A merchandise quality inspection checklist is useful when it connects a measurable requirement to a specific sample, a recorded observation, and a named decision. This guide provides a downloadable sample approval scorecard and quality calculator, then explains how to use it without confusing a favorable score with permission to manufacture or ship. Procurement teams, distributors, brand owners, and campus store operators can adapt the same process to their own products and inspection plans.
Download the sample approval workbook
Download the English sample approval workbook — XLSX, version 1.1. Download the companion English record template — CSV, version 1.0.
The workbook contains six worksheets: Instructions, Sample Brief, Inspection Log, Scorecard, Decision Record, and Lists. The CSV is a blank exchange template for transferring selected fields into another system. It does not contain formulas, validation lists, or the workbook's decision logic. Keep the workbook when you need calculations and use the CSV when you need a portable record structure.
This is an original operational model dated September 14, 2026. Its example limits and weights demonstrate the arithmetic; they are not market benchmarks, contractual promises, or a recommended sampling plan. Before inspecting goods, replace or explicitly approve those assumptions with the buyer and the responsible quality specialist. Save a separate working copy for each product, artwork revision, and inspection round so that subsequent corrections cannot overwrite the earlier evidence.
The inspection area has thirty rows. Use one inspected item per row, or consistently defined groups of equal size with comparable scoring methods. A larger or statistically designed inspection needs an appropriate inspection system; do not quietly extend ranges or claim that this small workbook supplies a formal sampling procedure.
Separate four decisions that often get combined
Artwork approval confirms the proposed wording, version, dimensions, and position of decoration. It can establish whether a name is spelled correctly or whether the intended logo fits the specified area. It does not establish the feel of a textile, the adhesion of a transfer, or the condition of a shipped carton. Record the approved artwork reference and the exact scope of that approval.
Physical sample approval evaluates an identified object. The sample should match the intended material, decoration process, finish, and packaging to the extent stated in the brief. If its packaging or material is provisional, record that limitation rather than describing the entire product as approved. Keep photographs and, when practical, retain the controlling sample with an identifier that the supplier can match.
Lot inspection examines goods from a defined production population using an agreed method. A beautiful development sample cannot establish the quality of goods produced later. Equally, inspecting a few convenient units from the top carton does not justify a claim about every carton. Record who selected the inspected units and what population they represent.
Final release combines the inspection decision with outstanding conditions, quantity, documents, and shipment instructions. Give one named person authority to release the identified goods. A worksheet recommendation remains an input to that decision; it does not notify the warehouse, revise a purchase order, or establish that regulatory obligations have been met.
Define the requirement before assigning a score
Start with a product identity that survives an email handoff: item code, supplier, purchase or sample reference, artwork revision, decoration method, and lot quantity. Add the intended use and the feature being assessed to the approved tolerances field or its linked specification. “Premium finish” needs a physical reference or observable acceptance condition before two reviewers can apply it consistently.
For logo position, identify a stable reference point on the actual product. State the nominal distance, unit of measurement, permitted deviation, and orientation. For a bag, measuring from a shifting fold produces a different result from measuring from a defined seam. Record the measuring method with the tolerance so another inspector can repeat the assessment.
For color, agree on the physical reference and viewing conditions. A screen image alone cannot settle every disagreement about a finished product. Record the material, surface, lighting arrangement, and any approved boundary sample. Avoid assigning a numerical color tolerance unless the responsible team has defined how it will be measured and interpreted.
For packaging, distinguish appearance from protection. An attractive presentation box can still require a separate shipping configuration. The approval record should identify the actual insert, closure, protective materials, and carton arrangement being reviewed. If transport testing is required, link its report and acceptance criteria; an appearance score does not replace that test.
Use a defect catalog with observable examples
Define critical, major, and minor categories for the specific product before inspecting it. A critical category should capture conditions the responsible team considers unacceptable because of their consequences. Major and minor classifications need equally concrete descriptions. Severity is a decision about the defect and its context, not a synonym for how expensive the order is.
Create a short catalog with a photograph, requirement, measurement method, and severity for each recurring issue. A misspelled university name, a visibly rotated mark, and a small scuff on an outer carton need not share a category. Record borderline examples explicitly. Otherwise, one reviewer may count a defect that another treats as acceptable variation.
Count occurrences consistently. If one unit has a printing problem and a packaging problem, this workbook can record two defect occurrences against one inspected unit. Its displayed observed defect rate is therefore defect occurrences divided by inspected quantity. It is not automatically the percentage of distinct defective units. Maintain a separate unique-unit count if your inspection plan requires that measure.
Do not compare these two measures as though they were interchangeable. Two occurrences among twenty inspected items equal ten occurrences per hundred items. They may affect one item or two. That distinction matters when estimating the number of replacements, designing sorting work, or discussing supplier responsibility.
