Good Services: a bloody good read (Part 2)

Good Services by Lou Downe tells you how to design good services that work.

I’m sharing my book notes so that you can enjoy the highlights for now, and hopefully be inspired to go on and buy it.

I’m not on commission. I just think that if you work in a job where you’re responsible for the delivery of services to other humans — you should read this book. It will help you to make your services better.

Good Services — How to design services that work. Lou Downe

The book contains 15 principles, which if followed, will help you to design services that work. Many of the examples in the book are from Lou’s experience of working in UK government organisations.

My notes are personal to me. I’ve noted the things that resonate, or that I want to remember and in places, added my own examples. I recommend you read the book to understand things in more depth and to read Lou’s stories of good and bad service design.

I’ve separated my notes into a three parts.

Part one: pre-service stuff (last Wednesday)

Part two: during service delivery (this one)

Part three: post-service stuff (coming next Wednesday)

15 Principles of Good Service Design

(Principles 1–3 are in part one)

4 — A good service enables the user to complete the goal they set out to do

You can’t change a user’s goal.

That goal might be ‘learn to drive’ or ‘move house’ and starts from the moment they consider doing something to when they’ve achieved their goal, including follow up support. Your user will define your service in relation to what they want to achieve. This is likely to be different from how you think of and describe your service.

If your service only provides part of your users goal, you need to help them orientate themselves, so they know what order to do things in. Tell them what they need to know before they reach you.

When we add up the costs of all fragmented services, they can become very costly: Malcolm Gladwell’s Million Dollar Murray example of how it cost the state $1 million over 10 years to keep Murray on the street because he fell through the cracks between services.

It’s all too easy to only consider those things that fall strictly within the scope of your existing service. But this makes the goal harder for the user to achieve and more expensive overall for organisations to provide.

When providing only part of a service, it is important that your element works with all the other elements of the end to end service. This requires awareness and collaboration with people outside your organisation. E.g., in the UK local authority housing registers need to collaborate with social housing providers to provide a seamless service.

What can we do differently?

• Remain focused on the end goal of your user.
• Ask yourself, ‘is this really where the service starts’ and ‘is this where they reach their goal’?
• Share data between orgs. to make a better service for the end user.

5 — A good service works in a way that’s familiar

People base their understanding of the world on previous experiences. If there’s an established custom for your service that benefits a user, your service should conform to that custom. But be mindful that not all customs benefit users — some have been put there to benefit the organisation. Avoid customs that negatively affect your user or those that are inefficient or outdated.

Trying to do something new without testing it with users risks creating something that doesn’t work. E.g., the ‘new’ locks on automated toilet doors on British Rail trains which embarrassingly exposed people half way through their visit to the loo because people couldn’t operate the locks.

What can we do differently?

• Look for patterns in how your competitors provide services
• Understand if there is an easier way to do them
• Test them with users
• Make changes that are intuitive
• Share the changes so that they become the new norm (E.g., Government Digital Service in the UK Design Principle: Make things open: it makes things better)

6 — A good service requires no prior knowledge to use

There is no service that will just be used by people who have used it before.

Services which require expert knowledge to understand and use attract parasitic industries to help the real end users navigate the system. Things like personal accountants, legal experts, tax specialists, conveyancers. Because these services have been so complex for so long, the service providers begin to think that these ‘experts’ are their users.

What can we do differently?

Design services so that they can be used by someone using it for the first time:

• How does your user know your service exists?
• Can they find it using searches based on the user’s logic, not the logic of the org. (this links with principle 1-
• Good services are easy to find)
• Does it work in a way that users expect, or have clear explanations on how to use?

7 — A good service is agnostic to organisational structures

The service must work in a way that does not expose the user unnecessarily to the internal structures of the organisation providing the service. Or in other words, we need to get better at collaborating and communicating across boundaries.

Services in the internet age don’t obey organisational boundaries. Getting something done is always more important than who is providing the service. When we define our services this way — by what a service user is trying to achieve — our services will likely stretch across multiple organisational boundaries.

Many of our organisations were designed for the industrial age. Services were devised and delivered consistently, with minimal changes over a long period of time. Delivered by organisations that too didn’t change. This led us to create organisations that are rigidly structured around a very specific set of tasks.

