Weeknotes s01e05: what do they really want?

6–12 February 2017

Yeah we’ve skipped a couple of episodes. I decided I couldn’t go on being two weeks behind. I may go back and fill in episodes 3 and 4 sometime, but probably best not to hold your breath. Think of it as an unfortunate Sky+ box malfunction.

The week was dominated by a lab-based user research session on Monday, and the write-up and analysis for that. Rather than do a play-by-play of the week — which largely goes: do lab, analysis, analysis, write-up, write-up — I thought I’d talk about the research itself, the insights we’re gathering, and some reflections on process.


The insights mostly confirmed what we already knew, though in a couple of places I think we got some extra depth of insight about what’s really going on when people look for a new GP:

People’s first priority when choosing a GP is convenient location

That’s not just proximity but also how easy it is to get to. They might prefer one that’s a little further away but is a pleasant walk along back streets, over the nearer one that involves crossing two busy roads. We’ve had people say they choose their GP partly because it was on the way to the supermarket.

After that people rely of personal recommendations from people they know

They ask the neighbours. They accompany a relative to an appointment and think how much nicer that surgery is than their own.

Ratings and reviews are no substitute for that personal recommendation

The people we spoke to were mostly at best ambivalent and in one case openly skeptical and mistrustful of online reviews and ratings. “Anyone could write that” one said (quite rightly) .

Official stats don’t fare much better.

“Will I be able to get an appointment” looms large

Of our 4 users, 3 were changing GP not because they were moving house, but because they had had a bad experience with their current practice — usually involving it being difficult to get an appointment.

The personal interaction with staff — especially receptionists and doctors — is really important

We asked people to pick out the 5 most useful stats from the bi-annual GP Patient Survey. They overwhelmingly choose helpful receptionists, and doctors who take the time to listen and explain things. Despite all the discussion about how difficult and awkward it can be to get an appointment, those things were more important. However, that does need to be set in the context that the receptionist is seen as the gatekeeper to appointments. This is only 4 people though, so we need to do some larger scale followup to validate.

When they look at things like how many doctors are at a surgery, how many patients, when is reception open etc… it’s often a proxy for this softer stuff about human interaction

For example a perception that smaller surgeries will be more friendly and personal. Or that it will be easier to get an appointment at a large surgery because there are more doctors available. Or a helpful receptionist with help you get the appointment you need. In short there’s a lot of extrapolation going on when people look at the factual information currently available about GPS, which goes something like:

Users make assumptions and extrapolate meaning from limited facts. Digging into the reasons why they choose particular facts reveals the deeper need they are really trying to meet.

People are trying to derive meaning from dry facts. They’re making assumptions based on those facts about whether a particular GP practice will meet their needs, which may not be correct.

This is a good example of why you need to really dig into the whys behind people’s surface reaction to a design, or their ‘expressed needs’. The surface ‘need’ is “I want to know how many doctors there are”, “I want to know the doctor to patient ratio of the practice”. But that’s not the really the need they’re trying to meet when they look at those numbers.

Low take up of digital services like online appointment booking is due to a double-whammy of not knowing it’s an option and once you do know it’s an option, finding the system difficult to use

Of the latter problems registering in the first place, and having yet another ID and password to remember, have come up a lot across our research.


We got some good stuff out of the research, even though overall it was scrappier than I’d have liked due to last week’s compressed prep. We also had a no-show which was frustrating.

It was the first lab we’d done together as a largely new team. That means our processes and ways of working together as a team aren’t very mature, which compounded the problems of lack of prep time. But I’m confident we can get this all humming along nicely pretty soon — it’s just a matter of time and repetition.

On the process front things that worked well were:

  • Allowing half an hour for reflection and analysis between each session. It meant not least that everyone had an opportunity to reflect on what they’d just seen, even if they weren’t present next day for analysis. Also to capture extra notes or clarify wording. I think this was worth it even though it meant we could only fit in 5 sessions, not 6.
  • Using a different coloured post-it for each participant. Makes it much easier to spot a) duplicates and b) things that appear to be patterns but are actually just one person going on about a topic at length (“it looks like we have lots to support this insight, but all 10 post-its are pink, so it’s not a pattern”)
  • Getting people to include time-codes on their stickies so it’s easier to find video clips later.

