The football at the weekend was good! Also Crystal Palace had a eagle flying from goal to goal before the game kicked off! An eagle! On the pitch! Shame we lost, but 1. Crystal Palace as a first division team would be expected to beat a team from the fourth division, 2. Crystal Palace as a first division team would be expected to beat a team from the fourth division by more than one goal, 3. there’s clearly something good happening at the moment with that Grimsby team. Cannot wait for the league game against Macclesfield this coming weekend. Anyway. How can this week follow that weekend?
Bit of a hectic week, ending the week spinning three plates for the project I am. (The project I am on is a system/product used by DWP staff to administer cases for a form of help DWP offers citizens. Think of it as a product that helps fulfill a wider service.) You’d prefer to take one problem/feature, get that as far as you can (in a reasonable amount of time), test your thinking, then decide to revise what you’re doing — or move on to whatever problem is next. One, not more than one. So, I came into the week hopping between two features:
- Case controls
To avoid a sense of doing everything at once, having a feeling of no-to-little progress and getting brain-ache from constant context switching, on Monday morning I decided to divide my days for the week into two halves. Into a half would be some work for one of the features and one only. This would help me plan — to do manageable amounts of work — but also help the team I am on (there’s about twenty of us) know what I was doing and when they would expect stuff, to review, to test with staff, to see if/how they could be made. People simplify the strategy is delivery to mean get something out there. The strategy is delivery is more about continually doing stuff that moves you forward with the aim of getting the right thing out there. You are always delivering by working with a focus, a purpose.
Last week I had worked with the delivery manager (DM) and lead business analyst (BA) to see when we were planning to deliver the features and then work out how well our designs for the features were working. Were they doing the absolute basic? (And what is “basic”?) Had we got the designs to the point we knew through research they met user needs and were usable? Is there anything we can build? What work do we need to do to get to that point?
To help out, to spread the load, one of the developers — Gary — had pulled together a couple of screens for letters. In the nicest way we said these would be like a version zero, a sketch of what we knew and thought. We can then use the design zero as a talking point for a few members of the team to find gaps and problems. I’ve tried to instil in the team it’s better to talk around some thing than to talk in the abstract. Some simple rules:
- Doing not talking.
- Show the thing.
- Make the thing better together.
Gary just gave me a heads up before he got on: Was I OK with him pulling some thing together if I was so busy and we’ll get together when we’ve all got an hour in our calendar to make whatever we had better? Of course, but by Gary asking out loud we knew what was happening. (Good communication is clear communication.) Gary’s work on letters gave me the chance to concentrate on changes to deductions and charges (similarly to get our design zero down).
Once I was clear at midday on Monday I took Gary’s letters starting point and set my status on Slack to a form of do not disturb.
I put my headphones on, opened my pad, typed
npm start into a terminal and off I went creating some in-browser prototypes to talk round. Here’s some grabs of some of the work I did on Monday afternoon. This is just for the “letters index”. (We’ll look at the rest of the journey for this task for users another time.)
If that is too small, view a larger version. [5MB]
Here’s what those screens are, left to right:
The original list made by Gary, the links take users to a page that allows them to generate a letter. (Remember: We’ll look at the following bits of this journey another time.)
My first iterations, a revision of number 1, but with the list divided into more groups. My reckon was this was a very, very, very, very basic level of acceptable. Think of it as tin if you had a tin/bronze/silver/gold grading of features.
Taking number 1, I spent 15 minutes looking over the original list and trying to work out what scenarios they represented. You can see the matrix. I then went over this with a couple of people in the team (Mark and Matt). In the first review it turned out I had missed two things, in the second review another thing missed popped up. Always always review your work with other people, especially when you are working quickly.
I then used the matrix to quickly pull together a more “exploded’ table layout of the links. The content of the link in the starting screen were very compact: I was looking to make the options more scannable, clearer to read.
Here’s four more screens:
If that is too small, view a larger version. [3MB]
Here’s what those screens are, left to right:
This is a copy of screen 4 from the previous collection of screen grabs. (These images are taken from a presentation I did.)
This is number 5. Using some simple logic (worked out from the matrix) I created a scenario: the citizen is English and they have not nominated an appointee (which is indicated to the user with a message on screen). And because of the matrix work we know some letters do not need to accessed in this scenario. We could know this from data about the case. We can make this list context appropriate for the member of staff, reducing their options to only ones that would be necessary in this scenario. (There are three other scenarios.) If the DWP staff needed any of the other letters from the fuller list we’ll give them that option, as a link (which goes to screen 4).
