What needs to be finished first
An existing customer journey for the service to blueprint, with clearly bounded phases and touchpoints.
List of involved roles, systems and external providers with one owner per role.
What needs to be ready before start
Large whiteboard or Miro board with five lanes (Customer Actions, Frontstage, Backstage, Support Processes, Physical Evidence); stickies in five colors; ruler cards for lines (Interaction, Visibility, Internal Interaction); access to process diagrams and system documentation.
One facilitator with service-design experience; one representative per frontstage role (sales, service, support); one owner per backstage system; one notetaker; one UX lead for the visibility line.
Customer journey; list of all touchpoints (digital, phone, physical); system landscape with ownership; known incidents or complaints from the last 90 days; SLAs of external providers.
4-6 h for one service slice, then 1 h maintenance per quarter
Horizontal lanes: Physical Evidence at the top, then Customer Actions, Line of Interaction, Frontstage, Line of Visibility, Backstage, Line of Internal Interaction, Support Processes. Vertically split into journey phases.
The one question this method answers
At which invisible backstage steps do fail points emerge that the customer experiences as friction at the frontstage?
Marker: Phase
| Step | Duration | Action | Hint |
|---|---|---|---|
1Phase 1: Customer Actions | 45 min | Enter concrete customer actions per journey phase, verb-oriented and with touchpoint ("opens app", "calls hotline"). Take them from the journey, do not reinvent them. | If this lane is created from scratch instead of coming from an existing journey, a precondition is missing. Pause and clarify the journey, otherwise the blueprint becomes speculation. |
2Phase 2: Frontstage | 45 min | Add the visible employee or system actions below each Customer Action. Include role and tool. Put Physical Evidence (emails, receipts, app screen) above the actions. | Frontstage is only what the customer directly experiences. Internal validations already belong behind the visibility line. |
3Phase 3: Backstage and support | 60 min | Add backstage actions and support processes per frontstage step. Name owner and system. Connect with arrows, mark asynchronous handoffs. | A backstage lane without owner is a blind spot. Better to write "unknown, clarify by ..." explicitly than assume an owner. |
4Phase 4: Fail points and wait times | 45 min | Mark fail points with red stickies (handoffs, external calls, manual steps). Quantify wait times and SLA breaches where possible. | Black-box fail points without frequency or impact are subjective. Require at least one observation or metric per fail point. |
5Phase 5: Action slice | 30-45 min | Prioritize three to five fail points by frequency x pain. Define one concrete improvement per point with owner and success metric. | If the improvement is "more people", it is not one. Look for structural levers: automation, remove handoff, renegotiate SLA. |
What comes out at the end
Service Blueprint as photo or Miro board with all lanes and lines plus accompanying report with identified fail points, frequency, owner and prioritized action slice.
One blueprint per service variant with version date. Service changes (new systems, new touchpoints) create a new version, old one gets archive tag. Action slice as linked roadmap item.
Compact working template for Service Blueprinting with context, input, output artifacts, and next step.
What is this method used for?
Which question should be answered at the end?
Which data, observations, or materials are available?
Owner, date, success signal.
Concrete filled scenario, fictional example
Concrete filled scenario, fictional example
Phase 1 (claim submission)
Phase 2 (assessment)
Action slice (by 2026-09-30):
Recognize symptoms and steer against them
Arrows point to systems or teams, nobody can speak for them with authority.
Complete owner list before starting. If a system has no owner, the blueprint is incomplete. Mark the gap explicitly rather than guessing.
Internal checks are entered as frontstage because an employee sees them.
Draw the line strictly by customer experience. What the customer does not see is backstage, even if an employee works on it.
Red stickies are placed from gut feeling, frequency unknown.
Require at least one observation or metric per fail point. If no data is available, run a data-collection spike before action.
Blueprint covers acquisition, delivery, service and cancellation in one map, everything stays superficial.
One service slice per session (for example claim submission only). Other phases get their own blueprints. Depth before breadth.
Every friction point is solved by increasing headcount, no structural change.
Force one structural lever per fail point: remove handoff, automate, new order, SLA. More people only as last lever with ROI rationale.
After 6 months systems no longer match, new providers are missing, actions cannot be verified.
Schedule quarterly maintenance. For service changes, make blueprint update part of the Definition of Done.
Done signals checkable in under a minute
Go to the profile for purpose, similar methods, and sources or continue to the next method in the catalog.