Cost per impression · denominator · redemption · retention · framework · defensible build

Cost of Merch Per Impression in 2026: A Defensible Build

The cost-per-impression concept in B2B merchandise is borrowed from paid-media planning and applied to a category where the mechanics are different. A digital impression is a single, countable event with no physical carrying cost and no retention behaviour; a merch impression is a multi-touchpoint event that depends on the item being kept, being used, and being seen by people the buyer did not choose. The result is that a simple cost-per-impression figure for a merch programme is usually wrong, and a marketing lead who quotes one without naming the assumptions is talking past the question. This page sets out what belongs in the denominator, why the metric is easy to misuse, and how to build a defensible version of it that the marketing team and the procurement team can both sign off on.

Talk to us about measurementAll B2B notes

Why a merch impression is not a digital impression

The simplest definition of a cost-per-impression metric for merch is the all-in cost of the programme divided by the number of impressions the programme produces. The definition is identical in structure to the definition used in paid-media planning, which is the reason it gets borrowed. The mechanics are not identical, however, and that is where misuse starts. A digital impression is served once and counted once; a digital impression's denominator is built from server logs and is largely vendor-validated. A merch impression is produced over the life of an item and counted through observation; the denominator is built from a behaviour pattern the buyer has to estimate and from data the buyer has to collect.

The mechanics that make the borrowed metric hard to apply are three. First, a merch programme produces impressions over months, not at the moment of receipt, and the count depends on where the item goes and who sees it. Second, an item may be discarded before any impression is produced, in which case the cost of producing it is sunk without a return on the metric. Third, an item may be retained for years and may be seen by hundreds of people the buyer did not select, in which case the denominator is far larger than the buyer's distribution list suggested. None of these mechanics is bad in itself; the issue is that none of them is captured by the simple definition, and the marketing lead who uses the simple definition produces a figure that is either too flattering or too pessimistic depending on which mechanics are assumed.

Digital impression
Single, countable event, server-logged, vendor-validated
Merch impression
Multi-touchpoint over the life of the item, observation-counted
Discard behaviour
Cost sunk without impression; affects denominator
Retention behaviour
Prolonged impressions outside the buyer's distribution list
Implied denominator
Always depends on stated assumptions

Four decisions that determine whether the metric is defensible

A cost-per-impression figure for a merch programme can be defensible if four decisions are made explicitly rather than implicitly, and if every internal stakeholder signs off on the same version of each decision. The four decisions are: what counts as an impression, how retention is measured, how the unit is costed, and how the figure is reported. Each one is described below with the alternative choices and the trade-offs they introduce.

What counts as an impression

An impression is a single counted instance of a person seeing the item. The decision is whether to count an impression at receipt, at first use, at every use, or at the cumulative total over the item's life. The most generous count (every use over the item's life) produces the largest denominator and the lowest cost-per-impression figure, and is the count that flatters the programme most. The most conservative count (receipt only) produces the smallest denominator and the highest cost-per-impression figure, and is the count that flatters the programme least. A marketing lead who wants a defensible figure names the count explicitly — "second use and beyond" is a useful convention because it captures the impressions that matter for brand reach and excludes the receipt moment, which is an event the buyer chose rather than an impression the recipient earned — and applies the same count across programmes that are being compared.

How retention and use are measured

The retention figure is the largest source of variance in any cost-per-impression calculation for merch, because retention determines how many impressions the item produces in total. A retention measurement has three choices: self-reported (a survey sent to a sample of recipients), observed (a usage log on a small share of items, typically impractical at scale), or modelled (a retention-rate assumption based on prior programmes). Self-reported is the most common and the most reliable if the survey is short, the sample is large enough, and the response rate is above a defensible threshold; modelled is the only practical choice when the buyer cannot run a survey, but the model has to be named and the source of the assumption has to be stated. The honest move is to choose one method, name it, and apply it consistently.

How the unit is costed

The unit cost in the denominator is the all-in landed cost of the programme at the buyer's named Incoterm, including freight, packaging, decoration, setup amortised across the order, artwork preparation, project coordination, and any account-management line that the supplier has applied. The decision is whether to use the headline per-unit figure, the line-item per-unit figure, or a fully loaded figure that includes internal coordination time. The most defensible figure is the line-item per-unit figure because it can be replicated at any future programme against the same supplier pricing structure; the fully loaded figure is the most accurate, but it requires time-tracking and cost-allocation discipline that most internal teams do not maintain. A marketing lead who uses the headline figure is using a figure the supplier chose to lead with, which is a figure designed to flatter the supplier rather than to inform the buyer.

How the figure is reported

The reporting decision is the one most often skipped, and it is the one that determines whether the metric survives internal scrutiny. The minimum discipline is to report the metric with three numbers visible: the unit cost, the impression count, and the cost-per-impression result, with the assumptions on each stated in the same paragraph. The figure should also be reported with the comparator (a different programme, a different category, a different retention assumption) so that the reader can see what the figure is sensitive to. A single number reported without its assumptions is a number that cannot be defended in a meeting, even when the number itself is correct.