Things I’d refine for next time:

  • Try getting different observers to focus on different things — for example one person capturing quotes, one capturing emotional responses, another capturing specifically what they do not what they say. The latter is often about spotting quite subtle things — moments of hesitation, excessive or insufficient scrolling, small cues which suggest content or interactions aren’t as clear as they might be. Because they’re small, and the user is still managing to complete the task, these opportunities to refine the design aren’t always captured.
  • Have an idea of the high-level themes observations will fall into already identified and up on the wall. That would make sticking up notes between sessions faster and clearer. I’ve relied on print outs of the screens in the past, but as a chunk of each session was a broader interview we could have done with some extra headings or print outs of non-prototype pages like Google searches.
  • Give ourselves more time to set up first thing. We’d allowed half an hour which thanks to Monday traffic making several of us late wasn’t enough.

and finally…

I’d say the rest of the week was spent as much trying to do analysis and write-up, as actually doing it. Odds and sods of meetings and discussions kept punctuating my attempts. The meetings were all relevant and useful but inevitably the context switching and lack of a good run of time to really get into the flow of analysis and write-up meant less than an ideally productive week. It’s not just for making things that you need maker time.


Elsewhere this week I let Andy trick me into visiting a guitar cave in his eternal, fruitless quest for the “perfect” acoustic guitar.

Also had some early discussions about the idea of writing a book on designing user journeys. I am torn between “OMG! Someone wants me to write a book! I must totally do it!” and [Looks at diary] “OMG! Doing All The Things! How am I ever going to find time to write a book as well?! I must be mad!”.

What’s next?

The book dilemma lead to more existential discussions of where I should be going next. In my talk at Dare Conference a couple of years ago I cited Neil Gaiman’s advice about doing things that get you closer to the mountain. Unfortunately this requires knowing what mountain it is you want to reach. Both personally and professionally I’ve basically climbed the mountains I’d been aiming at for a long time.

So I have a dilemma:

  • Do I aim for the specialist “Head of…” mountain building a team skilled in a particular branch of user-centred design? But which branch: interaction, content, service design, research? Tough when your strength is as a generalist.
  • So I might rather head for a broader, more strategic “Head of UX/User-Centred Design” mountain, working out how those specialisms come together to deliver better services, probably alongside specialist leads.
  • Or do I go broader still, more into product and service design strategy, or programme delivery, leveraging my generalist strengths and ability to understand not just users and design, but also business and technology?

And within each of those:

  • Do I continue taking a series of longish (6+ months) 4–5 day a week contract/interim/freelance roles?
  • Would I rather build a consultancy business providing short-term and ad hoc training, workshops and consultancy for multiple clients? I do some of this now, but there’s a limit to how much I can take on alongside fairly full-time contracts. I’d want to free up at least 2 days a week from contract work to grow that side of my business.
  • Would I consider going perm somewhere? Maybe for the absolute perfect role — but that requires knowing that the perfect role is.

The fact that all those options may be open to me is a great privilege, but I’m reminded again that constraints and a certain lack of options can be more useful than they seem.


Country Miles Walked: zero.

Meditation run streak: I haven’t been as diligent with this towards the end of the week. I’ve also been trying to do the “going to sleep” medication without listening to the guided recording in the app, without that much success. My mind wanders so much I lose track of the exercise. I think I’m still having to concentrate too much on remembering how the exercise goes. Back to using the app I think.

Reading/watching/listening: we started on The Strain. Or rather I started on The Strain. Andy had been watching during the day before Christmas and had got to season 3. He didn’t mind re-watching it so we’re now happily box-setting our way through it. Overall really good and thoroughly recommended. A modern take on accidental kick-arse vampire hunters with a nicely developing plot. I’m particularly pleased to find well-drawn characters who manage to all have personalities, not be all holier-than-thou, and yet not just go around being nasty to one another all the time.

And I reckon I could totally rock Nora’s sword-slung-over-the-shoulder look in meetings…

Nora from The Strain rocking the sword-wielding vampire hunter look, with oh-so-on-trend distressed leather cross-body shoulder holster.