Service Designers —Running

Act as a service designer working with ops. For [insert live service], design a service health dashboard covering 4 dimensions: User outcome, User experience, Operational health, Business outcome. For each dimension: 2–3 metrics, definition, source, target, healthy/concerning thresholds, review cadence. Present as a dashboard spec.

Act as a service designer. For [insert service], distinguish output metrics (volume, throughput) from outcome metrics (did users get what they came for). Propose 5 outcome metrics that capture whether the service is actually working for users. For each: definition, how to measure, target, why it's an outcome not an output. Present as a table: Metric, Definition, How to Measure, Target, Why It's an Outcome.

Act as a service designer. [Insert service] looks healthy on top-line metrics but users complain. Help me identify the hidden friction metrics — completion time, retry rate, abandonment, calls per case, repeat contacts, escalation rate. For each: how to measure, what it tells us, where the friction probably sits. Present as a table: Metric, How to Measure, What It Tells Us, Likely Friction Point.

Act as a service designer. For [insert service], define service level targets across: timeliness, accuracy, accessibility, recovery time, communication. For each: the user-facing promise (in plain English), the operational target, how it's measured, what triggers intervention. Match the 15 principles. Present as a standards table: Area, Promise, Target, Measure, Trigger.

Act as a service designer. Audit this list of metrics being tracked for [insert service]: [insert]. For each: is it measuring something real? Is it actionable? Does it predict user satisfaction? Recommend which to keep, which to retire, which to replace. Present as a metrics audit.

Act as a service designer. For [insert service], design a continuous user feedback loop. Cover: where to capture (post-interaction, mid-journey, on-exit), what to ask (max 3 questions), how to route insights to the team that can act, the cadence for review, the rule for when feedback triggers a change. Present as a one-page feedback-loop design.

Act as a service designer. Design a system for frontline staff of [insert service] to feed insights upward without it becoming a complaint stream. Cover: the format (structured note, ticket, weekly synthesis), who reads it, how it gets prioritised, how staff see action taken from their input. Reference Lou Downe's principle that good services respond to change.

Act as a service designer. Synthesise this VoC data: [insert mix of survey scores, verbatims, support tickets, social mentions]. Identify: the top 5 themes, where pain is concentrated, what's getting better, what's getting worse, the 3 highest-leverage interventions. Present as a VoC synthesis.

Act as a service designer. Triangulate these inputs for [insert service]: [insert NPS], [insert support ticket themes], [insert frontline observations], [insert ops metrics]. Identify: where they agree (high-confidence issues), where they conflict (needs investigation), what's only visible from one source. End with a prioritised issue list.

Act as a service designer. For this service failure: [insert incident]. Run a blameless post-mortem covering: what happened (timeline), what the user experienced, what failed (frontstage, backstage, support, evidence), why it failed (root cause), how we recovered, what we learned, what changes prevent recurrence. Present as a post-mortem doc with a timeline table.

Act as a service designer. For [insert service], design recovery protocols for the top 5 failure modes. For each: how the user finds out, what we say, what we do, who owns it, the timeline, the goodwill gesture (if any), how we close the loop. Reference Lou Downe's principle that good services make it easy to recover from mistakes. Present as a table: Failure Mode, User Comms, Action, Owner, Timeline, Gesture, Close-out.

Act as a service designer drawing on John Seddon's failure demand concept. For [insert service], analyse the inbound contact volume. Categorise: value demand (user actually wants the service), failure demand (contact caused by something we did or didn't do). For failure demand, identify the root causes and the redesigns needed to remove them. Present as a table: Contact Type, Demand Category, Root Cause, Redesign.

Act as a service designer. Audit [insert service] for dead ends — places where a user gets stuck with no path forward, no clear help, no recovery. Cover: digital dead ends (404s, broken flows, expired sessions), procedural dead ends (no fits-our-rules path), human dead ends (bounced between teams). For each: where it occurs, the fix. Present as a table: Dead End, Type, Where It Occurs, Fix.

Act as a service designer. For [insert service], identify the workarounds frontline staff have built (spreadsheets, side-channels, manual lists, unofficial scripts). For each: what it does, why it exists (what the official service fails to do), the risk it carries, whether to formalise or fix the underlying gap. Present as a table: Workaround, What It Does, Why It Exists, Risk, Formalise or Fix.

Act as a service designer. Audit the capability of frontline staff to deliver [insert service] well. Cover: knowledge gaps, tool gaps, authority gaps (where they need to ask permission for things they should just decide), time gaps. For each: the impact on user experience, the fix. Present as a table: Gap Type, Gap, User Impact, Fix.

Act as a service designer. For [insert service], structure a staff knowledge base that helps frontline staff serve users. Cover: top 20 user scenarios with scripts, decision trees, escalation paths, edge cases. Make it scannable in 30 seconds during a live interaction. Avoid corporate language. Present as a knowledge-base outline with a sample entry.

Act as a service designer. For [insert service], identify the backstage activities that consume time without serving users (duplicate data entry, hand-offs between systems, sign-offs that never get rejected, reports nobody reads). Estimate the time cost. Recommend what to remove, automate, or simplify. Present as a table: Activity, Time Cost, Remove / Automate / Simplify.