The misuse patterns the marketing lead should anticipate

The cost-per-impression metric for merch is misused in five predictable patterns, and a marketing lead who knows the patterns can pre-empt them. The patterns are: inflated impression counts, suppressed unit cost, opaque retention assumptions, single-programme reporting, and the comparison-with-paid-media fallacy. Each is described below with the move that corrects it.

Inflated impression counts happen when the impression count includes receipt, unprompted recall, social-media mentions and third-party observation without naming the count source. The corrective move is to settle on a single impression definition and to apply it consistently. Suppressed unit cost happens when the per-unit figure excludes freight, packaging, project coordination or amortised setup, which produces a flattering denominator but a number that cannot be replicated. The corrective move is the line-item cost build. Opaque retention assumptions happen when the retention rate is assumed without naming the assumption, which produces a metric that no one outside the marketing team can audit. The corrective move is to publish the assumption and to state the sensitivity of the metric to the assumption. Single-programme reporting happens when the metric is produced for one programme only, which produces a number that cannot be benchmarked. The corrective move is to compute the metric for every programme in the same cadence.

The comparison-with-paid-media fallacy is the misuse pattern with the highest reputational cost. A cost-per-impression figure for merch is not directly comparable to a cost-per-impression figure for paid digital, because the impression mechanics are different and the value of the impression to the brand is different. A marketing lead who claims a merch programme produces impressions at one-tenth the cost of paid search is making a claim that does not survive the impression-mechanics question, and the claim damages the credibility of the merch programme rather than defending it. The corrective move is to compare merch programmes against merch programmes, in the same category, at the same cadence, with the same assumptions, and to avoid cross-category comparisons that conflate different impression mechanics.

Building a defensible cost-per-impression figure for a programme

A defensible build for a merch programme cost-per-impression metric is a four-step process: name the assumptions, build the cost, count the impressions, and report the figure. Each step has a defined output and a defined owner, and the four together produce a figure that the marketing team, the procurement team and the finance team can sign off on. The steps are short; the discipline is in the naming, not in the calculation.

Step one: name the assumptions. The programme lead writes a half-page document with the impression definition, the retention measurement method, the unit-cost basis (line-item per unit), the calculation window, and the source of each assumption. The document is shared with procurement and finance before the campaign starts. Step two: build the cost. Procurement produces the line-item all-in cost per unit against the final specification, including freight and packaging, and confirms it in writing. Step three: count the impressions. Marketing commissions a short survey at the appropriate measurement window (typically ninety days for a corporate gifting cohort; six to twelve months for an apparel cohort), computes the retention rate and the average impressions per retained item, and multiplies through to the total impression count. Step four: report the figure. The figure is reported in the same document as the assumptions, with the comparator programme at the same cadence and the same metric definition. The four steps produce a number that survives questions, and a number that survives questions is a number the programme can be defended on.

Questions buyers ask about the cost-per-impression metric

What is the most common mistake in a merch cost-per-impression build?

Opaque retention assumptions. The unit cost is visible because the supplier's invoice is visible; the impression count is visible because the survey can be run again; the retention assumption is the variable that tends to be invisible because it is built into the calculation without being documented. The corrective move is to publish the retention assumption and to test the sensitivity of the final metric to the assumption. A figure whose answer moves tenfold when the retention assumption moves ten percentage points is a figure that cannot be used for budgeting.

Can we use the same metric for different categories of item?

In principle, yes, but only if the impression definition is consistent across categories. A bottle, a notebook, a hoodie and a tote have different retention patterns and different per-use impression counts, and a metric built on one category cannot be compared to a metric built on another without restating the impression definition. The discipline is to use the same definition and to report each category separately rather than aggregating them into a single programme-level figure that hides the category variation.

Is a single cost-per-impression figure enough to defend the programme?

No. The figure is one of three numbers a programme needs to be defensible: the cost-per-impression figure, the retention rate at the measurement window, and the programme's share of the marketing budget. The three together provide the answer to the question that a CFO or a procurement director will eventually ask: what did the programme cost, how much of it is still working, and how does the cost compare to the alternative uses of the budget. A programme that can answer only one of the three is a programme that is not yet ready for board-level scrutiny.

Is Corp-Merch the official supplier or affiliate of any measurement body, brand, or industry reference on this page?

No. Corp-Merch is an independent B2B merchandise supplier operating in Armenia. References to impression definitions, retention measurement methods, paid-media conventions and any measurement body or trade term are informational only and imply no partnership, sponsorship, endorsement or official-supplier relationship with any of them. Nothing on this page is a guarantee of measurement outcome; every measurement is dependent on the buyer's data collection, sample design and assumption discipline.

This page is written for marketing, brand and procurement leads who are asked to defend a corporate merch programme against a cost-per-impression question. Corp-Merch is an independent B2B merchandise supplier. The framework on impression definition, retention measurement, unit-cost build and reporting is a reasoned synthesis of practice in this category as we understand it; it is not marketing research, legal or measurement advice, and the framework should be calibrated to each programme's actual data collection discipline and policy. Named measurement conventions, paid-media categories and trade terms are informational only and do not imply partnership, endorsement or recommendation by the bodies that publish them.