Preview the inputs and outputs
Workbook field guide: enter observed facts separately from buyer choices and calculated recommendations.
| Area | What the buyer records | How it affects the decision |
| Sample Brief | Identity, artwork, tolerances, quantities, owner, date, evidence, limits, weights | Defines the specific review and its assumptions |
| Inspection Log | Observed quantity, defect counts, four scores, item reference, evidence | Supplies the facts used in calculations |
| Scorecard | Calculated occurrence rate, weighted score, completeness, recommendation | Highlights results requiring an accountable review |
| Decision Record | Disposition, decision owner, reasons, corrective action, release evidence | Preserves the actual decision separately from the recommendation |
| Lists | Decoration and action-status choices; illustrative arithmetic cases | Supports consistent entries without prescribing buyer policy |
The yellow input cells remain editable. Enter zero when an assessed defect category has no observed occurrences. Leave an unassessed score blank and document what is missing. Do not enter a favorable score merely to make the recommendation appear complete. Evidence fields should point to records reviewers can actually open, not only to a folder name or an inaccessible personal drive.
The workbook checks several necessary fields, but it cannot judge whether a photograph proves the claimed condition or whether the person named has authority. Review those matters explicitly before release. Treat imported or pasted data with the same care as manually entered data because application validation controls can behave differently across spreadsheet software.
Understand the calculation before changing thresholds
The occurrence rate adds critical, major, and minor defect counts, then divides by total inspected quantity. Zero inspected quantity produces no rate. The quality score averages each of four score columns and combines those averages using the weights entered in the Sample Brief. The sum of weights normalizes the result.
Occurrence rate = (critical + major + minor occurrences) / inspected quantity
Quality score = (color average × color weight
+ placement average × placement weight
+ dimension average × dimension weight
+ packaging average × packaging weight) / total weight
Each populated score row contributes equally to its column average. Quantity does not weight those row averages. Mixing one row for a single item with another row for ninety-nine items can therefore distort the intended interpretation. Use comparable rows and disclose the unit of observation. Separate materially different products instead of combining them into one apparently precise score.
With complete inputs, a critical count above its configured limit or an occurrence rate above the rejection boundary produces Reject. Acceptance requires the rate to be at or below its acceptance ceiling, the score to meet its minimum, and major and minor counts to meet their limits. Other complete results require Review. Equality at the rejection boundary alone does not trigger rejection; the formula uses a strict greater-than comparison.
Choose buyer limits without turning examples into policy
The demonstration settings are a two-percent acceptance ceiling, a five-percent rejection boundary, zero critical occurrences, two major occurrences, five minor occurrences, and a minimum score of eighty-five. These are illustrative inputs only. Their suitability changes with product risk, inspection quantity, defect definition, and the agreement between buyer and supplier.
A fixed count limit behaves differently when the inspected quantity changes. Two major occurrences among ten items are not equivalent to two among a thousand. Before using the calculator, make the inspected quantity and decision plan explicit. Do not take the sample size from this article or select a convenient quantity merely because its arithmetic produces a favorable rate.
Weights also encode preferences. Giving color thirty percent and packaging twenty percent does not prove color is objectively more important. Ask the brand owner to explain the consequence of a poor result in each dimension. Where a requirement is mandatory, keep it as a separate release condition rather than allowing high scores elsewhere to compensate.
Document any concession against the actual affected quantity and requirement. Keep the original limit and record who approved the exception, why, for which units, and for how long. Changing a threshold after seeing the inspection result makes it difficult to distinguish a legitimate commercial concession from an altered record.
Hypothetical case one: campus tote samples
A campus store is reviewing a proposed order of three hundred totes. Before production, its buyer chooses to examine twenty development samples, one per row. This is a hypothetical development review, not a statistically justified acceptance sample for the future order. The brief identifies the artwork revision, an agreed logo-position tolerance, and a retained reference sample.
The inspector observes no critical defects, one major placement defect, and no minor defects. The occurrence rate is one divided by twenty, or five percent. Average color, placement, dimension, and packaging scores are ninety, eighty, ninety-five, and ninety. At weights of thirty, thirty, twenty, and twenty percent, the weighted score is eighty-eight.
Using the demonstration limits, the recommendation is Review. The score exceeds eighty-five, but the occurrence rate exceeds the two-percent acceptance ceiling. It equals the five-percent rejection boundary, so that condition does not by itself produce Reject. The buyer now has a specific question: is the placement problem isolated, or does the production setup need correction?
The chosen response is to hold physical sample approval, ask the supplier to correct its placement reference, and examine a newly identified sample round. The alternative is a documented concession if the brand owner accepts the deviation. That option preserves timing but changes the approved appearance; it must not be treated as an invisible tolerance adjustment. The release evidence is the revised reference, new measurements, and a dated decision.
Hypothetical case two: onboarding notebook pack-out
A company plans an onboarding kit containing a notebook and separate accessories. The quality owner examines twenty-five notebook packages from a defined preparation round. The example quantity is chosen for illustrating the workbook only. Each row represents one package, and the reviewer records both the presentation condition and the relevant packaging evidence.
Suppose three packages have a classified minor scuff and there are no major or critical occurrences. The rate is three divided by twenty-five, or twelve percent. Four average scores of ninety-five would still produce a weighted score of ninety-five. With the demonstration five-percent rejection boundary, the result is Reject. Strong scores cannot cancel that rate condition.
