Key Takeaways
Design services deliberately, because accidents have real human and public costs
Failure is designed in. Lou Downe argues that services often become harmful through accumulated choices: a policy here, a staff target there, and a technology that nobody revisits. In the foreword’s tree story, a resident is sent between separate city websites to remove a stump and request a replacement tree, then gets rejected repeatedly for choosing unsuitable trees. The service ends with no tree at all. Elsewhere, student-loan callers are pressured to defer payments because operators have a seven-minute call limit, even when another option would better serve borrowers.
Good has three tests. A service should work for its user, be sustainable for its provider, and avoid harming society. Treating design as deliberate work makes failures visible, so teams can examine the rules and incentives creating them instead of blaming users for getting stuck.
The central move is from viewing service failure as isolated bad luck to examining the system that repeatedly produces it. That shift changes what an organization investigates: not only whether a page works, but whether policies, staff incentives, and handoffs combine to defeat the user’s goal. The three-part test also resists a narrow definition of success based on convenience or cost alone. Its challenge is practical: these interests can conflict, so teams need to make tradeoffs explicit rather than claiming that every improvement benefits everyone equally.
Call the service what people are trying to do
Start with the user’s words. A service may be internally known by a policy name, acronym, or technical label, while users search for an everyday task. A rural council learned this when a resident brought in a dead badger because he could not find instructions online. The UK government’s Fee Remissions service had a similar problem: people who needed help with court costs often did not understand the term remission. Research showed that they described the need as help with court fees, and the service was renamed accordingly.
Make the task searchable. Users often begin with a search engine, not an organization’s homepage. Name the thing people want to do, avoid unexplained acronyms and technical labels, and account for the different words users may try. Internal names can help staff coordinate, but they should not become a vocabulary test for the public.
Naming is not cosmetic. It shapes who discovers a service, what they think it offers, and whether they assume they qualify. In that sense, a title functions like a doorway: a precise internal label can still be a locked door if users do not know the organization’s language. Yet user vocabulary is not always uniform, and a single plain-language name cannot anticipate every search. Good naming therefore belongs with research into synonyms, search behavior, and the words people use when they do not yet know what help exists.
Explain the service before asking people to trust it
Set the terms early. Users need to understand what a service does, why it exists, how it works, and who it is for. They also need practical expectations, such as cost, timing, eligibility, and what happens after they act. When a city service let residents report potholes but did not say how long repairs might take, people called to ask why nothing had happened. The reporting tool had raised expectations without explaining the process behind it.
Purpose prevents wasted effort. Around 40 percent of users reporting medical conditions to the UK driving authority were reporting issues that did not need to be reported, including broken bones and ingrown toenails. The service had not clearly explained that it focused on conditions affecting safe driving. Explaining the reason for the questions helped route relevant cases to review and spared others unnecessary work.
Clear expectations give users agency: they can decide whether the service fits their situation and plan around its constraints. That is different from managing expectations merely to make a poor experience more tolerable. The distinction matters because an honest timeline does not excuse unreasonable delays, and an eligibility statement does not justify an unnecessarily narrow rule. Transparency is most useful when it helps people make a real choice and prompts the provider to examine whether the underlying process should change.
Own the user's outcome, even when other organizations provide the steps
Follow the whole journey. A user’s goal may cross several providers, even though each provider sees only its own task. Buying a house, for example, can involve estate agents, surveyors, lenders, and solicitors. A survey is not the user’s final goal; it is one step toward a safe purchase. Services should help people understand where they are in that larger journey and coordinate with the other parts where possible.
Fragmentation shifts costs. In a story about homelessness in Reno, the author recounts how one man’s emergency care, shelter stays, and time in prison cost more than a million dollars over ten years, while failing to provide a secure home. The lesson is not that one provider must deliver everything. It is that improving one small component may simply move costs and effort elsewhere unless the overall outcome remains in view.
This is a strong antidote to organizational tunnel vision. A provider can perform its assigned task efficiently while the full journey remains slow, expensive, or unsuccessful. Mapping the user’s outcome helps reveal those displaced costs and creates a basis for coordination. But end-to-end responsibility should not be confused with unlimited scope. A team may lack authority to control every partner, so it still needs a manageable role, clear handoffs, shared information where appropriate, and a way to expose gaps it cannot directly repair.
