Weeknotes s01e11: design tactics, design strategy
20–26 March 2017
Exploring nav and IA for GP profiles
This week we’ve been thinking about how to improve wayfinding and IA on the new NHS.UK pages we’re creating for GP practices. We have been pursing a long scrolling page design: basically everything about a practice on a single page, with clear headings for each section.
But unlike the information on symptoms and conditions our London colleagues have been working on, the information we’re presenting doesn’t have a natural linear order. There is no narrative to follow, or obvious hierarchy. People come to practice pages for lots of different discrete tasks — from checking what time they can call for an appointment, to leaving feedback, or finding out what their rights are when they’re having trouble registering with a new doctor. There also is some information that is totally irrelevant to most people, but extremely important to a few — like wheelchair access, facilities for people who are deaf or hard of hearing, or information to help people with poor English find a doctor who speaks a language they’re more fluent in.
So we thought we should review whether a long scrolling page was the best approach. In particular we had been using anchor links at the top of the page to help people quickly jump to the information they needed rather than having to scroll. We found in testing that people assumed the links took them to a new page. They then became disoriented by the scrolling and had difficulty spotting specific bits of information — like whether there as a car park.
So from our previous research lab we had a few things we wanted to look at:
- user interface design — what I call the navigation mechanics — tabs, top menus, sidebar menus, sticky nav, category pages etc… etc…
- information architecture — how do we group information into sections and what are those sections called (labelling)?
- information design — within those sections, is the information presented in a way that’s easy to understand and scan?
Finding patterns
I spent some of the week hunting down and collating example of sites tackling similar navigation challenges — thanks to everyone on Twitter who sent suggestions. We’ve been looking at hotels, which similarly need to collate a lot of information about a service (an individual hotel) some of which is only relevant to some guests. I also looked at patterns from the BBC, GOV.UK and local councils. Common patterns were:
2. Making the page visibly scroll rather than jump when using navigation links, so it’s clear to people they have gone to a different point on the same page, and not to a new page (AirBnb, SACO).
3. Creating a ‘hub’ page primarily made up of links to sub-pages and tasks — particularly good if most people are interested in one task or topic at a time (BBC Bitesize, Bristol City Council)
I had some fun capturing the videos of these interactions using GIPHY Capture. I wanted animated gifs as they’re dead easy to share in Slack or Trello. GIPHY capture is a super-simple app that captures what’s happening on any section of the screen and saves the video as an animated gif. You can set it to show or not show the cursor, and there’s a few controls for framerate, image size, and trimming the start or end of the clip. A nice example of a simple app that does one thing really well.
Rapid user research
Wednesday afternoon we went off to a couple of local coffee shops for some pop-up guerrilla testing (we tend to use “pop-up” rather than guerrilla in government circles). Pop-up research is a good way to assess individual interactions and quick tasks like this — anything you can get feedback on in 5 minutes or less. We wanted to see the kind of terms people might use to look up their own GP, and also how the long practice pages worked without any links at the top at all — did people naturally scroll, and did they still get a bit lost?
We structured our questions so we could learn something useful whether people had two minutes or 10. We spoke to about 20 people over the course of an hour and half, and got some useful if unsurprising feedback. It was interesting how different the customers were between the two coffee shops though. The one on the ground floor of our own office building had a largely 30–40 year old professional clientele. The one in the city centre shopping centre had mostly older shoppers, many with low or non-existent digital skills. Going to both was good as it prevented us making assumptions based on insights from a limited user group.
We’re pretty lucky to be working on something as universal as GP surgeries, which means almost everyone we speak to can relate to the task we’re investigating. Checking our assumptions with ‘normal people’ in this way stops us getting bogged down in arguments about our own opinions. I reckon that’s a good return on the investment of a few hours of four team members’ time.
An expert view
I was also lucky enough to be able to tap into the big brain and many years of experience of Caroline Jarrett this week.
Shock news for all of us in the design/UX community was that, hard on the heels of her husband coming out of hospital following a kidney transplant, Caroline herself has been diagnosed with leukaemia. She is currently in hospital receiving treatment, and will be isolated in a hospital room for several weeks.
In her announcement on her website she asked us all to please keep asking her questions about forms and surveys. So I did, and she generously gave me an hour of her time, both for a good gossip, and to give me the benefit of her wisdom on various nav and IA patterns. It was particularly good to see our work through her eyes. After almost 3 months I’m too close to both the problem and our solutions. Caroline’s enthusiastic endorsement helped me see how far we’ve come, and that the page as it is today is nothing like as long and confusing as it maybe was a few weeks ago. Testament to the continual iteration and improvements the team have been making.
As I say Caroline is marooned in hospital for some weeks to come. She tells me she often has some hours each day to do with as she likes, and as you know one of the things she likes best is chatting about forms. If you have questions about form design, survey design, or wider interaction and information design please do send them her way. Just be aware that anything her medical team needs to do takes priority, and she may not always be able to answer quickly.
Design workshop
We tied this all off with a design workshop on Thursday afternoon, identifying two or three options to work up over the next week and a half before another round of lab-based user research on 5th April.
