PRODUCT UNITS / COMPLETED PURCHASES / MONEY
Count the event your decision needs.
A public-listing buying brief for understandable vending sales metrics.
01 · METRIC REVIEW
Introduction
A vending dashboard shows “120 sales.” Before the operator uses that number to compare cabinets or estimate payment costs, one question needs an answer: does it mean 120 products, 120 completed purchases or something else? When a shopper can buy several products together, those counts can be very different.
The WM22 listing describes multiple purchases in one transaction. The AI fridge also describes multiple-item purchases. These public statements support a real purchasing question about report definitions. They do not establish how the supplied reports name, count or aggregate those events. A two-cabinet station adds a further scope question because its actual transaction and reporting configuration require confirmation.
This guide concerns the denominator of a sales metric. It does not test product-recognition accuracy, reconcile bank settlement or establish a formal revenue-recognition policy. The buying goal is a report definition and demonstration that let the operator explain the same set of purchases without substituting units for transactions.
02 · METRIC REVIEW
Quick Answer
Request separate definitions for completed purchases, product units and monetary totals. For each metric, agree the event counted, included statuses, reporting period, cabinet scope and treatment of reversals. Then demonstrate those definitions with a small set of known purchases.
Do not infer transaction count from a unit total. Do not infer customer count from either. One person can make several purchases, and one purchase can include several units. A report may not identify people at all, and there is no need to request unnecessary personal data to verify this arithmetic.
If the actual software exposes only some of the required metrics, ask which supported records can answer the remaining question. Do not promise a calculated average or an export field until the supplier has shown the necessary inputs and their definitions.
03 · METRIC REVIEW
Comparison Table
| Metric in the buying brief | Proposed meaning to confirm | What it cannot establish alone |
|---|---|---|
| Completed purchases | The count of purchases in an agreed completed state. | Units sold, unique customers or settled bank receipts. |
| Product units | The total quantity of included product units. | How many separate purchases occurred. |
| Included monetary total | The agreed amount for the included purchase records. | Cash in the bank or profit. |
| Units per purchase | Included units divided by the matching purchase count. | Recognition accuracy or customer satisfaction. |
These are proposed definitions to discuss, not verified WEIMI report labels. The same reporting scope must apply to the numerator and denominator. A calculation is not meaningful merely because both inputs are visible somewhere in the portal.
04 · METRIC REVIEW
Who Should Buy This
Use this review when comparing connected vending proposals, planning multi-item shopping or preparing a fleet reporting handover. It is useful for an operator who wants to compare buying patterns across locations without confusing a busy item counter with a high number of separate purchases.
The operator, commercial analyst and supplier should agree the definitions together. A payment contact can explain which external record counts chargeable transactions for a particular service. That external definition should not be assumed to match the vending report automatically.
For a small pilot, a few controlled examples may be enough to expose an ambiguous label. For a larger fleet, retain a metric dictionary so that later dashboards and exports can be interpreted consistently. No specific reporting or export feature is established here.
05 · METRIC REVIEW
How We Evaluate Smart Vending Machines
We compare three real WEIMI public listings as a procurement shortlist, not an independent equipment test. The public descriptions establish multi-item shopping on two candidates and a two-section configuration requiring confirmation on the third. They do not provide a complete metric dictionary or a demonstrated transaction-count report.
Our proposed review starts with known examples: one completed purchase of one unit, a second completed purchase of several units and a third completed purchase containing repeated units of the same product. Write the expected totals before the supplier demonstration. Ask the supplier to show how the offered records represent each example.
Only after the basic definitions are clear should the review add the relevant exception cases. Ask about cancelled, failed, refunded or partially reversed purchases where the supplied system supports them. Their names and treatment must come from the actual offered workflow, not from assumptions about a generic vending platform.
06 · METRIC REVIEW
Key Buying Factors
1. Name the event counted
Ask whether a counter refers to initiated shopping sessions, completed purchases, product lines, delivered units or another event. A short label such as “sales” is insufficient. Retain the supplier’s exact definition alongside the metric used in the buying decision.
2. Separate product lines from physical units
A purchase of two identical bottles might contain one product line with quantity two. Ask which count the report exposes. A line count cannot be substituted for units unless the actual data guarantees one unit per line, and no such guarantee is established by these public pages.