Design for newcomers, not insiders who already know the rules
Experience is not instruction. People arrive with assumptions learned from other services, but they may have no idea how this one works. On British trains, some toilet doors used unfamiliar electronic locks. Operators added signs and instructions, yet the design still failed because riders had no familiar cue for whether the door was secure. Conversely, a paper passport application had taught some users to rely on tracked mail as proof of receipt, so they distrusted a digital renewal that offered no equivalent reassurance.
Expectations come in three forms. Universal expectations are standard features people count on, such as basic banking access. Assumed expectations are guesses users make when a service has not explained its process. Outlier expectations are newer features some users have encountered elsewhere, such as instant banking alerts. Meet the basics, clarify important unknowns, and watch for new expectations spreading.
The useful insight is that first-time use is never a blank slate. Familiar conventions can make a service easier to learn, while a departure from them can create anxiety even when the new process is technically sound. But familiarity is not automatically virtuous: an old convention may be confusing, exclusionary, or needlessly burdensome. Change is justified when it improves the experience, provided the reason is understandable and the new pattern is tested with people beyond the design team.
Keep steps only where users need a meaningful choice
Let decisions set the pace. A service should have roughly as many steps as users need decisions, not as many as an old form or internal process happens to contain. Some breaks give people necessary visibility and control. Others are merely friction. The useful question is whether an action could happen automatically without the user losing a choice or understanding what comes next.
Design the pauses too. The number of steps is the service’s rhythm; the time between them is its tempo. A medical decision may need room for reflection, while an ordinary payment may be better completed quickly. A UK rule required a 14-day cooling-off period after switching some suppliers, but prevented providers from contacting customers during that wait. Users became confused, showing that a pause placed after a decision can create uncertainty instead of supporting deliberation.
Reducing steps is not the same as removing all friction. A confirmation before an irreversible choice may protect user control, while a repeated identity check may simply consume attention. The service designer’s task is to distinguish productive friction from accidental friction. This principle also warns against treating speed as the only measure of quality: an efficient path can still be poor if it rushes a consequential decision or leaves people unsure what they have agreed to.
Make every channel and handoff strong enough to preserve trust
The weakest link governs. A service can look polished online and still fail when a user switches to a phone call, email, or in-person interaction. Trainline discovered that its booking confirmation email was hard to locate in users’ inboxes just before they needed to collect tickets. Tesco added a gender-neutral title to its website, but a phone agent did not know how to address a customer using it, revealing that the change had not reached the whole service.
Build a complete service. Downe distinguishes consistency from uniformity. Users should receive a coherent experience across steps, channels, time, and user groups, but staff may need to respond differently to different circumstances. Developing a minimum viable service means making the essential journey work from beginning to end, rather than launching one impressive feature that leaves users stranded.
Consistency is best understood as dependable principles, not identical scripts. The distinction matters because rigid uniformity can make a service feel inhuman, while uncoordinated exceptions can feel arbitrary or unfair. The examples show that a local improvement can alter expectations across the organization, so changes need operational follow-through: training, updated systems, and clear ownership. A strong service is not simply the sum of well-designed interfaces; it is the ability of people and processes to uphold the same promise at each handoff.
Build alternate routes before users hit a dead end
Plan for real constraints. A dead end occurs when someone cannot continue, whether because they are ineligible, lack a required document, cannot use a particular channel, or cannot perform an assumed task. Sarah, who lost her phone in a ride-hailing car, could not use the app’s recovery route because logging in required access to that same phone. A separate phone-tracking tool helped her recover it, but the service itself offered no path forward.
Inclusion is practical design. Linda, a worker recovering from cancer, missed support because a benefits letter went to an old address and she could not afford travel to an assessment. Services should examine what users can do, who they are, and what resources they have. Offer alternatives for common barriers, explain routes for people who do not qualify, and avoid making one channel or document the only way through.
Designing around edge cases is not a niche accessibility exercise. A temporary injury, unreliable internet, limited money, caring duties, or a changed address can make an ordinary requirement impossible. Considering these situations exposes brittle assumptions that may affect many people. Still, alternatives need boundaries: security, legal obligations, and service capacity can limit what is feasible. The responsible response is to make those limits clear, provide a genuine route where possible, and avoid presenting a single imagined user as the norm.