Some type/layout tweaks: Spacing and drawing out a message that tells the member of staff about English/Welsh and post to citizen/appointee.
Removed the column indicating “English”: It is indicated in the message on screen all communication is in English and there are no situations where the case might need English and Welsh communication.
We reviewed the work as a team, found some possible issues, but I started planning going through this work with staff, find what doesn’t work and what does work for them as the people who would use this. Monday done. Tuesday morning I could explore the rest of the journey.
Everything was going to plan until the afternoon of Tuesday: I suddenly had another plate. Two become three with the addition of changes to deductions and charges. The three features are now:
- Case controls
- Changes to deductions and charges
There was too much to do. We needed to take a pragmatic approach to our work. Again working with the DM and the lead BA we did some analysis of where we were, what was needed, when, how, why. Essentially though: What can wait? We worked it out: changes to deductions and charges was the priority as we knew far less about it than the other two features. We needed to do some design zero work there to get that going. So we did. We huddled for a “briefing” of the problem, talked through a possible starting point, and I went away to pull something else together to talk through the next day. Hopefully we’ll go over some of that work in next week’s notes.
It felt good the team had a show and tell later in the week, to go through the design and technical work on letters. We haven’t had a show and tell in months. The absence has pained me. And in that twenty minutes together we showed to the wider team how each bit of work on letters complemented the rest as we’d worked together and shared our thinking and what we were doing. Together it showed a possible roadmap for the feature. Doing the work we all agree it’s not about sticking the very first thing we do out there: We now have a degree of knowing what is good enough to be released and the possibilities of what is better afterwards.
The rest of the week was ping-ponging between those three features. It’ll be more of the same next week but hopefully we’ll be on top of it. More good teamworking like this week makes it better. Fries yer brains though dunnit.
Elsewhere in the week I had catch ups with a couple of designers. I try to do catch-ups with at least two designers a week, preferably three, each lasting between 30 and 60 minutes. Sometimes these aim for a few targets: How they getting on in life, what work have they been doing, how have they gone about it, how did they find it. Sometimes these are one-on-one reviews of work, talking through problems. Whatever the designers need though that time is for them.
A few times this week heard/read people fighting for their work, fighting so something happens. Ur. Fight. I’ve spoken and written before about my aversion to the word ‘fight’. It’s the language of conflict. There are better words, more constructive words. Maybe a post I wrote three years ago helps explain my thinking.
I tell you what though: HDMI cables are hard to come back in our place. Gotta keep a tight leash on any you have. Also: Toilet roll.
I’ve been trying to make sure I go for a walk every lunchtime. Stretch the legs and get some outside air into my lungs. One lunch this week I had a quick look at the Year of the Dog exhibition at Colours May Vary (which is just round the corner from Quarry House, where I work). I took some snaps and put them on Instagram.
Finished a re-read of Tim Harford’s Adapt. I dragged a quote out of it for a Leeds Gov Design meet a couple of months back and been reading a chapter here and there since. It was lingering as unread in my Kindle. Now it is not.
Started reading The Entrepreneurial State.
Someone (might have been Iain Broome) mentioned on Twitter he’d been reading 25 pages a day, which used to be my minimum daily aim until last year. Going to aim for that again. Still tracking stuff on Goodreads.
Got through a couple more episodes of the excellent first season of Runaways. Just two more episodes left.
Next week continuing the plate spinning, and getting responses to this week’s work from users. There’s a day trip to Blackpool to see the designers and teams there, and a DWP Digital design systems meet in London on Friday.
Got to gather as many of the header and footer options on the DWP Digital products and services and get the first draft of the combined thinking down to get further input. (We’ve got an issue already set up in our [design examples log(https://github.com/dwp/design-examples/issues) to record all this at.) And do the same with How Do We Tell Users There’s Been A Change This Product/Service (which will need a new issue setting up).
Keep an eye on prep for the DWP Digital interaction design meet in a couple of weeks.
Prep for Leeds Gov Design meet #7, which is at the end of this month.
Prep for Leeds Service Jam, which is at the end of March.
Ch-check it out
Real stories about getting stakeholder buy in to fund UX by Carl Burton, the head of product over at Co-op Digital.
Erika Hall’s excellent Research Questions Are Not Interview Questions.
A fine thread on Twitter from Tom Bridges off the back of the NY Times piece What if Cities Are No Longer the Land of Opportunity for Low-Skilled Workers?.
The rabbit hole of what people put in their bags.
And to mark it being 20 years since the Sopranos started HBO’s Twitter account was offering people their Sopranos name.
Brought to you by…
A mug of green tea and The Fall.