Sign in

Blog · Territory, capacity and quota planning

Agree success criteria for a customer supplier trial

Define a bounded supplier trial with agreed scope, operational measures, exceptions and a decision date. Separate trial orders from approval to expand the relationship.

The short answerA customer supplier trial needs an agreed product and site scope, a fixed observation window, observable delivery and quality criteria, exception handling and named reviewers. Decide how results will support a go, revise or stop decision before the first order. Passing a small trial supports only its tested scope; it does not prove full rollout readiness, annual savings or an analytical platform's capabilities.

A trial order is an opportunity to learn whether a proposed supplier arrangement works for the customer. Without agreed success criteria, the seller may call it successful because an order shipped while the customer sees a late delivery or an unusable product. Define the decision before collecting the observations.

The customer-gap conversation can lead to a trial. This article owns operational acceptance of that proposed supply, not another analytics-pilot specification or account-growth experiment.

Bound the trial to a real operating question

State the category, specifications, sites, expected order pattern and duration. Describe what the customer wants to learn: compatibility, usable delivery performance, service responsiveness or a particular cost assumption. A trial that attempts to answer every question will be difficult to review.

Confirm that the customer has authority to conduct it under its existing arrangements. Use the incumbent-switching review to identify relevant constraints. Do not treat a seller's trial invitation as purchasing approval.

Distinguish trial scope from proposed rollout scope. One site using one item does not test every location or every variant of the category.

Agree inputs and measurement rules

Useful records include trial order IDs, agreed quantities, specifications, requested and confirmed delivery dates, actual receipts, quality observations and customer acceptance. Specify who records each event and who can resolve a disputed observation.

For delivery performance, define whether the unit is an order, line or shipment. Identify permitted changes to the requested date and how partial delivery is treated. The OTIF comparison guide explains why delivery measures are not interchangeable.

Choose thresholds for this customer's requirements. They are negotiated trial criteria, not universal supplier benchmarks. Critical safety, compliance or suitability questions need the customer's appropriate technical reviewers and separate acceptance rules.

Review a synthetic trial

The DEMO-601 trial below is invented. Twenty orders are eligible under an agreed order-level rule. The declared delivery gate requires at least 19 of 20 to arrive on time and in full; the trial also requires no critical quality failure.

Observation Recorded result Agreed gate
Eligible orders 20 All planned eligible orders reviewed
On time and in full 18 At least 19
Critical quality failures 0 None
Orders with a minor quality issue 1 Review before expansion

Delivery attainment is 18 / 20 = 90%, below the declared 95% gate. The absence of critical failures does not cancel that result. A minor quality issue needs its recorded disposition rather than being hidden in a summary score.

The review can decide to investigate the two delivery exceptions and conduct a defined retest. It cannot report that the original trial passed by retrospectively reducing the delivery threshold to 90%.

Understand what the observations do not establish

Twenty orders are a small operational sample. The result describes those orders under those conditions, not a guaranteed long-run delivery rate. A trial conducted during quiet weeks may not test peak demand, remote sites or a different product mix.

Record missing observations explicitly. If a receipt time is unavailable, distinguish unknown performance from a confirmed late receipt. The decision may require recovering the record or extending observation, rather than assigning a convenient result.

Use association and cause when explaining apparent improvements. A before-and-after change can reflect demand, staffing or other conditions as well as the proposed supplier.

Make the review decision actionable

Record go, revise, retest or stop with the evidence, accountable reviewers and next scope. A conditional go might permit the same item at another similar site after the minor issue is resolved. It should not silently authorize every category in the proposal.

Revisit the customer's switching business case using observed costs where available. Preserve assumptions the trial did not test. The UK government's contract-management principles offer contextual guidance on ownership and controlled changes, rather than these specific thresholds.

Keep the trial separate from a sales achievement

Report trial orders, accepted scope and later repeat purchasing separately. The repeat-demand guide follows the category after its first purchase. A trial receipt alone does not establish ongoing demand.

Bring the decision beside the customer-growth review example. Share an authorized trial sample through Covirage contact to agree the required analytical view and review scope.

Questions people ask

Is a supplier trial the same as an analytics pilot?

No. A supplier trial evaluates a proposed product or service in the customer's operation. An analytics pilot evaluates data, calculations and analytical outputs. Each needs its own acceptance criteria.

Can a successful trial be presented as proof of annual savings?

Only report what the observation supports. Short or unusually favorable conditions may not establish annual costs, seasonal performance or results at additional sites.

Should a failed trial always end the opportunity?

Follow the agreed rules. A remediable noncritical issue may justify a bounded retest, while a disqualifying failure can require stopping. Do not change the threshold retrospectively to manufacture a pass.