Measure the outcome, not a proxy that rewards harmful behavior
Targets shape conduct. A government target required GP surgeries to offer appointments within 48 hours. Some surgeries stopped taking advance bookings, so patients had to call repeatedly within a narrow window. The target appeared successful for those who obtained appointments, while those who could not get through were largely invisible in the measure. The metric had replaced the actual goal: helping patients see a doctor when needed.
Align the whole system. Good service behavior should benefit users, staff, the provider, and society. Call-time limits, sales targets, or financial models can reward actions that undermine the service’s purpose. Design must therefore examine more than user behavior. A company cannot fairly ask customers to act responsibly while its own incentives encourage harm, concealment, or avoidable consumption.
A target is never merely a neutral report card; once attached to rewards or pressure, it becomes part of the service itself. The GP example shows how a measure can define success in a way that excludes people who never make it into the data. Better evaluation asks whether the intended outcome actually occurred and whose experience is missing. Even well-chosen measures need review, because staff and users adapt to them, sometimes in ways that were not anticipated.
Let users update what matters when their lives change
Change is normal. People move, change names, lose access to a phone, or experience changes in health and relationships. Services should be designed to respond rather than treating existing information as permanent. Ryanair’s staff scheduling system did not account for overlapping summer leave, contributing to the cancellation of flights and leaving nearly 80,000 passengers stranded. A predictable seasonal pattern had been ignored by the tool.
Use information in context. Not every personal change needs to be shared with every organization. Facebook’s sharing controls illustrate why a user may want different information visible to different groups. Services should explain why they need information and what will happen to it. When a detail becomes relevant, such as a phone number before confirming an appointment, checking it then may work better than relying on a user to report every change in advance.
Adaptability involves both service reliability and respect for personal control. The Ryanair example concerns predictable operational change; the privacy example concerns information that users may reasonably want to disclose selectively. Treating both as simple profile updates would miss that difference. Services need to distinguish what is required for a decision from what is merely convenient to collect. Asking for confirmation at the moment of use can reduce stale records, but only if the request is clear and does not create a new burden.
Explain decisions and keep humans reachable for exceptions
Make decisions contestable. A user should be told what information shaped a decision, why it mattered, and how to challenge the result. Automated sentencing tools in the United States produced racially unequal risk assessments, while people subject to those decisions often had little ability to inspect the reasoning. Opaque decisions make errors difficult for users and staff to identify, and can leave a person unable to correct a mistaken record.
Reserve a human route. Gina Haynes found inaccurate fraud-related records blocking job offers, but struggled to find anyone able to help. Laura Park, after her husband died, eventually reached a company bereavement team that could change account details outside the usual process. Human assistance matters most for complex, high-risk, or unusual situations. Staff need authority to explain, review, and resolve cases, not merely repeat a system’s answer.
Explanation and human support reinforce one another: a reasoned decision can be questioned, while a capable person can recognize circumstances that a standard process misses. This is not an argument against automation. Routine steps may be handled efficiently by technology, leaving human judgment for ambiguity and harm. The danger is offering a nominal contact route without empowering staff to act. Conversely, discretion without consistent guidance can create unfairness, so organizations need both authority and accountability.
Analysis
Good Services treats service design as a form of institutional responsibility. Its most important move is to define the service from the user’s goal rather than from the organization’s chart. That reframing makes familiar design failures visible as governance failures: a call-time rule, a disconnected database, a poorly explained eligibility decision, or a missing human contact route all shape what people can accomplish. The book’s examples are persuasive because the harm often emerges from ordinary operational choices, not a single villain or spectacular technical breakdown.
The principles also connect user experience to organizational architecture. Fragmented teams tend to produce fragmented journeys, while measures and incentives determine which parts of a journey receive attention. This makes service quality a useful lens for examining power: the provider controls the steps, evidence requirements, pace, and explanations, while the user often bears the cost of navigating them. Accessibility and clear communication therefore are not decorative additions. They affect whether a person can exercise meaningful choice at all.
The framework’s strength is also its limitation. Fifteen principles offer a practical baseline, but they do not resolve conflicts between speed, privacy, fraud prevention, fairness, and cost. Nor does a seamless journey always mean a desirable one. A highly integrated service can make consequential choices harder to notice, while friction can sometimes protect deliberation or consent. The book’s emphasis on outcomes and transparency points toward the needed discipline: make tradeoffs explicit, explain them, and preserve ways to correct mistakes.