These tasks tend to be highly specialised at completing one specific part of the end to end journey. In a world where systems are expensive to change, the more specialised you are, the cheaper you become.

(* remember Malcolm Gladwell’s Million Dollar Murray from Principle 4)

Barriers to collaboration & what we can do about it

Separation of data

Review what data flows through your end to end service. Spot places where data is collected multiple times and not shared.

Design in data from the start, not as an afterthought.

Incompatible processes

Consider what activities your users have to do, and when do they have to complete these things.

Do all the timelines match up?

Incompatible criteria of use

Map out all of the mandatory requirements in your service and see if these are consistent across teams or departments.

Inconsistent language

Look at all the nouns in your service — what do you call documents, processes, activities and things a user needs to do.

What can we do internally with staff?

• Permission — allow people to ‘step outside of their day job’ to work together (e.g., org infrastructure and budgeting practices)
• Shared standards — some things we need to establish collectively to enable people to work together
• Shared goals — having a combined sense of what the future looks like is vital. Call it a vision / purpose or whatever. The important thing is that everyone is involved in creating it.
• Shared incentives — don’t let financial incentives ruin all your work in the other areas. The wrong KPI’s, profit-sharing or cost codes sharing rituals can become a stronger tide than communication.

8 — A good service requires as few steps as possible to complete

When thinking about how many steps your service needs, there is a simple rule: the number of steps in your service should be equal to the number of decisions your user has to make no more and no less.

8-A good service requires as few steps as possible to complete

There are two things to consider:

• Rhythm — # of steps = # of decisions a user has to make
• Tempo — speed of those steps should be relative to the size of the decision

There is a difference between involved services and transactional services:

• Involved services — where more time is required between steps to make big decisions (e.g., a cancer patient will require time to absorb shocking new information and decide on the appropriate treatment plan)
• Transactional services — need to be done quickly and as painlessly as possible (e.g., paying tax, buying groceries)

What can we do differently?

• Review where decisions are made in your service. Align steps with the users need to make a decision.
• Allow users to focus on one task at a time
• Think about how fast or slow your steps need to be

9 — A good service is consistent throughout

It should feel like one service (consistent language, visual styles and interaction patterns), regardless of the delivery channel.

A good service is only as good as its weakest link. It’s no good putting all your effort into one part of the service if it falls down in another part. E.g., Trainline’s booking service in the UK was well designed, but when users tried to retrieve their email, they couldn’t find it in their inbox. The service failed just before the user went to retrieve their ticket from the machine. This is highly stressful — I speak from experience!

Sometimes a service will be well designed in one channel, yet this new way of working hasn’t been communicated to other parts of the organisation. (E.g., Tesco digital team added Mx — a gender neutral honorific. Yet when a user received a call from the call centre, they didn’t know how to pronounce Mx.)

Being able to complete a service from start to finish is far more important than having a great experience in one moment, then not being able to complete the rest of the journey.

Instead of thinking about the minimum viable product, think about the minimum viable service. If you’re starting a new service, start with a usable service and incrementally improve parts of it. Don’t create a desert-island of a product and expect your users to swim to it.

Good services are consistent, not uniform:

• From end to end — ie, there are no gaps
• Across different communication channels — ie, consistent experience on digital and face to face channels and ability to swap between them.
• Over time — ie, neglecting a moment in the user journey will create a week link
• Between users — ie, consistency regardless of ability

What can we do differently?

No more computer says no!

Empower staff to make individual decisions about how to resolve challenges for users or to fix a problem in the best way they see fit.

10 — Good services should have no dead ends

A service should direct all users to a clear outcome, regardless of whether the user is eligible or suitable to use the service. No user should be left behind or stranded within a service without knowing how to continue.

There are four main reasons why a user might face a dead end:

• They’re not eligible to use your service. In which case, let them know as soon as possible, and give them an onward link to relevant support.
• They’ve strayed off the beaten track — into complexity that means they can’t get what they want to do done.
• They can’t do something (accessibility)
• They don’t have access to something. E.g., a landline phone number, utility bills to prove ID

What can we do differently?