The buyer compares replacing the sleeves, changing the pack-out, or accepting the scuffs under a limited concession. Replacement sleeves address appearance but may fail again if rubbing inside the parcel caused the damage. Changing the pack-out can require new materials and another review, while a concession preserves the schedule at the cost of the agreed presentation.
The team chooses a revised protective arrangement and a separate reinspection record. It retains the failed observations and asks for evidence of the changed configuration. If the team only substitutes three visibly damaged sleeves without examining the cause, it has corrected those samples but has not established whether the remaining packages face the same problem.
Run the review with clear ownership
Procurement opens the working copy and confirms the product and purchase references. The brand owner supplies the controlling artwork, visible tolerances, and any approved appearance references. The quality owner defines the observation method, severity catalog, quantity, and scoring rubric. These inputs should exist before the inspector records results.
The inspector records facts with item or carton identifiers. For dimensional observations, retain the actual measurements in the evidence record, not only the resulting score. For subjective scoring, use anchors such as clearly acceptable, borderline, and unacceptable, with examples agreed by the brand owner. Ask a second reviewer to resolve disputed classifications before the summary is finalized.
The decision owner reviews the calculated recommendation alongside unmet requirements and business constraints. A supplier can propose a correction but should not silently act as the buyer's approval authority. If responsibility is shared, record the scope of each sign-off: artwork, product quality, commercial concession, and physical shipment release.
The logistics owner receives an explicit release reference with product, quantity, destination, packaging revision, and exceptions. A screenshot of a green cell is insufficient instruction. Require acknowledgment that the dispatch team received the correct version, particularly when a later artwork or packaging change might otherwise travel in a separate email chain.
Handle missing evidence and corrective action
When the workbook shows Incomplete, first check required identity fields, numeric thresholds, weights, observed quantities, scores, and evidence references. A missing assessment should remain visible as missing. Entering zero is appropriate for an inspected defect category with none found, but zero is a real score when entered into a score column; it is not a substitute for “not assessed.”
The completeness indicator is an input check, not a document audit. Open the linked evidence, confirm its revision and item identity, and make sure all four scoring dimensions use comparable observations. If a dimension cannot be assessed, hold the recommendation and arrange the assessment or have the quality owner design a suitable alternative model. Avoid editing formulas merely to suppress an inconvenient warning.
For a failed requirement, record the affected population, immediate containment, proposed correction, owner, due date, and verification method. Identify whether the issue arose from artwork, material, decoration setup, handling, or packaging. Keep the original failure record so the next review can establish what actually changed.
The overdue flag compares the corrective-action due date with the current date while the action remains uncompleted. It is a reminder, not an escalation workflow. Someone must review outstanding actions, contact the responsible person, and update the release decision. Closing an action means recording the verification evidence, not merely entering a completion status.
Keep the workbook separate from formal sampling standards
The official ISO 2859-1:2026 page describes sampling schemes indexed by acceptance quality limit for lot-by-lot inspection. Last verified September 14, 2026. This workbook supplies an original operational scorecard and does not implement that standard's sampling tables or certify compliance.
When does the team need a separate inspection plan?
Use an appropriately designed plan when deciding whether a production lot meets a specified acceptance scheme. The responsible specialist should establish the population, selection method, required observations, acceptance rules, and response to failure. Preserve that plan and its results alongside the workbook. A small development review can inform a sample decision without establishing a statistical conclusion about future production.
Do not describe the percentage in this calculator as an acceptance quality limit. It is an observed occurrence rate calculated from your entries. Similarly, a high weighted score does not prove product safety, durability, material compliance, or the performance of untested goods. Keep specialist tests and mandatory requirements as explicit evidence requirements outside the appearance score.
For the broader sourcing sequence, see the bulk corporate gifts guide to samples, customization, quality, and delivery. Use this workbook for the specific sample record while retaining the commercial and logistics decisions in their appropriate order documents.
Release a traceable decision
Before authorizing the next step, confirm that the record answers four questions: what was examined, against which requirements, what was observed, and who decided what happens next. The recommendation is useful only when another person can reconstruct its inputs and connect it to the physical goods.
-
Product, artwork, packaging, and sample identifiers match the evidence.
-
Limits and weights were agreed before reviewing results.
-
Observations, scores, and evidence use a consistent unit of review.
-
Unmet mandatory requirements and concessions are explicitly resolved.
-
Corrective actions have owners, deadlines, and verification records.
-
The named decision owner has released the exact quantity and revision.
Keep the blank master separate from completed buyer records. Restrict access to working files that contain supplier details or commercial information, and share only what the recipient needs. When a new version of the template changes formulas, retain the previous completed record and document the migration rather than replacing historical results.
For teams connecting approved merchandise to a larger gifting program, Giftpack's Global Swag Factories provides a commercial starting point for discussing custom merchandise execution. Bring the approved specification, sample record, and unresolved conditions into that discussion so sourcing and delivery work can proceed from the same decision evidence.