90% archaeology: my notes and reflections on Service Design in Government 2015

Some rights reserved - Jess McMullin

Some rights reserved – Jess McMullin

There’s never been a more exciting time to be designing services in the public sector. But it can still be a lonely existence – in any organisation, a small number of advocates may find themselves trying to shift a large mass with plenty of inertia. The Service Design in Government conference that I attended last week has an important role to play. It’s a place where people can share their triumphs and frustrations, and form a common view of what we should be aiming for in the design and redesign of public service.

Thanks again to all the speakers, the other participants, and the organisers at Software Acumen. I was delighted to be part of the programme committee. These are my notes and reflections…

Everyone’s a designer.

Jess McMullin and Alex MacLennan have been building service design practice in the government of British Columbia since 2010. Along the way they’ve developed an awe-inspiring array of interventions across the government’s many services, intensively training cross-disciplinary, cross-department teams and moving up the design maturity ladder – from no conscious design, through a focus on style, function and form, up to using design to solve already-identified problems and frame new ones. Recognising that their own team is small (“we are not the official designers for the BC government”) they focus on getting other people to think like designers. Now they have a UX library that all ministries can use.

Those Hogwarts Moments – I love those too…

Anna Whicher and Adrian O’Donoghue carried on the “public servant as designer” theme with the story of the trans-national SPIDER project, and its application in Ireland’s Northern and Western Regional Assembly. An ambitious peer learning platform between local governments across North-western Europe, it covered public service co-ordination, youth unemployment, ageing populations and culture change within authorities. The scale of the capacity building is impressive: 1478 people attending local workshops, working on issues where people will benefit. Stand-out quotes: “Co-production works. It scares the public sector” and “Hero designer is not suitable in this way of working.”

Several other talks picked up the same themes:

  • Housing manager Amanda Pujol worked with designer Kathryn Woolf under the Design Council’s leadership programme to prevent life-threatening trips and falls among older people in Teignbridge (and together they bubble wrapped a whole GP’s reception to dramatise the issue!)
  • Transport for London, with Ben Reason‘s Live|Work, seconded 20 station staff to facilitate workshops, bringing an honesty and credibility that could only come from frontline workers
  • Gavin Bell of the Ministry of Justice told how they seconded a deputy prison governor to work with digital specialists on the Prison Visit Booking exemplar
  • Jean Mutton, an “inside-out service designer” at Derby University, brings students into her team on 12-month paid internships because “we get a much richer picture from students talking to students”
  • The Satori Lab‘s Jo Carter and Esko Reinikainen got housing association staff across Wales into conversation with the world cafe method
  • For sexual health community interest company SH:24, Glyn Parry and Unboxed Consulting‘s Martyn Evans had to create a core team including clinicians, public health professionals, agile project managers, designers and developers.

It’s 90% archaeology.

Louise Downe, service design lead at the Government Digital Service, outlined how  service failure is still one of the biggest costs in government. Time is taken up with unnecessary processing – roughly 40% of people declaring medical conditions to DVLA have a condition they didn’t need to report – user contact, casework, and manual handling of exceptions in policy. Change needs to happen “in hearts and minds of everyone who works in government. It’s not sitting in a room and ideating, it’s finding out why we’re doing what we’re doing.”

  • On the SPIDER project, this became a kind of “double ethnography on the end users and the system to understand how you can make an impact”.
  • Gavin detailed his own learning curve at MoJ – all that paper, printing and cabling! “It’s not done until you follow the transaction all way through the court process… It’s not done until the user has finished paying their fine… Understand the context in which you’re operating, understand how to get all the way to the end.”
  • Jean told us the story of the bicycle books, dutifully filled out and filed for decades even though their purpose in administering wartime rationing was long since redundant.
  • Jess and Alex “unearthing the decisions of the people that came before us”, connecting with the experience and legislation cultures, norms, values and power structures: “There is no more powerful tool than the road trip.”
  • Glyn and Martyn working with NHS trust information governance boards to devise a better way of keeping personal information by anonymising it instead of sending it around the system in the mail.

For TfL, it was about how to improve the customer journey (literally!) in creaking, crowded stations on a 152-year-old network that last saw a full day of good service on all lines in April 2010. (and in heritage buildings to boot – Earl’s Court Station, “beautiful escalator!”)

Change is hard.

TfL used the inspiration of the 2012 Olympics to prove that things were possible. But to put this into practice, they used a design approach to engage staff, build readiness for further change and reduce potential conflict (avert just half a day of Tube strike and the initiative would pay for itself.) Covent Garden station supervisor Pele Bapere told a powerful story about his own role in the “reachback” communications to colleagues at the station: “I’ve worked for the Underground 16 years and I’ve seen many things brought in… This was the first time they’ve gone out in a systematic way and engaged staff.” Sure enough, staff repayed that investment by highlighting priorities and changes that could be put into action quickly, such as modernising the approach to lost property.

