Scoping and Success Criteria

Topics Covered

From Frame to Scope

What scope means

Starting from the frame

Thin slices

Choosing the first slice

In scope and out of scope lists

Assumptions and dependencies

Estimating effort

Scoping with the customer

Scope and the contract

Scope for the vendor team

Signs of poor scope

Scope and team size

Scope for pilots

Revisiting scope after discovery surprises

Scope for data and integration work

Scope for users and training

Scope in different deployment models

Testing the scope with a dry run

Writing scope in the language of the customer

Scope that survives the first review

A good frame says what problem to solve. Scope says how much of it the deployment will solve, for whom and by when. Without clear scope, deployments drift, deadlines slip and customers feel let down even when the team works hard. This lesson explains how to turn a frame into scope, write success criteria, plan milestones, manage change and agree the plan with the customer.

What scope means

Scope defines what the deployment will deliver: which users, which workflows, which data sources, which outputs and which quality levels. It also defines what the deployment will not deliver, at least in this phase. Both halves matter. A clear list of exclusions prevents misunderstandings later, when someone assumes a feature was included all along.

Starting from the frame

Scope starts from the frame agreed in discovery. If the frame says planners need a validated plan by seven each morning, scope lists which planners, which regions, which data feeds and which checks are included. Every item in scope should trace back to the frame. Items that do not serve the frame belong in a later phase or nowhere.

Thin slices

A useful approach is to scope a thin slice: one user group, one workflow, end to end, with real data. A thin slice delivers working value quickly and tests every part of the system: data access, processing, output and adoption. It is better to deliver one workflow fully than five workflows partly, because partial workflows rarely get used.

Choosing the first slice

Choose a first slice that is valuable, visible and achievable. Valuable means users care about it. Visible means the sponsor and other stakeholders will notice the result. Achievable means the data and systems are accessible within weeks. A slice that scores well on all three builds momentum and trust for the larger work that follows.

In scope and out of scope lists

Write explicit lists. In scope: daily plan for the north and south regions, using order and inventory data, delivered in the planning tool. Out of scope for phase one: east region, carrier data, mobile access, automated rerouting. Short, concrete lists like these are easy for everyone to read and to point at when questions arise later.

Assumptions and dependencies

Scope rests on assumptions and dependencies: access to a database by a certain date, a customer engineer available two days a week, a security review finishing on time. List them next to the scope. If a dependency slips, the scope or timeline may need to change, and having the dependency written down makes that conversation straightforward rather than awkward.

Estimating effort

Estimate the effort for each item in scope, ideally in rough ranges such as two to three weeks. FDE estimates carry more uncertainty than product estimates, because customer environments are unpredictable. Add explicit allowance for integration surprises, approvals and data problems. Experienced FDEs often double their first estimate for anything that touches a system they have not seen.

Scoping with the customer

Scope should be built with the customer, not presented to them. Walk through the frame together, propose a first slice and discuss what is in and out. Customers often have strong views about priorities and constraints that change the plan. Scope that the customer helped write is scope they will defend when pressure arrives.

Scope and the contract

Check scope against the contract. Some contracts list specific deliverables, while others describe outcomes loosely. If the agreed scope differs from what the contract implies, raise it early with the account team. Scope disagreements discovered near renewal are painful, while those discovered at kickoff are easy to resolve.

Scope for the vendor team

Scope also matters inside the vendor. Your manager, the account team and the product team need to know what you are committing to. Share the scope internally, especially if it depends on product features or other teams. A scope that surprises your own company is as risky as one that surprises the customer.

Scope turns a frame into a thin end to end slice with explicit lists of what is in, what is out and what it depends on.

Signs of poor scope

Poor scope shows itself early: vague items such as improve reporting, no out of scope list, no dependencies listed, estimates without ranges and no link to the frame. If you inherit a deployment with scope like this, rewrite it before building more. A day spent clarifying scope can save weeks of confusion and rework.

Scope and team size

Scope must fit the team. A single FDE can deliver a thin slice in a few weeks but not a multi region platform in the same time. If the customer wants more scope, discuss adding people, extending the timeline or phasing the work. Pretending that one person can deliver everything quickly sets up the deployment to fail.

Scope for pilots

Pilots need especially tight scope. A pilot tests whether the product works for the customer before a larger commitment. Scope it around the question the pilot must answer, with a short timeline and clear success criteria. Pilots that grow into open ended projects lose their purpose and often end without a clear decision.

Revisiting scope after discovery surprises

Sometimes discovery continues to reveal surprises after scope is set, such as a data source that cannot be accessed. Revisit scope promptly when this happens, rather than hoping to absorb the problem quietly. Explain the change to the sponsor, propose an adjusted scope and record the decision. Early adjustments are far less painful than late ones.

Scope for data and integration work

Data and integration items deserve extra care in scope, because they are where surprises hide. Name each source system, the fields needed, the expected volume and the refresh frequency. A line such as load order data daily from the warehouse system, around two million rows, by six each morning is far easier to plan and test than a line that just says integrate orders.

Scope for users and training

Scope is not only technical. Include the people side: how many users will be trained, who will run the training, what support they get in the first weeks and which teams will be onboarded later. Deployments that scope only the software often deliver a working system that few people know how to use, which looks like failure to the sponsor.

Scope in different deployment models

Scope looks different depending on how the vendor works. In a product led deployment, scope is mostly configuration and integration of an existing platform. In a services heavy deployment, scope may include custom code the vendor will maintain. Be explicit about which parts are product, which are custom and who owns each part after launch, since this affects support and cost for years.

Testing the scope with a dry run

Before committing, walk through the scope as a dry run. Imagine the first day of use: a planner opens the tool at seven, sees the plan, checks two exceptions and publishes. Does the scope cover every step of that day? Dry runs often reveal missing pieces, such as a login method, a permission or a report the planner needs to send to their manager.

Writing scope in the language of the customer

Write scope in words the customer uses, not in vendor jargon. If planners talk about the morning plan and the exceptions list, use those terms. Scope written in customer language is easier to review, easier to agree and easier to check later. It also signals that the FDE understands the work, which builds confidence in the plan from the very first meeting.

Scope that survives the first review

The first review of scope usually adds detail rather than removing items. Expect questions about edge cases, permissions and reports. Answer them in the scope document itself, so the next reader sees the same answers.

Thin, explicit and traced

Turn the frame into a thin end to end slice that is valuable, visible and achievable. Write explicit in scope and out of scope lists, list assumptions and dependencies, estimate in ranges with allowance for surprises and build the scope with the customer so they will defend it.