Scoping is the most under-invested part of most agency engagements. This is the playbook we use — adapted from years of getting it wrong before getting it right.

The principle

A good scope answers three questions clearly: what's included, what's not, and what happens when reality disagrees with the plan. Everything else in the proposal is supporting evidence.

Phase 1: Pre-engagement triage

Before you spend any real time, qualify the engagement. Reject projects where:

  • The client can't articulate the business outcome they're paying for.
  • The budget is unknown or "we'll figure it out as we go."
  • The decision-maker isn't part of the conversation.
  • The timeline is set by an external constraint that doesn't allow for proper scoping.

Saying no to bad engagements is the highest-leverage thing you'll do in scoping.

Phase 2: Discovery week

For engagements above a threshold size (we use £25K+), we run a one-week discovery as a paid engagement. The deliverable is the scope document. Five days, structured as:

  • Day 1: Stakeholder workshop and one-on-one interviews.
  • Day 2: Technical and design audit of existing systems.
  • Day 3: Architecture and design exploration. Sketches, not commitments.
  • Day 4: Draft scope, milestones, budget, and exclusions.
  • Day 5: Working review with client, revisions, delivery.

Phase 3: Writing the scope document

The document itself follows a strict structure:

  1. Executive summary. One page. What we're building, why, key decisions.
  2. Detailed scope. Page-by-page or feature-by-feature breakdown of inclusions.
  3. Exclusions. Equal prominence to inclusions. Anti-scope.
  4. Milestones and timeline. What's delivered when.
  5. Budget and payment schedule. Total, broken down by phase.
  6. Assumptions and dependencies. What needs to be true for this scope to hold.
  7. Change order process. What happens when scope changes mid-project.
  8. Post-launch. Support, retainer, or handoff.

Phase 4: The estimation move

For each line item, we estimate three numbers: best case, expected, worst case. We bid the expected. We share the range internally. When reality bends toward worst-case, the team knows to flag it early.

Phase 5: Change orders

Scope creep kills projects. We pre-build a change order process into the scope document itself: any in-flight scope change above a threshold (e.g., 4 hours of work) gets a written change order — even if it's a one-paragraph email. Both sides sign off. Then we build it.

What we'd do differently

Earlier in our career, we under-priced discovery because clients pushed back on paying for "just a document." That was a mistake. Discovery is the most valuable week of any engagement. Price it accordingly, and the rest of the project goes smoother.