And there were loads of other great change tips from presenters:

  • “Make sure you have ‘Do-ers’, not strategists” – Amanda Pujol
  • “Design language to some is really not helpful” – Andrea Siodmok, Cabinet Office Policy Lab
  • “The Gov Whisperer: not a change manager, more empathetic and focused on the needs of government” – Jess and Alex
  • “Intelligent challenge: Can you help us understand?” – Gavin
  • “Epic failure: Using empathy tools with psychopathic organisations” – Jo and Esko

Design is often first to join the dots.

The work of people-centred change frequently starts by helping those who do only a small part of the process to see the whole picture from the end-user’s point of view. Shockingly this often doesn’t happen until designers get involved. That’s what Jean Mutton did at Derby when she issued Flip cameras to new first year students – she helped the university to move from a component process review to the holistic student experience. Her team found departments tripping over each other to send letters, each of which “wanted to be the first” to welcome the student. And they developed a 40-point action plan that covered everything from signage to staff awareness.

Perhaps the most compelling story of joined up service – not just between organisations but across sectors – was the story Pele told from Covent Garden Station. With the opening of Britain’s largest Apple Store just across the road, there was an increase in the number of blind and partially sighted people coming through the station. So tube workers made an arrangement to call Apple, who now send a staff-member to meet people at ground level and escort them through the crowds.

Now shift from actual, physical underground railway platforms to the digital metaphor of the moment, “government as a platform”. Imagine, as Louise is starting to, how we could work when it is easier and quicker to make better, more user-focused public services: “When services are easier to make we’ll probably have more of them, not less.” But they’ll be “made of the internet”, “small pieces loosely joined”. Rather than having one monolithic piece of the benefit system, we can create a customised user journey that meets people’s individual needs. The potential is massive, but we’ll have to up our game. We’ll need a new clarity of thinking, not just “if we can’t fix it with a form, we create a portal.”

Needs are diverse, complex and quirky.

TfL’s Pele: “This is not Singapore, this is not New York, this in London, we’re quirky.”

Here’s a “best of discovering user needs” compilation I jotted down at the conference:

  • A GP decided to go through the front door of his surgery as if a customer – and re-worked his reception area based on what he learned – Amanda
  • Some prisoners have complex love lives – the prison visit booking system can end up being the forum where their rivalries play out – Gavin
  • If the admissions department puts its address on letters, that’s where students will show up on their first day – not the campus 20 miles away – Jean
  • There’s a new digital divide – between people online with basic skills and the “next gen” users for whom internet is interwoven with life – Liz Stevenson of Cambridgeshire County Council
  • Look for the verbs: “Bad services are nouns, good service are verbs.” – Louise

Everyone has tools, but prototyping is where it’s at.

I lost count of the number of toolkits, frameworks, canvases, cards, and variations on a process that showed up during the conference. (Top marks to Satori Labs for their “double diamond with knobs on”.) And there seemed to be general agreement that toolkits have their place.

But meeting up on the Thursday evening with a fifth column of Global Service Jam and Global GovJam hosts from around the world, I was reminded that for all the analysis we need to keep public service design real.

So it was great to hear references to prototyping – including one reportedly by Home Secretary Theresa May. Andrea: “If in a year people are talking about prototyping and they weren’t before, we’ve made an impact.”

  • In British Columbia, Alex runs a Public Services Dragon’s Den with a budget for creating prototypes and pilots
  • In Teignbridge, the bubble-wrapped reception was one of three alternatives tested for real in GP’s waiting rooms before settling on a single direction to raise older people’s awareness of trip and fall risks
  • Smeared and splattered iPhone screens made it bleeding obvious to the SH:24 team that video wasn’t the best way of showing people how to give a blood sample.

What did I miss?

Service Design in Government was a dual-track conference with a line-up so good that I inevitably didn’t get to see all the great presentations, like the one on design patterns by GDS colleagues Caroline Jarrett and Tim Pauland Lesley Thomson‘s reflections on design in the Scottish Government. I’m sure there was more that I missed!

See also:

The Last Target Operating Model You’ll Ever Need™

I first wrote this as a comment on Joel Bailey’s excellent blog post titled ‘This thing called agile might kill us all’ but thought it worth re-hashing and expanding here.

