An operator opens a vending site and performs several agreed purchases to check selection, payment and product collection. The first sales report later includes those purchases alongside customer activity. The cabinet may have recorded genuine operating events, but the report population is wider than the customer-demand question the buyer wants to answer.

This distinction matters when the early total is used to choose an assortment, add a second cabinet or compare locations. A controlled commissioning purchase has a known demonstration purpose. It should not silently become evidence that an ordinary customer wanted the same product at that time. Equally, excluding it from a demand view does not make the underlying stock movement or payment disappear.

The public dual-cabinet WEIMI listing asks buyers to confirm sales reports and demonstrates a purchase sequence. WM22 and the AI fridge also describe multi-item purchasing and management functions. These pages support a practical reporting question; they do not establish a native test flag, a commissioning filter or a particular accounting treatment.

Plan how controlled purchases will be identified before performing them. Agree the demonstration with the supplier and payment provider, retain the actual references available, and ask how those events appear in stock, purchase and financial records. Then define a separate customer-demand view using the identified population.

If the offered reporting tool supports an appropriate filter or label, have the supplier demonstrate it. If it does not, ask whether an approved export and a separately documented analysis can meet the requirement. Neither route is established by the public listings. Preserve the source records and the explanation of each adjustment.

Treat the classification as an analysis boundary. It is not a direction to delete sales, issue refunds, change tax records or alter financial recognition. The responsible finance team should decide how real payments and other movements are recorded. A successful demonstration and a customer-demand result can both be useful while answering different questions.

These are proposed evidence distinctions. They do not establish which fields or export interfaces exist in the quoted software. Ask the supplier to name the actual reports and their definitions instead of copying the table labels into a product specification as verified features.

Use this brief when an operator plans opening-day checks, a distributor demonstrates equipment to a buyer or a fleet team compares an initial pilot with existing locations. It is useful whenever controlled purchases occur inside the same period or reporting environment used for commercial review.

Include the person arranging the demonstration, the operator reviewing demand and the finance contact responsible for payments and stock. If a third party supplies the terminal or reporting service, identify its role too. One person may perform several tasks, but the distinctions should remain clear in the evidence file.

A free mechanical delivery exercise may create stock movement without a paid purchase. A provider-approved paid demonstration may create different records. Ask which process is actually proposed before applying this guide. No free-vend capability, sandbox, payment reversal or refund entitlement is assumed for any machine.

We compare three real WEIMI public product listings as a procurement shortlist. We have not made purchases, inspected private reports or measured demand at an installed site. The following method is a proposed supplier demonstration and reporting review, not an independent product test or a finding that a platform misclassifies activity.

Start by defining the business question: for example, ordinary customer unit demand during a stated opening period. Next, identify the controlled commissioning activity that will happen inside that period. Agree the safe equipment procedure and the payment-provider arrangements before the demonstration, including who supplies products and who handles the resulting records.

For each agreed example, retain the references the offered system actually provides. Ask the supplier to show its appearance in the relevant records and any supported analysis method. Compare the unchanged source total with the explained customer-demand subset. Record unresolved identification limits; do not manufacture missing references or assume a report is complete from one visible total.

Specify the purpose before the event

Define which actions are part of commissioning or demonstration and who authorises them. Write the expected equipment result and reporting follow-up in advance. Retrospectively calling an inconvenient purchase a test can bias the demand review; a planned evidence list gives the team a clearer basis for classification.

Use references rather than a time window alone

Retain the actual supported purchase reference, cabinet identity, product and quantity where available. A narrow time window can help locate an event, but it may also contain real customer activity. A time-based exclusion should disclose that limitation instead of pretending every event in the interval was controlled.

Define the quantity being compared

Keep purchase count, product units and monetary totals in separate columns. A commissioning purchase containing several products can affect those measures differently. This guide focuses on which events belong in the demand population, but the chosen metric still needs a clear definition before totals are compared.

Separate exclusion from financial reversal

Removing an identified event from a demand calculation does not prove that a charge was voided, refunded or never settled. Ask the payment provider to explain the actual record and any supported test arrangement. Keep financial follow-up with the responsible team instead of using an analysis filter as evidence of money movement.

Keep stock movements visible

Products collected during commissioning are physical movements even when the analysis excludes them from ordinary demand. Record the approved stock handling separately, including any returned or removed samples. Do not automatically add units back because the associated purchase is classified as a demonstration.

Ask what the reporting tool actually supports

Request a demonstration of any proposed label, filter or export. Ask whether the analysis changes the source record or creates a separate view. If a manual working file is needed, preserve the source copy, the identified event list and the calculations. No automatic test-transaction feature is verified here.

Agree an owner for incomplete evidence

A controlled purchase may lack a matching reference in one of the reviewed records. Assign the follow-up and retain the uncertainty. If the team cannot distinguish some demonstration events from customer activity, state that the opening-period demand evidence is mixed rather than claiming a precise adjusted result.