• Provide onward routes for people who aren’t eligible
• Evenly distribute the complexity of your service
• Ensure it’s inclusive
• Minimise the number of requirements of users
• Build alternatives
• Let your service degrade gracefully — treat each new technology as a progressive enhancement on the last so that when services fail, you can rely on the technology that came before it.

11 — A good service is usable by everyone, equally

We need to go beyond thinking about accessibility, with all the inherent biases that come along with creating a baseline of ‘normal’ versus those with ‘access needs’, and start to think in terms of ‘inclusion’ of a full spectrum of needs instead.

11-A good service is usable by everyone, equally

If we relied only on people who look like us and have the same experiences as us, we wouldn’t have many users!

Nothing replaces the need to do research with your users to understand what specific barriers they might face when dealing with your service. Barriers might exist in relation to:

• Who they are — age, sex, ethnicity, race, religion
• What they can do — hearing, sight, reading ability, remember things
• Things they have access to — time, money, transportation, self belief, civic literacy.

Inclusion is a necessity, not an enhancement. Making your service inclusive doesn’t just make it usable for diverse users, but for everyone. This is because:

• The needs of users often converge — e.g., someone who is deaf, autistic or works full time may not be able to use the phone
• Testing your service with the most extreme ‘edge case’ needs makes it better for standard needs. Product designers have been doing this for years — e.g., crash test dummies test car safety standards in the most extreme of circumstances.

What can we do differently?

Instead of thinking about designing for the 80% of needs that are easily met, we should spend our time with the 20% of needs that are these so-called outliers. By understanding how those in the most difficult circumstances carve a pathway through our services we can make it easier for everyone.

Inclusion isn’t an afterthought. It’s much easier to design this in from the beginning.

Beware! Lack of inclusion in your team = lack of inclusion in your service.

12 — A good service encourages the right behaviours from users and staff

Staff shouldn’t be incentivised to provide a bad service to users, for example through short call-handling time targets. What gets measured gets done. If you give people a target, they will inevitably try to meet that target, regardless of whether it helps to achieve the thing they’re setting out to do.

The target was met but the point was lost

In 2005, Tony Blair appeared on BBC Question Time in the general election campaign. An audience member said that she couldn’t get a GP appointment. She’d been trying for 3 weeks! She was only allowed to make an appointment within the next 48 hours. Why? The GP surgery was trying to meet government targets to give all patients an appointment within 48 hours of calling the doctor.

Blair responded with “well, obviously it wasn’t supposed to work like that”.

The GP surgeries responded by not allowing people to make appointments beyond 48 hours. Those requesting an appointment beyond this time limit weren’t recorded at all, meaning that the target was met but the point was lost.

Although targets might be set with the best of intentions with underlying objectives that are well understood by those at the top of the hierarchy, they are rarely communicated to staff who deliver the service.

Instead, upper management generally tries to ensure that objectives are met by setting a target, forgetting entirely that those delivering the service might be the best at knowing how to measure whether they’re being effective, so long as they understand the objective they’re trying to meet.

If these targets stay in place too long, we can soon forget what the objective was in the first place.

The right behaviours are those which:

• benefit the user,
• benefit your staff,
• helps the organisation to be financially sustainable and
• benefits the world (or at least does no harm).

Some organisations have started to use targets focused on what a user is trying to achieve by using the ‘jobs to be done’ methodology of Anthony W Ulwick — which tasks orgs. to organise their metrics based on whether or not they are achieving an outcome for a user.

What can we do differently?

• Remember what you’re trying to achieve: Incentivise staff in a way that does not run counter to delivering benefits for your user.
• Encode the outcome you want to achieve in every layer of the organisation.

In the final installment, we’ll look at principles 13–15 and consider the things you need to get right after service delivery. If you can’t wait until next week, why not buy the book?

EDIT: Link to final part 3.

Follow Jo or Service Works on Twitter or LinkedIn.

Take a look at our website www.weareserviceworks.com and sign up to our newsletter or get in touch jo@weareserviceworks.com

We help government and third sector employees to design services that work through training, consultancy and workshops.

Our flagship service is ‘Service Design in Practice’. An eight month learning and development programme, which takes people with little or no experience of using service design to applying and practicing these tools and approaches on real work based challenges.