For context, Joel writes about “working for a big high street bank. The brief is to redesign the ‘end to end mortgage experience’. The timescale is to reach a business case, with a roadmap of delivery waves to achieve minimum viable product, within 6 weeks. ”

He floats the idea of a Target Customer Experience as counterpoint to that change management staple, the Target Operating Model.

I’ve had recent experience with a “TOM”, attempting to intercept with an agile, digital project. It left me puzzled, and I’m grateful to Joel’s post for helping me clarify my unease.

In case you haven’t come across one before, the TOM is a Thing in the world of “change management,” defined on Wikipedia as:

a description of the desired state of the operations of a business. Typically a TOM also includes the roadmap over time that specifies what the company needs to do to move from the “as is” to the “to be”.

For the service designers among you, a typical TOM covers similar turf to Alex Osterwalder’s Business Model Canvas, only with fewer sticky notes and more spreadsheets.

As an aside on his nascent agile project, Joel writes about the toll it takes on participants:

someone needs to write a Marxist evaluation of agile. Yes the outcome is better and it’s all very sexy and new and ‘oh so right’, but I suspect the cost on the worker is high as essentially it speeds production and works the asset of production (you and me) harder.

… which immediately set me thinking that if people are using “agile” to mean doing the same process only faster, even at the risk of burning out their people, then they’re Doing It Wrong.

I reached for the 8th of the Agile Manifesto Principles:

Agile processes promote sustainable development. The sponsors, developers, and users should be able to maintain a constant pace indefinitely.

And that’s when I realised the real challenge to peddlers of TOMs and the like: agile transformation isn’t a one-off thing that you do to get from A to B – it’s a continuous culture of iterative improvement.

Agile organisations succeed through sensing, not planning.

  • They are in touch with their actual customer experience (not just some brand fantasy). This is the dirty secret of much Target Operating Model work. A warts-and-all “as is” picture is far more valuable than any amount of “to be” prognostication – but even if that’s what executives secretly wish for, no consultant can afford to say out loud “I’ll tell you the time if you show me your watch”. Sadly the picture TOM processes do generate is often missing empathy, the key ingredient that spurs the organisation’s people on to make things better for their customers.
  • They truly understand their operating model (clue: it won’t look like a flow chart). Organisations are nothing more than systems made of, and by, people. They’re complex social constructs that operate on emotional as well as financial planes. This is what agile understands when it says “individuals and interactions over processes and tools”. To map an organisation by decomposition is to follow in the footsteps of the early Cartesians, dissecting a dog to prove it has no soul.
  • They have the capacity to make very frequent adaptations in response to their ever-growing understanding of customer needs. Being able to respond quickly to what you learn beats any amount of predicting and planning. Embracing diversity means pushing decision-making to the frontline. This in turn reduces the waste inherent in standardised processes. Let’s cultivate this as a core competency of every organisation. If we never get stuck in a rut, we’ll never require a “change programme” to jolt us out again – and that should come as a relief to all concerned.

All of this poses problems to an organisation addicted to discontinuous change. We’ll have to break down the Berlin Wall between the bits of an organisation that create “strategy” and the bits that do “operations”. Likely, product development can no longer be capitalised, so the balance sheet might appear worse before it gets better.

But I’ve come to the conclusion that this is the only sane way to run an organisation.

Learning by doing: it’s the Last Target Operating Model You’ll Ever Need™

Seeing over the next hill – a service design pattern

Over the years I’ve worked with digital services in different spaces, from sports performance to house buying to students on campus and training in the workplace. And there’s this one picture that resurfaces in service after service. I need to get it out of my head and into the world, where I hope others will help me develop it further.

image

It’s a picture of a pattern that goes something like this:

Seeing over the next hill

We meet much of the most valuable service when facing a change or challenge for the first time. But unless we know what to expect, it’s hard for us to make decisions in our best interests, or to trust others seeking to support us.

Sometimes the change is related to a life-stage – choosing a school or college, having a baby, retiring from work. Sometimes it’s simply something we may only do a few times in a lifetime – buying a car or home, opening a bank account or reporting a crime.

On one side, the person at the centre of these events has lots to think about, many (possibly conflicting) choices, short and long-term implications to consider.

On the other, the people delivering service are likely to see and do the same things time and time again. Their experience is valuable, but can easily give rise to jargon and preconceptions that obstruct communication and empathy with first-time users. Often the first step to improving service is to recognise and reduce this asymmetry of understanding.