Ultimately, the book asks organizations to stop treating services as invisible connective tissue. A service is an ongoing arrangement among people, rules, technology, and resources. Designing it well means maintaining that arrangement as circumstances change, rather than polishing one screen and assuming the job is done.
Review Summary
Good Services receives mostly positive reviews, praised for its insightful principles on service design. Readers appreciate its accessible language, real-world examples, and practical advice. Many find it a valuable resource for both beginners and professionals in the field. However, some criticize the book's typography and formatting, noting that the bold font and color choices can be difficult to read. Despite these design issues, most reviewers recommend it as an essential guide for understanding and improving service design.
People Also Read
Glossary
Service
Help toward a goalA service is something that helps someone achieve an outcome, whether simple or complex. It may involve multiple organizations, channels, and interactions, but users experience it as a connected journey toward their goal. The provider’s internal boundaries do not determine what counts as the service; the user’s intended outcome does.
Universal expectations
Expected basicsUniversal expectations are features or behaviors that users commonly take for granted because comparable services usually provide them. For example, a bank customer may expect to access money through an ATM. Providers do not generally need to advertise these basics, but failing to meet them can seriously damage confidence.
Assumed expectations
User guessesAssumed expectations are beliefs users form when they do not know how a service works. These guesses may be based on another provider, limited experience, or convenience, and may be wrong. Services should identify important assumptions through research, then either change the process to match reasonable expectations or explain the difference clearly.
Outlier expectations
Emerging user expectationsOutlier expectations are features or behaviors that only some users expect, often because they have encountered them in a newer or different service. An innovation such as instant transaction alerts may begin as an outlier expectation, then become common. Providers should monitor these expectations without neglecting established basics.
Experience integration
Joined-up service deliveryExperience integration describes providing a coherent user experience across multiple organizations or services. One provider may combine separate components into a journey that feels connected, even when the organizations delivering those components remain independent. This can improve coordination for users, but can also change the role of existing providers.
Minimum viable service
A complete usable journeyA minimum viable service is a basic but end-to-end version of a service that allows most intended users to complete their goal. It contrasts with launching one polished product feature while leaving the wider journey incomplete. Teams can improve the service incrementally, but should first ensure that essential steps and handoffs work.
Conway's law
Organization shapes systemsConway's law is the idea that systems tend to reflect the communication structures of the organizations that create them. When teams are isolated and do not coordinate, the services or software they produce are likely to be fragmented in similar ways. The concept highlights communication as a key condition for joined-up service design.
Service of last resort
Fallback human assistanceA service of last resort is a human-supported route that remains available when ordinary channels or technologies fail to help someone complete their goal. It recognizes that digital and automated systems cannot anticipate every circumstance. A capable person can use judgment to assist users whose situations do not fit standard pathways.
Dark patterns
Manipulative interface choicesDark patterns are interface or interaction choices that steer users toward actions they may not intend, or make it difficult to refuse or leave. Examples include obscuring an unsubscribe option or making account deletion unnecessarily difficult. The term highlights how design can manipulate behavior while appearing to offer free choice.
FAQ
What's "Good Services: Decoding the Mystery of What Makes a Good Service" about?
- Focus on service design: The book by Lou Downe explores what makes a service good and how to design services that work effectively for users.
- 15 principles: It introduces 15 principles of good service design, providing a framework for creating services that are easy to find, use, and understand.
- Real-world examples: The book uses real-world examples to illustrate how services can fail and succeed, offering practical insights into service design.
- User-centered approach: It emphasizes designing services that meet user needs, are sustainable for organizations, and beneficial for society.
Why should I read "Good Services" by Lou Downe?
- Practical guidance: The book offers actionable advice for anyone involved in designing or managing services, making it a valuable resource for professionals.
- Comprehensive framework: It provides a structured approach to service design, helping readers understand and apply the principles effectively.
- Improving user experience: By focusing on user needs and expectations, the book helps readers create services that are more user-friendly and efficient.
- Broader impact: It highlights the importance of designing services that are not only good for users but also for organizations and society as a whole.
What are the key takeaways of "Good Services"?
- 15 principles of good service design: These principles serve as a guide to creating services that are easy to find, understand, and use.
- User-centered design: Emphasizes the importance of designing services that meet the needs and expectations of users.
- Consistency and integration: Highlights the need for services to be consistent across different channels and integrated across organizational boundaries.
- Adaptability and responsiveness: Stresses the importance of services being able to respond quickly to changes in user circumstances.