Highlight of the afternoon though was surely using our new Surface Pro screen to capture notes as we went. Things I love:
- Pens that never run out of ink
- Including images, screenshots and photos of paper sketches on the whiteboard
- Live annotating and sketching on top of a web page
- Getting a high-res image of the final whiteboard
The built-in whiteboard app is still a little basic — I really want it to do handwriting recognition, and fake sticky notes would be a nice addition. But it was incredibly useful to be able to share a really good quality version of the notes from the session with the team, and most importantly the product manager who couldn’t be there with us. The image it exported was vastly superior to the usual smartphone photo of a whiteboard, with faded whiteboard pens and scribbled post-it notes.
Top tasks
My other job this week has been first pass analysis of the survey we ran recently. We’ve just about got all 1,150 responses categorised now. I spent Tuesday wrangling Excel functions and cursing pivot tables. I love a good spreadsheet as much as the next former delivery manager and content strategist, but this has frankly taken it all a little far.
Still we’ve got some useful high-level insights from it. In particular we’ve confirmed a major top task we’ve currently dropped from our pages — reading and leaving reviews. I suspect a good chunk of response bias here — if your intent is to leave feedback, then you are going to be more inclined to fill in a survey asking for feedback than someone who is just after a phone number. But we clearly do need to dig into people’s need and motivations around reviews, feedback and complaints in more detail at some point, and hook up with the work already done by NHS England and other colleagues on this.
Old friends and colleagues
As well as Caroline, I also got to catch up with various former DWP colleagues. I swapped nearly but not quite buying a house stories with Mel and Rebecca from the content team, raised a glass to Eddie Shannon as he heads off to his next adventure in government design, had a cathartic lunch with fast-streamer Hannah Roe, and caught up properly with Ben Holliday at last.
It was good to hear that the strategic piece I was working on before I left is bearing fruit. Even better, it’s enabled Ben to make the case to hire a permanent, senior service designer to work on that programme long term.
Making the case for service design and design strategy
A big lesson from my time at DWP is that, when you are trying to make the case for service design to an organisation, there is nothing quite like having someone in there and doing the job to convince people of its value.
It is surprisingly tough to make the case for service design — and more particularly ‘service designers’ and why you should hire them — to people who don’t already understand what it is. Partly it involves a leap in perception and understanding of the role of design and user research in an organisation. Design and research are still perceived as largely tactical, delivery functions. Whereas service design — certainly at senior level — is primarily a strategic one. We’ve successfully got design and research embedded in the day-to-day delivery of digital services — the are we working in the right way level. They’re not yet well embedded at the are we working on the right things level, where they can actually make the most impact.
But also when you try to describe the kind of tasks a service designer might do — end to end journeys, identifying pain points, future vision — people reply that those are things their user researchers, business analysts, technical architects and above all product managers, are (or should be) doing. Why do they need this extra ‘service designer’ person in there?
I think the service designer is there, not because these aren’t things the people you’ve already got can do, but because when tackling big, broad problems having someone who covers all those things and can look at the whole from a ruthlessly user-centred perspective is invaluable.
My thinking on this is similar to David Bland’s three circles. At a product strategy level you want a balance of three perspectives:
- desirable
- feasible
- valuable
with leadership around all three.
Your service designer — or design lead, experience strategist, UX architect or whatever job title we’re using this year — fills the ‘desirable’ function. Of course, in government services ‘desirable’ isn’t the right word. But nonetheless I think the core trio here of product/business, tech/ops and design/user experience is valid.
Importantly, of course, your design lead, tech lead, and product lead must all value and understand the importance of the other circles. The point is not that your product manager doesn’t understand users, or your design lead is incapable of making trade offs between user needs and business goals. If these people don’t understand the importance of all three perspectives I’d question their fitness to lead at this level. The point is to bring together experts on those three areas, and in the healthy tension that will inevitably arise from the particular view you’ve asked each to take of the problem, to identify the very best approach.
Metrics
Blog posts published: 2. Year to date: 8.
Country miles walked: 11 miles.
Wooooot! Spring has well and truly sprung so Saturday we headed to our usual spot at Langsett Reservoir. We made it nearly to the top of Cut Gate, which accounts for 8.7 of the 11 miles. We turned back before the top in case our legs weren’t as up to going 10 miles as our liberated brains were, but we’re determined to make it to the top and the views over Derwent some day soon though. The final 2.3 miles were an early Sunday evening stroll in celebration of the clocks going forward. It being light past 7 is definitely a bonus when trying to get the miles up.
Meditation: been really lax this week. Need to get back into it properly.
Reading/listening/watching: Luke Jones shared this good piece from The Atlantic on The Management Myth (2006) about management ‘science’ and whether you can, in fact, learn just as much about management from reading Descartes and Shakespeare. We tried the original John Wick as part of our 9 mile walk recovery programme. I can confirm that the reviews saying it’s very stylish, very silly, and within those parameters really rather good are pretty much spot on.