Deliver service so that people can always see over the next hill, so they know what to expect, what good looks like, and who they can trust to help them along the journey. Specifically:

  • Cultivate empathy among people who deliver and design service day in day out. Find ways for them to see the service through the fresh eyes of a first-time user as frequently as possible.
  • Put first-time user personas at the centre of your work. Ask yourself what will shape their expectations and how might those differ from the way insiders perceive the service?
  • Test your service with people who have never seen it before. Over and over again. They can’t unsee what they’ve seen so it has to be new participants every time.
  • Follow up with service users before, during and after their experiences. See how their needs, wants and behaviours change as they go through their journey.
  • Record real-life experiences to share with future users. It’s much more compelling to hear from someone like you who has been there before.

As with all the reckons I post to this blog, I’d love to know what you think. Have you seen this pattern too? Who handles it well? What else could we do with it? Or am I making a mountain out of a molehill?

The Lost Robot Manoeuvre

The lovely thing about designing for service is the intangibility. You can prototype it in conversations. You can act it out. No tin required – the virtual is so much more pliable.

Then again, the maddening thing about designing for service is the intangibility. People have trouble getting their heads round it. How will service interact with users? How will it meet their needs? The solid is so much more familiar.

To re-tie the frayed ends of this creative tension, I’ve found myself using a technique that deliberately introduces a physical actor into the process, a service avatar to stand in for the stuff we can’t see.

Useful robots workshop

The Lost Robot Manoeuvre emerged by accident when Marc Fabri asked me to run a service design workshop for students as part of Leeds Met’s Futures Fest. The ever-inspiring Emma Bearman suggested that we link it with her March of the Robots series.

At first the robot felt like a cuckoo in the nest; I still wanted to talk about intangibles. But as I developed the workshop plan I realised it could be a powerful thought experiment.

robot by Ludo

Put simply, the method goes like this…

  1. Quick, draw a robot, a robot to help people. Work out what problems it solves. Maybe write some user stories.
  2. Take your robot out of the building. It’s a great conversation starter for some guerilla research. Re-write the user stories based on what you learned. Re-draw the robot.
  3. Now pivot. Lose the robot. It never existed anyway. But what if you met those needs with service instead?

THERE IS NO ROBOT. (SORRY.)

My guinea pig participants rose admirably to the challenge. One group created a robot to help their fellow students de-stress at exam times, The other focused on exercise and encouraging people to be active. In both cases the robot was the starting point, but not the end.

I put the workshop outline up on Speakerdeck. I’d love to run it again some time if anyone will let me…

It can be these, but…

Our economy will not grow bigger in scale, but we will see it become more specific, more diverse, more adapted to individual needs and desires. The economy that served us well is giving way to what I call the informative economy.

According to my dictionary, “to inform” means to “imbue or inspire with some specific quality or value.” Practically speaking, information is not merely data, telecommunications, or a computer network. It can be these, but it is also the knowledge added to resources to make them valuable. It is design, craft, utility and durability — everything that makes a product more useful, longer lasting, easier to repair, lighter, stronger, and less energy-consuming. Information is nothing more (or less) than how to make or accomplish something the best way.

A Chevrolet requires ten to twelve times more expense on warranty repairs than an American-built Honda does. The difference is information in the form of design, workmanship and quality. Twenty-five years ago Honda was a “small” business. It became a big business not by building bigger cars, or cars with more gadgets, but by building a car with more information…

— Paul Hawken‘Growing a business’ (1987)

Thanks to Andy Bell for the recommendation.

Which part of “the customer is always a co-producer” don’t these people understand?

For the third time in the past few months I’m assailed by a survey so shockingly poor that I wonder why the service provider in question has bothered at all.

First it was East Coast trains with a lengthy paper questionnaire about my journey, conducted entirely in mind-boggling forced-choice price/quality trade-offs.

Then came a letter from an Ofsted inspector slipped into my child’s book bag at primary school. “Your views about the school are important to us,” said the letter. The less-than-24-hours’ notice to go online and complete a survey suggested otherwise.

This time, as I log out of my online account, my bank butts in with an entreaty to help them develop new features. Like this one…

Imagine you could search and sort through transactions...

Let’s leave aside the dubious value of any question in user research starting “imagine if…” We’ll also charitably disregard the fact that all the bright ideas my bank is asking about have been standard features of their competitors since the days when the Internet sounded like a fax machine.

What really winds me up about this – and the examples before it – is the complete absence of a space to explain or qualify my choices in free text.

The East Coast one went on for 14 A4 pages without so much as a simple text box for me to have my say.

And when Ofsted states

By sharing your views, you’ll be helping your child’s school to improve. You will also be able to see what other parents have said about your child’s school.

… they don’t actually mean said the way you or I, or a child in Key Stage 1, would understand the word. What they mean is clicked. Only strengths of agreement/disagreement and yes/no answers are permitted.