What are the 15 principles of good service design according to Lou Downe?
- Be easy to find: Services should be easily discoverable by users without prior knowledge.
- Clearly explain its purpose: Users should understand what the service does and how it works.
- Set user expectations: Clearly communicate what users can expect from the service.
- Enable goal completion: Help users achieve their intended outcomes seamlessly.
- Work in a familiar way: Services should align with established user expectations and patterns.
- Require no prior knowledge: Users should be able to use the service without needing prior expertise.
- Be agnostic of organizational structures: Services should not expose users to internal organizational complexities.
- Require minimal steps: Streamline the service to minimize user interactions.
- Be consistent throughout: Ensure a uniform experience across all service channels.
- Have no dead ends: Provide clear outcomes and next steps for all users.
- Be usable by everyone equally: Ensure accessibility for all users, regardless of their circumstances.
- Encourage right behaviors: Promote safe and productive behaviors from users and staff.
- Quickly respond to change: Adapt to changes in user circumstances promptly.
- Clearly explain decisions: Make decision-making processes transparent to users.
- Make it easy to get human assistance: Provide accessible human support when needed.
How does Lou Downe define a service in "Good Services"?
- Simple definition: A service is something that helps someone to do something.
- User goal-oriented: Services are defined by the user's goal, whether it's short and straightforward or long and complex.
- Continuous journey: To a user, a service is a continuous set of actions towards an end goal, regardless of who provides it.
- Historical context: The book provides a brief history of services, showing how they have evolved with technology.
What is the significance of the principle "Be easy to find" in service design?
- User accessibility: Services must be discoverable by users without prior knowledge of the task they set out to do.
- Naming importance: The name of the service should reflect what users are trying to achieve, using language they understand.
- Avoiding jargon: Services should avoid legal or technical language that users may not comprehend.
- Google as a homepage: Recognizes that users often start their search for services online, emphasizing the importance of searchability.
How does "Good Services" address the issue of consistency in service design?
- Uniform experience: Services should look and feel like one service throughout, regardless of the delivery channel.
- Avoiding weak links: Inconsistencies can create weak links in the service chain, leading to user frustration.
- Minimum viable service: Focus on creating a consistent service experience rather than isolated product features.
- Empowering staff: Ensure staff are empowered to make decisions that maintain consistency across the service.
What does Lou Downe mean by "A good service should have no dead ends"?
- Clear outcomes: Every user should be directed to a clear outcome, even if they are not eligible for the service.
- Avoiding user frustration: Services should not leave users stranded without knowing how to continue.
- Multiple pathways: Design services with multiple pathways to accommodate different user scenarios.
- Continuous journey: Ensure that users can always find a way forward, regardless of their circumstances.
How does "Good Services" suggest services should respond to change?
- Adaptability: Services should quickly adapt to changes in a user's circumstances.
- Consistent updates: Ensure that changes are reflected consistently throughout the service.
- User control: Allow users to update their information easily and understand the impact of those changes.
- Proactive design: Anticipate common changes and design services to accommodate them seamlessly.
What role does human assistance play in "Good Services"?
- Complex decision-making: Human assistance is crucial for complex or high-risk situations where automated systems may fall short.
- Empowered staff: Staff should be empowered to make decisions and provide personalized support.
- Accessibility: Ensure that users can easily access human assistance when needed.
- Balancing automation: Find the right balance between automated and human support to meet user needs effectively.
What are some of the best quotes from "Good Services" and what do they mean?
- "Good services are designed": Emphasizes the importance of intentional design in creating effective services.
- "Google is the homepage for your service": Highlights the importance of online discoverability for modern services.
- "Make things open: it makes things better": Advocates for transparency and openness in service design to improve user experience.
- "A good service is good for everyone": Stresses the need for services to benefit users, staff, organizations, and society as a whole.
How does "Good Services" address the importance of user-centered design?
- User needs focus: The book emphasizes designing services that meet the needs and expectations of users.
- Understanding user goals: Services should be designed around what users are trying to achieve, not just organizational capabilities.
- Inclusive design: Ensure services are accessible and usable by everyone, regardless of their circumstances.
- Continuous improvement: Encourage ongoing evaluation and adaptation of services to better serve users over time.
Download PDF
Download EPUB
.epub digital book format is ideal for reading ebooks on phones, tablets, and e-readers.