3. Keep the period and location scope identical
Use the same reporting dates, time basis and cabinet selection for all inputs to a ratio. A fleet unit total divided by one cabinet’s purchase count produces an unrelated number. If the supplied report has filters, ask how the active selection is made visible and retained in the evidence.
4. Define included states and reversals
Ask what happens to each metric when a supported purchase is cancelled, fails or is refunded. A report can show gross and net views, but the offered definitions must be confirmed. Do not invent a universal rule that every refund removes one transaction or returns a product unit to stock.
5. Reconcile detail to the displayed total
Where the offered system provides purchase detail or an export, compare the included records with the headline. If it does not, ask what demonstration evidence can verify the count. Do not claim that a CSV download or transaction identifier exists until it has been shown.
6. Avoid turning purchases into people
A purchase count is not a unique-customer count. Demonstrating the difference needs no real names, payment-card details or personal profiles. Use a controlled supplier demonstration and retain only the information necessary to connect the examples to their metric results.
7. Keep payment-provider definitions separate
A payment service may have its own chargeable-event definition. Ask the provider to explain its actual fee schedule. Use the appropriate event count for that schedule rather than multiplying a per-transaction charge by product units simply because the vending dashboard calls them sales.
07 · METRIC REVIEW
Best Smart Vending Machines
WM22 Touchscreen Snack and Drink Vending Machine
Read the WM22 public listing. The page describes a 21.5-inch touchscreen, multiple purchases in one transaction, inventory management and remote operation. The multi-item statement makes a units-versus-purchases demonstration relevant to the quotation.
Ask how the supplied software represents a purchase containing three products and one containing two units of the same product. Confirm the available record and report fields instead of inferring them from the inventory feature. Product fit, selected dispensing mechanisms and local payment support remain separate purchasing checks.
Single-Door AI Vision Smart Fridge for Packaged Drinks
Read the AI fridge public listing. This cabinet uses camera-based recognition for compatible packaged drinks and snacks. Its public description supports multiple-item shopping and calls for testing actual packaging, shelf arrangements and shopping scenarios. Five shelf levels and five baskets describe its listed layout, not its sales denominator.
Ask how a completed shopping session becomes the reported purchase count and how recognized quantities become the unit count. Recognition validation and report aggregation are separate questions. A correct multi-item example does not establish statistical accuracy, a unique-customer metric or a particular payment fee.
Two Cabinets, More Choice | Snack & Drink Vending Station
Read the dual-cabinet public listing. The station has a main display and a secondary visible spiral-lane compartment beneath a menu and payment area. The public page asks buyers to confirm the controller, transaction flow and whether reports distinguish the two selling sections.
First establish which purchases the actual configuration supports. If the project requires a purchase containing items from both sections, ask whether that function is offered before defining a station-wide average. Two physical sections do not prove one basket or two transactions. Retain the supplier-confirmed scope in the metric definition.
08 · METRIC REVIEW
Feature Comparison
| Candidate | Relevant public statement | Specific report question |
|---|---|---|
| WM22 | Multiple purchases in one transaction. | Are completed purchase count, product lines and units separately identifiable? |
| AI vision fridge | Multiple-item shopping and recognition validation. | How does the completed session relate to purchase count and recognized unit quantities? |
| Dual-cabinet station | Two selling sections; transaction/report scope to confirm. | What is the actual purchase boundary and location scope of each count? |
This table compares evidence questions, not measured reporting quality. No common software platform, export format or analytics feature is inferred across the three listings. Select the complete configuration after confirming the required records.
09 · METRIC REVIEW
Cost & ROI Analysis
All examples are fictional assumptions. Assume three completed purchases contain respectively one, three and two product units. Assume their included monetary amounts are US$3, US$8 and US$5. Exclude tax, refunds and all other adjustments in this simplified example.
| Illustrative purchase | Product units | Included amount |
|---|---|---|
| Purchase A | 1 | US$3 |
| Purchase B | 3 | US$8 |
| Purchase C | 2 | US$5 |
| Matching totals | 6 units / 3 purchases | US$16 |
Under those assumptions, units per purchase are 6 ÷ 3 = 2. The monetary amount per purchase is US$16 ÷ 3, approximately US$5.33 when displayed to two decimals. The amount per unit is US$16 ÷ 6, approximately US$2.67. Those averages answer different questions; neither is a unique-customer value or a profit figure.