WM22 Snacks and Drinks Vending Machine

Read the WM22 public listing. It describes a 21.5-inch touchscreen, inventory management and multiple purchases in one transaction. Optional spiral, conveyor, direct-push and hanging arrangements are listed. Confirm the supplied mechanism and the actual products before designing a purchase demonstration.

For this reporting review, ask how one agreed commissioning purchase appears when it contains several product units. Retain the product and purchase evidence under their actual definitions. The listing does not establish a test flag, sales-report exclusion, automatic refund or native separation of commissioning activity from customer demand.

Single-Door AI Vision Smart Fridge for Packaged Drinks

Read the AI fridge public listing. Customers select compatible packaged products directly and close the door for automated checkout. The page calls for actual packaging, shelf arrangement and multi-item recognition testing. Its listed structure has five shelf levels and five baskets, with usable capacity depending on the products and approved layout.

A recognition demonstration can produce evidence about product identification and quantity without demonstrating ordinary customer preference. Ask how its actual purchase and stock records can be located for the later demand review. Do not assume that a recognition test is a paid sale, or that a paid demonstration has no financial consequence. This cabinet does not prepare fresh juice.

Two Cabinets, More Choice | Snack & Drink Vending Station

Read the dual-station public listing. It describes a main product display and a secondary visible spiral-lane compartment. The page asks buyers to demonstrate selection and delivery from each section, and to confirm stock and sales tools, including whether they distinguish the two cabinets.

Identify controlled purchases from both sections in the evidence plan. A combined early total can otherwise obscure how much activity was performed only to demonstrate each selling area. Confirm the actual reporting and transaction scope. A shared basket, independent section report or common test filter is not established by the two-cabinet appearance.

This is an evidence-request comparison, not a ranking of report quality or a claim that any candidate already supplies the proposed analysis. Ask for a worked demonstration in the actual quoted configuration and retain the supported method in the project file.

Fictional arithmetic only: assume an opening-period source report contains 30 product units. A reviewed evidence list identifies six of those units as controlled commissioning purchases. Under the example’s defined scope, the remaining ordinary-customer population is 30 − 6 = 24 units. No real report, sales result or site performance is represented.

The six commissioning units are 20% of the all-event total, because 6 ÷ 30 = 20%. That is a different denominator from the 25% comparison in the last row. Neither percentage measures a machine benefit or proves that a future forecast would differ by the same amount.

For a separate fictional cost allowance, assume the evidence review takes 15 minutes at US$28 per hour. The planning amount is 0.25 × US$28 = US$7.00. This is not a measured duration, WEIMI service price or payback estimate. Actual demonstration, product, payment, travel and review costs require their own scope.

Do not multiply commissioning activity into projected demand. Likewise, do not treat the adjusted subset as a complete investment case. Expansion decisions need representative operating evidence and actual equipment, site, product and service costs. Finance should review monetary totals independently under the applicable business records policy.

A new dispensing cabinet has opening-day purchase checks

Begin with WM22’s confirmed mechanism and product list. Agree the controlled purchase examples, preserve their references and ask how they will appear in the records. Use a separately explained demand population when reviewing the first operating period.

A smart fridge needs product recognition demonstrations

Use the AI fridge’s packaging and shopping-scenario context. Record expected and observed products for the approved examples. Their role is to evaluate the workflow; customer preference needs its own evidence. Keep any real payment handling visible to the responsible parties.

A two-section station needs both product areas demonstrated

Use the dual station with a section-specific evidence list. Confirm whether the reporting tools can identify each area and how the team will distinguish controlled activity. A successful demonstration in one section does not supply the missing evidence for the other.

At quotation stage, ask what records and analysis routes are included in the offered service. If the buyer requires a native commissioning label or filter, identify it as an unconfirmed requirement until the supplier demonstrates it. Do not assume it is bundled into a generic management-system statement.

During commissioning, use the agreed supplier procedure and provider-approved payment arrangement. Retain an event list with the actual available references and the purpose of each example. Keep the observed physical outcome, system record and financial follow-up distinct. This guide performs no live purchase or payment test.

During the first demand review, preserve the source report and document the subset calculation. Have the responsible reviewer confirm the event population and note any uncertainty. If more demonstration purchases occur after a software or configuration change, add them to the same reviewed process rather than treating the initial launch date as the only possible boundary.

Approve a clear commissioning evidence plan before the first controlled purchase. Identify the events, retain the original stock and payment evidence, and define which population the customer-demand review will count. An unadjusted opening total can be accurate for its own scope while being unsuitable for a different question.

WM22, the AI vision fridge and the dual station provide three real public starting points for the equipment shortlist. Their purchase formats create different demonstration needs. None is presented as an independently tested reporting winner. Choose the configuration and supported evidence method that make the intended commercial review defensible.

Source basis: the three linked WEIMI public listings, reviewed 15 October 2026. No private report, transaction, native test filter, payment reversal or actual demand result was verified. The evidence plan is a procurement proposal. All numeric examples are fictional assumptions.