Act as a service designer. Run a quarterly review of [insert service]. Cover: performance vs targets, user feedback themes, frontline feedback themes, incidents and recoveries, what changed last quarter and what it moved, what's planned next quarter, what needs leadership input. Present as a 1-page review.

Act as a service designer. From this improvement backlog: [insert]. Score each on: User impact, Frontline impact, Operational saving, Effort, Risk if delayed. Recommend the top 3 to deliver this quarter. Present as a prioritised list.

Act as a service designer. Design an A/B test for [insert service touchpoint] that respects service-level constraints (regulation, fairness, no-disadvantaged-cohorts). Cover: hypothesis, control, variant, success metric, guardrails, sample size, exclusions, ethical review. Present as a service-safe test plan.

Act as a service designer. Take this insight: [insert]. Convert it into an improvement brief: what's wrong, why it matters, what good looks like, the proposed change, the success measure, the team that needs to deliver it, the risks, the dependencies. Present as a 1-pager.

Act as a service designer. Set up governance for [insert service] crossing multiple teams. Cover: the service owner role, the regular forum (what, who, cadence), how decisions get made, how disagreements are escalated, who has authority over the user experience end-to-end. Avoid making it a meeting circus. Present as a one-page governance charter.

Act as a service designer mediating tension between teams owning different stages of [insert service]. The tension: [insert]. Help me: 1) name the underlying conflict (interest vs position), 2) reframe around the user outcome both teams should be serving, 3) propose a path forward, 4) define what we'd track to know it's working.

Act as a service designer briefing leadership. For [insert service], write a 1-page health update: top-line health (are users getting outcomes?), 3 things working, 3 things broken, 1 systemic risk on the horizon, 2 asks. Avoid jargon. Be specific about what's actually happening.

Act as a service designer. Make the business case for investing in [insert service improvement]. Cover: the user pain, the operational cost of failure demand, the business outcome at stake, the proposed investment, the projected return, the risk of doing nothing. Present as a 1-page investment case.

Act as a service designer. For [insert service] launching in [insert new market], identify what needs to change. Cover: regulation, language, channel preferences, payment norms, service expectations, frontline staff context, supplier ecosystem. For each: the local difference, the adaptation, the test. Present as a table: Area, Local Difference, Adaptation, Test.

Act as a service designer. Plan a phased rollout of [insert service change] across [insert N] sites or markets. Cover: pilot site selection, success criteria for each phase, the rollback plan, the comms plan, the staff training, the monitoring. Present as a phased rollout plan with kill criteria.

Act as a service designer. For [insert service] delivered across many sites, design for healthy variability — the bits that should be standardised vs the bits that should flex locally. For each component: standardise (why), flex (within what guardrails), measure (how we know it's working). Present as a standardisation matrix.

Act as a service designer. For [insert service], plan how to handle demand spikes (seasonal, viral, event-driven). Cover: what triggers the spike protocol, what scales, what gets temporarily reduced, how we communicate to users, how we protect frontline staff from being overwhelmed, how we return to normal. Present as a spike playbook: Trigger, Scale Up, Scale Down, Comms, Staff Protection, Return to Normal.

Act as an inclusive service designer. Audit [insert service] against 5 edge user groups it likely fails: users in financial difficulty, users in crisis, low-literacy users, users with assistive tech, users with limited access (no smartphone, no fixed address). For each: where the service fails them, the fix, what to test. Present as a table: User Group, Where It Fails, Fix, Test.

Act as a service designer. For [insert service], identify environmental costs: physical artefacts produced, travel required, digital infrastructure, paper waste, hardware lifecycle. For each: the current cost, the reduction lever, the trade-off. Present as a sustainability audit.

Act as a service designer. For [insert service], design for the relationship over time, not just the transaction. Cover: how the service learns from the user as they return, what becomes easier the second/fifth/tenth time, how trust accumulates, how we avoid re-asking the same questions. Reference Lou Downe's principle of consistency. Present as a relationship roadmap: First Time, Fifth Time, Tenth Time.

Act as an ethical service designer. Audit [insert service] for coercive or dark patterns: hidden costs, friction on cancellation, default opt-ins, manipulative language, missing choices, unclear consequences. For each: where it occurs, the user harm, the fix. Present as an ethics audit.

Act as a service designer. Stress-test [insert service] against 3 plausible futures: 1) demand doubles, 2) regulation changes (specify how), 3) a key technology becomes obsolete or central. For each: what breaks, what holds, what we'd need to redesign, what to start preparing now. Present as a table: Scenario, Breaks, Holds, Redesign, Prepare Now.

Act as a service designer. Audit [insert service] for strategic drift — where the service has slowly diverged from its original outcome through accumulated changes. Identify: where it's drifted, why each change made sense individually, what the current service is now optimising for vs what it should optimise for, the realignment needed. Present as a drift audit: Original Outcome, Current Behaviour, Drift, Realignment.

Act as a service designer. Plan a refresh of [insert mature service]. Cover: the user signals saying it's needed, what to keep (the things that work), what to change, what to remove, how to do it without disrupting users mid-service, the comms plan, the timeline. Present as a one-page refresh plan with a timeline.

Act as a service designer leaving [insert service]. Write a handover doc for whoever takes over. Cover: the service's purpose and outcome, the current health, the things you've tried, the things you haven't, the unresolved tensions, the people who matter, the metrics that lie, the metrics that matter. Be honest.