For a separate fee illustration, assume a provider charges US$0.10 for each of these three purchases, with no other fees. The assumed charge is US$0.30. Multiplying by six product units would produce US$0.60 and use the wrong denominator for that hypothetical schedule. Actual providers may define fees differently; obtain their complete terms.
These calculations do not establish a machine’s actual average basket, fee, sales volume or ROI. A real investment model needs actual prices, product costs, payment terms, site charges, software fees and operating evidence. A higher unit count alone cannot demonstrate better commercial performance.
10 · METRIC REVIEW
Best Choice by Scenario
Comparing single-item and multi-item shopping
Use the same metric definitions before comparing locations or formats. WM22 and the AI fridge publicly describe multi-item purchases, but their actual reporting fields require confirmation. Compare units per matching completed purchase rather than assuming each removed or dispensed unit is a separate transaction.
Reviewing a two-section station
Resolve the dual-station transaction boundary first. Then agree whether the report is station-wide or section-specific. Do not divide a two-section unit total by a single section’s purchase count without an explicitly supported and meaningful scope.
Estimating per-transaction service costs
Obtain the provider’s actual chargeable-event definition. Match the relevant record count to that definition. The vending report can be a useful input only after the relationship is demonstrated; it is not automatically the provider’s billing ledger.
11 · METRIC REVIEW
Applications
At quotation stage, include the metrics the operator actually needs and ask the supplier to identify their availability and definitions. Avoid demanding a generic dashboard screenshot as evidence for an unnamed denominator. A narrow supported metric is more useful than an ambiguous headline.
Before handover, retain the controlled example set, the report selection and the resulting totals. If the actual system supports export, confirm that the export preserves the relevant quantities and scope. A table header can make a number readable without establishing its business definition; verify both.
When a workflow or software version changes, review whether the definitions changed. Keep the accepted dictionary with the operating documentation. This guide does not promise an automatic version history, real-time updates, customer identification or an accounting integration.
12 · METRIC REVIEW
FAQ
Does six units mean six transactions?
No. In the fictional example, six units belong to three completed purchases. The actual supplied system’s counting definitions must be confirmed.
Is a product line the same as a unit?
Not necessarily. One product line can carry a quantity greater than one. Ask what each reported field counts.
Does purchase count identify unique customers?
No. One person can make several purchases, and a report may not identify people. No unique-customer feature is established here.
Should a refund always remove one purchase from the count?
No universal treatment is specified. Ask how the offered report defines its included statuses and gross or net views.
Does the dual station guarantee one cross-section basket?
No. Confirm the actual purchase and controller configuration before defining a combined metric.
Are the US$5.33 average and US$0.30 fee real performance figures?
No. They follow from explicit fictional assumptions and do not describe a machine, provider quotation or investment return.
13 · METRIC REVIEW
Final Recommendation
Agree what the report counts before using it to justify equipment or operating costs. Separate completed purchases, product units and money, and keep their period, location and status scope aligned. Demonstrate a few known examples and retain the definitions with the result.
WM22 and the AI fridge provide public multi-item starting points; the dual station requires its transaction scope confirmed. These are public-listing procurement candidates, not independently tested winners. Choose the format that meets the project and can provide understandable evidence for the metrics the operator will actually use.
14 · METRIC REVIEW
CTA
Request a reporting demonstration with defined counts
Send WEIMI your country, sites, machine quantity, products, package dimensions, storage needs and payment preferences. List the report metrics you need and the planned single-item or multi-item workflow.
Ask for a controlled example showing purchase count, unit quantities and the matching monetary total. Confirm the available reports, export scope and any software charges in the quotation, alongside product fit and service terms.
Get My Custom QuoteSource basis: the three public WEIMI product listings linked above, reviewed 15 October 2026. Definitions and demonstrations are proposed buying requirements. All numeric examples are assumptions. No private report feature, independent test, accounting conclusion, measured sales result or search-volume claim is established.