I’m not suggesting that large-scale, structured surveys are bad in themselves. But I do believe that asking any question without listening properly to the rich, human voice of the respondent does a disservice to surveyor and surveyed alike.

At the organisational level, asking only closed questions runs risks in two directions – gaining false reassurance or prematurely discounting profitable opportunities. In the bank example above, I do indeed value searching and sorting through my transactions, but much prefer to do so in an Excel spreadsheet or separate online personal finance service rather than on my bank’s own website. How am I meant to convey this subtlety in the survey? And how are the bank’s service managers to know this is what I want?

Maybe you think I’m only seeing half of the picture. Perhaps these three organisations also have sophisticated qualitative programmes wide open to unstructured feedback. Statistically speaking, I’m much more likely to be tapped up for ten minutes doing a quick online survey than for participation in an in-depth interview or ethnographic study.

Actually this make things worse, not better.

Consider the disempowering message sent to the thousands of travellers, parents and bank account holders on the blunt end of closed choice questionnaires. In signing off those questions, managers have assumed the sole right to structure the terms of conversation with the customers who are surveyed. “We want to know what you think,” they say, “but only so long as it fits within the narrow confines of our pre-existing plans and prejudices.” It’s as if they’ve rolled out the welcome mat to invite you into the conversation, only to snatch it away from under your feet.

Service dominant logic demands a dialogue, a collaborative learning effort between customers and service providers. In their essay ‘Co-creating the voice of the customer’, Bernie Jaworski and Ajay K. Kohli list the following features of a co-creating dialogue:

  • Is the conversation end point clear or unclear?
  • Do the comments build on those that came before them?
  • Is there a willingness to explore assumptions that underlie the dialogue?
  • Is the conversation exploratory: no topic is “off-limits?”
  • Is there an eagerness for new ideas?
  • Do the firm and the customer each shape the structure and content of the conversation?

It’s hard to do any of these things in a smash-and-grab raid to snatch a few data points on a five-point scale.

In 2014, organisations have no excuse for behaving so oafishly.

  • If you really need to ask closed choice questions, add an optional space where people can explain or clarify their answers. It shows you might be genuinely listening, not just engaged in a box-ticking exercise.
  • Worried you’ll be overwhelmed with more answers than you can read? What a great problem to have. Throw all the answers into a tool like Wordle so you can at least see common terms that crop up time and again.
  • Instead of a big survey upfront, try to gather user input a little and often. Ask for micro-feedback at relevant points in the user journey. That way you can adapt your questioning to context and find precisely the users who are grappling with the issues you want to know more about.
  • Spread the conversation out through your service design process. Think of every survey as a chance to recruit and screen users for deeper collaboration at the next stage. You may be surprised how many are prepared to give contact details for follow-up discussion on interesting findings.
  • Above all, keep an open mind – which is much easier to do when you ask an open question.

Update, February 2015: Inch by inch, the things I complained about a year ago are getting better:

  • My bank now does give survey respondents the chance to explain their answers to its closed choice questions. I intend to use this soon to give them my views on their recent “upgrade”, which seems to extend only to deploying a new and less legible webfont.
  • Ofsted reports (on page 29 of a PDF, sorry): “We are updating ParentView in response to feedback. Following a trial in the autumn term 2014, we will introduce a free text box, from September 2015, that will give parents the space to raise any specific issues.” Welcome news – though if a school that took that long to change, it would surely be in special measures by now.
  • My train company is approaching the end of the line, having been replaced by a private-sector bidder. Here’s hoping Virgin and Stagecoach are better at listening when they take over the trainsets next month.

In praise of the good enough

… what the designers and engineers see as “pain points” aren’t necessarily that painful for people. The term satisficing, coined by Herbert Simon in 1956 (combining satisfy and suffice), refers to people’s tolerance — if not overall embracing — of “good enough” solutions…

Frankly, I discover satisficing in every research project: the unfiled MP3s sitting on the desktop, ill-fitting food container lids, and tangled, too-short cables connecting products are all “good enough” examples of satisficing. In other words, people find the pain of the problem to be less annoying than the effort to solve it.

I’m about a third of the way into Steve Portigal’s Interviewing Users but this bit rings especially true.

So much of the buzz around “smart cities” seems to focus on subtle optimisations and efficiencies – catching a bus a couple of minutes sooner, or turning the thermostat down a degree or two. Big data focused on small problems.

But wouldn’t the world be boring if everything was uniformly perfect? Maybe the capacity to work around life’s little frustrations is in itself a form of empowerment.

What if – for a while – we left alone all the stuff that’s good enough, and focused on delivering services that support people in making big decisions and enduring differences?