that they have the view that open data is ‘just the right thing to do’, and they’re also really keen to share what they’ve learned on their mission so far.
Did some mind-bending to help with the technical aspects of modelling a new register which needs to link to a whole host of other registers and datasets in a way we haven’t handled before — weighing up the different aspects that need balancing around data modelling, stability and consistency with existing registers, whilst meeting the needs of the users of the data. A lot of core reference data hasn’t been originally conceived with wide use in mind, so the work done to analyse and prepare a register for public feedback can get pretty complex pretty quickly.
Realised, again, that I’m a visual thinker — increasingly find myself translating written or spoken words into visual analogies or diagrams to help me process things. Did some reading about models for making sense of complex systems and caught up on Leisa’s amazing reading list.
Continued some work to explain what we mean by ‘data infrastructure’, and how the products we’re working on right now help towards that vision, and solving real problems.
Tuesday’s commute read: How company culture shapes employee motivation a topic very much on my mind at the moment with two new members joining the team and a couple leaving, and working on a programme that’s growing in an organisation that’s had some changes recently, the downstream impact of company culture is something I’m really mindful of, and lapping up all the articles I can find on the topic, especially any that nudge me beyond what I think I know.
Tuesday started *early*, which meant forgoing my usual treble-snooze (but beating the morning rush at the station), and having cake for breakfast in the office cafe.
The day started with a really useful discussion about where we’re going as a programme, and the most effective ways of getting there. It was really valuable to get some outside views on programme structure and approach, and to abstract away some of the day to day to think about the absolute ideal and how to get there.
This formed a bit of a theme throughout the day, as it started me off thinking about planning, reporting and delivery, and how to make them complementary to (rather than conflicting with) positive team culture. Found this article on How Company Culture Shapes Employee Motivation really helpful and thought-provoking, and referred back to it throughout the day as my thinking progressed.
Had a really productive problem solving meeting with some of the data analysts, developers, designers and product people in Registers to work out how we could protect our fundamental characteristics without making an upcoming register either unusable or untrustworthy. Involved large amounts of me attempting to draw analogies about schema evolution, and a collective realisation that sometimes (perhaps especially in data) the most elegant solution may not actually be the most useful.
The afternoon saw the second session of my new favourite thing: design crit. Working on an infrastructure product can sometimes mean we feel a bit distant from our users in the day to day, and that’s a bad thing.
The infrastructure we’re building still needs to work for users, and still needs to drive interactions and user journeys that work. We started the design crits after learning about them from another team in data infrastructure, and they’ve proven to be a really great way for us to see how we’re supporting user needs, and how the ‘boring but necessary’ data we spend our days thinking about can be interpreted and put to use in real use cases. It’s also an opportunity for our team to look sideways, and understand how design disciplines can inform and support our work, as well as being a chance for our design team to get feedback from a new perspective. I love that these are already becoming an effective two way communication channel, and they are the highlight of my week.
Topped the day off with my first ever attendance at the Product & Service Community meeting, and felt lucky to land on a day when introductions were the focus, through the medium of Top Trumps. It was great to hear the really diverse range of experience and specialisms of others in the community, and I’m only sorry I was too tired to make it to the pub. Next time!
Wednesday’s commute read: Inside the Federal Bureau of Way Too Many Guns this was a tough (long!) read for a couple of reasons: my personal views on gun regulation and record keeping, and the offence to my structured data fangirl sensibilities. Also features an unexpected but v.cool segue into queuing theory!
Wednesday was one of those days that’s really packed and hectic but at the end of it you’re struggling to articulate what you got done. A lot of conversations were had, mostly around whether we might be missing opportunities to align our roadmaps with other teams doing related things to make the most of the people and skills available and check we aren’t reinventing things. I’ve started mapping this out to see what opportunities we might be missing, and plan to start asking others if/how they’ve approached this themselves.
Also had a really productive planning session with the team — we’ve been gently iterating how we do planning for a little while and seem to have found a rhythm and formula which works for us. It’s something like: I set high level product goals that aren’t technology focused, something akin to user stories (although we use job stories too), with an end goal as the ‘definition of done’. The tech lead leads the team to then break these down into the work needed and adds these to the backlog in a ‘forward look’ planning session. During this session the team also tags me into trello cards with any questions or clarifications that they’d need if they were to start the work tomorrow. I then have Wednesday morning to mop up those questions and prioritise the backlog, weed out any I don’t think we need or add any I think are missing, and then on Wednesday afternoons we get back together as a group and plan the next week’s work based on that prioritised backlog.
It sounds like a lot but in practice it’s relatively lightweight and we’ve found it gives us the right balance of focus and product vision, with practitioner autonomy and collaboration so that the team really feel ownership of what they’re working on and can step back ahead of time and see if they need input from others before getting started. It’s also reducing the number of ‘mini-kick-offs’ that happen when someone picks up a story that wasn’t as clear as it could have been, or had too many unanswered questions to be playable. I felt super energised after that planning session, and really positive about inducting two new team members next week.
Thursday’s commute read: 12 signs you’re an extroverted introvert Yeah, it me.
Thursday was another meetingful day, but there were two big wins for me. The first was a really fantastic service mapping workshop run by our service designer — he’s been running them in small groups across the programme all week to ensure a diverse set of perspectives and views. They basically involve looking at a user journey and breaking down the ‘frontstage’, ‘backstage’ and ‘tech’ actions, tools or tasks that underpin it. I found this so useful for stepping back and seeing how our service works in practice, and also for identifying opportunities to make it work better for users and for us.
The second win was last thing in the afternoon, because there’s nothing quite as satisfying as unpicking a complex problem. It was a group effort across product, tech and data science to identify the underlying problem driving the stakeholder’s request, and the ways in which we could (and couldn’t) address that problem. What had started as a seemingly unmoveable impediment eventually became a really interesting research angle for us, and we reached a solution to the immediate problem so that progress could continue.
Friday’s commute reads:
Human Centred Machine Learning — a lot of what this post from Google covers chimes with some stuff I’ve been thinking/drafting a blog post about around product development for data infrastructure. I found it really compelling and shared it with user-centred design colleagues as soon as I got in, to see what they thought.
Man up — there’s nothing in here that’s new to me (sadly) as a woman in the western world in 2017, but it was interesting to hear it from the perspective of someone who’s spent their life in hypermasculine environments and is still finding this stuff out. Caught between frustration that sometimes systemic oppression is only recognised when a member of the oppressive group talks about it, and happiness that the discussion of these issues is starting to get more widespread.
Friday managed to be almost entirely meeting free, which is a precious luxury that allowed me to catch up on a bunch of emails and to-dos, and clear the post-it farm forever growing on my desk/notepad/monitor.