A handover is not complete because the internal meeting happened. It is complete when the commercial promise, delivery boundary, customer responsibilities and evidence of first value are readable without the people who sold the work having to explain what they meant.
The gap often appears after everyone has behaved reasonably. Sales protects momentum. Implementation protects feasibility. Customer Success protects continuity. The customer remembers the outcome discussed in the buying conversation. Each person carries a sensible part of the story, but nobody owns the whole decision.
An illustrative case
The following is a composite teaching scenario, not a claimed customer engagement or outcome.
Harbourline has bought a twelve-week planning pilot for two sites. The signed scope includes a standard connection to its existing system. During the sales process, supplier automation was discussed as a future possibility. A later customer email describes it as part of the pilot. The September planning cycle creates a fixed date, and the customer has not yet named the person responsible for historic data.
The internal handover is marked green. The order is signed. A project manager is assigned. Kick-off is in the diary.
Yet four different definitions of success are already present:
- Sales: preserve the September start and the relationship.
- Implementation: connect the agreed system and load usable data.
- Customer Success: reduce weekly planning rework.
- Customer: automate supplier activity before the next cycle.
If the team treats these as one shared understanding, the disagreement will not disappear. It will emerge later, when correction is more expensive and feels more personal.
Reconstruct the promise before planning the work
Start with a single sentence that names the customer, scope, intended change and time boundary:
Harbourline bought a two-site proof of planning value before the September range cycle; supplier automation and national rollout are not part of the current commercial commitment.
This sentence does not settle every issue. It makes the central boundary challengeable.
Then separate the record into four columns:
| Status | Meaning | Example from the scenario | |---|---|---| | Contracted | Present in the signed scope | Standard system connection for two sites | | Customer-stated | Expressed by the customer but not contracted | Supplier automation in the pilot | | Assumed | Treated as true but not yet checked | Historic codes are consistent | | Unknown | Evidence or ownership is missing | Completeness of historic data |
This is not contract interpretation or legal advice. It is a practical way to stop different kinds of claim being flattened into one project plan. Material disagreements still need the appropriate commercial, delivery and legal review.
Separate launch from first value
“Go live” is a provider milestone. “First value” is an observable change in the customer's work.
In the scenario, a better first-value statement might be:
By week twelve, pilot planners reduce weekly forecast rework from the agreed baseline while forecast accuracy does not fall below the customer-approved safeguard.
The exact measures would need real customer confirmation. The structure matters: a user group, an operational change, a time boundary and a safeguard against improving one number by damaging another.
The GOV.UK Service Standard makes a related point for public services: define what success looks like, identify measures that show whether the service solves the intended problem, and use performance evidence to improve it. That guidance applies to government services, not automatically to a commercial implementation. The transferable principle is that delivery activity and user outcome need different evidence.
Give every gap an owner and date
Avoid assigning “the account” to one person. Assign the missing decisions:
- Sales owns correcting or confirming the supplier-automation expectation.
- Implementation owns reporting data completeness and connection feasibility.
- Customer Success owns continuity of the first-value measure and baseline.
- The customer programme lead owns naming the data owner and confirming the boundary.
Now add dates and evidence. “Customer to confirm” is not an action. “Customer programme lead names the data owner by day five in the shared mobilisation record” is.
The route forward can preserve momentum without hiding the conflict:
- Begin reversible discovery and connection preparation.
- Keep supplier automation outside phase one unless scope is formally changed.
- Do not load production data until ownership and readiness checks are complete.
- Confirm the baseline and safeguard with the planners who do the work.
- Review the decision at named day-seven, day-fourteen and day-thirty checkpoints.
This is conditional mobilisation. It is neither “press on and hope” nor “stop everything until uncertainty disappears”.
The next useful action
Take one live handover. Ask each function, separately, to complete this sentence: “The customer bought ___ so that ___ changes by ___.” Compare the answers. Where they differ, classify the statement as contracted, customer-stated, assumed or unknown. Give the highest-consequence gap a named owner, evidence requirement and date before the next irreversible action.
The Promise to First Value Package is the related catalogue tool for reconstructing that decision, clarifying functional boundaries and carrying a customer promise into accountable mobilisation.
Sources and transfer limits
- GOV.UK Service Standard: Define what success looks like and publish performance data — used for the distinction between activity, success measures and evidence of whether a service solves its intended problem. It is guidance for UK government services; applying the principle to a private commercial handover is editorial judgement, not an official endorsement.
- GOV.UK Service Manual: Measuring the success of your service — used for the value of combining performance measures with user research and considering the end-to-end journey. It does not prescribe the example measures or contractual roles in this article.
Sources reviewed 7 September 2026. The Harbourline example is explicitly illustrative and must not be represented as customer evidence.