Redesign an Internal Process End-to-End

Operations · 2–4 weeks · Intermediate

The job

Map an existing process, find waste and bottlenecks, and design a leaner version.

What the sponsor gets

Each canvas and map from this playbook — Service Blueprint, Empathy Map, Prioritisation Matrix, Experiment Canvas, Retrospective Canvas — exports as a PDF, or as PowerPoint on Pro, straight from the session, and the whole run exports as one PowerPoint for the sponsor. Nothing is redrawn.

The method

  1. Scope the process and gather data

    Define the start and end points of the process. Collect cycle times, error rates, handoff counts, and volume data. Interview 5–8 people who actually do the work, not just the people who designed it.

    Tip: The people doing the work always know where the waste is — they've just stopped mentioning it because nobody listened.

  2. Map the current process

    Create a Service Blueprint of the as-is process showing every step, handoff, decision point, and system interaction. Include wait times and rework loops.

    Tip: Colour-code steps by value-add (green), necessary non-value (yellow), and pure waste (red) — the visual makes the problem impossible to ignore.

  3. Understand the human experience

    Create Empathy Maps for the key people in the process — the requester, the processor, and the approver. Their frustrations reveal where the process serves the system instead of the people.

    Tip: Pay attention to workarounds — if people have built informal shortcuts, those reveal where the formal process fails.

  4. Identify and prioritise improvements

    List every improvement opportunity from the blueprint and empathy maps. Plot them on a Prioritisation Matrix. Focus on changes that reduce handoffs and wait times first.

    Tip: Eliminating one unnecessary approval step often saves more time than automating three others.

  5. Design the future-state process

    Create an Experiment Canvas for the redesigned process. Define what you're changing, how you'll measure improvement, and what the rollback plan is.

    Tip: Pilot with one team before rolling out — process changes that look good on paper often have unintended consequences.

  6. Pilot and measure

    Run the new process with a single team for 4 weeks. Measure the same metrics as your baseline. Gather qualitative feedback weekly.

    Tip: Track the workarounds that emerge during the pilot — they're telling you what you missed in the design.

  7. Retrospective and rollout decision

    Run a retrospective with the pilot team. Compare metrics against baseline. Decide whether to roll out, iterate, or roll back.

    Tip: If cycle time improved but people hate the new process, you haven't redesigned — you've just moved the pain.

The canvases and maps

Service Blueprint
Five lanes stacked from what the customer sees and does down to the backstage processes and support systems behind it, so the handoffs nobody sees become the thing everyone can point at.
Empathy Map
The goal first, then what the person sees, says, does, hears, thinks and feels, so the room separates what it has observed from what it is guessing.
Prioritisation Matrix
Impact against effort in four quadrants — quick wins, major projects, fill-ins and avoid — so the team trades off instead of calling everything a priority.
Experiment Canvas
Hypothesis, design, metrics and success criteria come before results and the decision, so the goalposts are written down before anyone sees the data.
Retrospective Canvas
What went well and what to improve, then start, stop and continue, so reflection ends in actions rather than a list of complaints.

A worked example

Fictional example: Northfold Supplies, operations.
The engagement
Order-to-dispatch handoff at Northfold Supplies.
The scope and the data
From "order confirmed by sales" to "left the warehouse". Eleven next-day orders missed the van last week, and every order is keyed in by hand twice. Six people who do the work were interviewed.
The current process
The service blueprint showed the delivery date confirmed before anyone checks stock, and the order sheet reaching dispatch at 5pm, after the stock check has already run.
The people in it
Empathy maps for the sales adviser, the warehouse picker and the dispatch lead. The picker phones sales about every missing line: the workaround showed exactly where the formal process fails.
What to change first
On the prioritisation matrix: check stock before promising a date, and send the order sheet at 3pm. Replacing the carrier system went into major projects.
The future state
On the experiment canvas: success means fewer than five late orders a week, and the rollback is the old 5pm sheet if dispatch cannot pick in time for the van.
The pack
Twelve contributors in the mapping session on Thu 12 Nov. That afternoon the whole run went to the operations director as one PowerPoint, with the pilot plan inside.
The pilot and the decision
Four weeks on one product line, then a retrospective with the pilot team: three late orders a week on average, and nobody wanted the old way back. Rolled out, with the team’s own "Fix a process that keeps breaking" session to keep it working.

What it doesn't do

It doesn't run the room for you or decide for the client. The steps, canvases and duration are a starting point you adapt; the judgement stays yours.