The short answer
Yes, if the available retailer data support a clearly defined sales outcome. Map which outlets, periods, and purchases the data cover before setting the MMM scope. Keep consumer purchases separate from shipments to retailers, and report missing sales channels explicitly. Results for observed retailers should not be presented as the brand's total sales impact.
This article is part of Asked by Marketers, a series answering real questions from marketing leaders.
Sellforte's team holds more than 1,450 meetings each year with marketing leaders in Ecommerce and Retail about Marketing Mix Modeling and incrementality testing. Each week, we anonymize at least one question from those conversations and answer it in depth, based on what marketers are actually struggling with, not what keyword tools suggest. About the series →
Why marketers ask this
A brand may have detailed sales data from its own website, useful reports from a marketplace, and limited information from specialist retailers. The missing retailers can still be important to a product launch. Marketing needs to understand what can be measured now and whether the available results will support decisions for the whole business.
What does partial retailer coverage mean for MMM?
It limits the sales outcome the analysis can describe. Begin by identifying exactly which consumer purchases appear in the data, rather than judging readiness from one overall coverage percentage.
A file might include all sales from a subset of retailers, selected products from every retailer, or a changing sample of stores. Those situations create different questions. A stable history for a clearly defined group is easier to interpret than a total whose composition changes without being recorded.
Marketing Mix Modeling estimates effects on the outcome supplied to it. If the outcome contains direct sales and one marketplace, the analysis can address that scope. It cannot establish the sales effect at retailers whose outcomes were never observed.
| Data question | Why it matters |
|---|---|
| Which stores, products, and sales channels are included? | Defines the part of the business represented by the outcome. |
| Does the date describe a consumer purchase or a shipment? | Determines which event the model would explain. |
| How often are sales reported, and at what geographic level? | Limits the time and geographic detail available for measurement. |
| Has the coverage changed during the history? | Helps distinguish changes in the dataset from changes in demand. |
| How do totals reconcile with the retailer's reporting? | Reveals gaps or definition differences before modeling. |
Can shipments to retailers replace missing consumer sales?
Shipments to retailers should not silently replace purchases by consumers. Sell-in records the retailer buying stock from the brand; sell-out records the retailer selling that stock to the consumer. The dates and volumes can differ.
A retailer can replenish inventory before a campaign, place a large order for several weeks of demand, or sell existing stock while placing no new order. A shipment series therefore does not describe the same event as the consumer-sales outcome marketing may want to measure.
Keep both measures clearly named when both are available. If the business wants to explain wholesale ordering, define that as a separate measurement question and assess the data accordingly. Do not label a shipment-based analysis as proof of consumer sales lift.
This distinction is also relevant to incrementality testing. A test intended to measure consumer purchases needs an outcome that represents those purchases. The event date and geographic coverage must support the comparison being made.
How should we choose an initial MMM scope?
Choose a scope with usable data and a decision the business can act on. An initial analysis might cover direct sales and selected retail partners, while a channel with inadequate history remains outside the measured outcome.
Write the scope into the reporting language. “Incremental sales through the measured channels” tells the reader more than “total marketing impact” when several retailers are absent. Keep the missing channels visible in the review so they do not disappear from the budget discussion.
Consider where the advertised products are sold. If a new product is concentrated in a retail channel with little consumer-sales data, strong coverage elsewhere does not resolve the launch question. A good model of other channels can still leave that decision unanswered.
As additional retailer feeds become usable, expand the outcome deliberately and reconcile the new coverage. Adding outlets can raise the reported sales total even if demand at existing outlets has not grown. Record the change so it is not mistaken for a marketing improvement.
The article on measuring digital advertising's effect on offline sales explains the broader modeling approach once the relevant sales data are available. The first decision here is which offline sales can actually enter that analysis.
Is there a minimum retailer-data coverage percentage?
A coverage percentage alone is not enough to judge whether the proposed analysis will answer the business question. Assess the available history and its composition against the specific outcome, campaign reach, and decision.
Ask the modeling team to inspect a representative data sample. Have them identify which outcomes can be supported, which limitations affect interpretation, and which additional data would materially improve the answer. Avoid accepting a percentage threshold without understanding what that percentage covers.
Do not scale an observed lift to the whole retail network solely by dividing by the share of sales observed. The missing retailers may sell different products, serve different audiences, or have different exposure to the campaign. Extrapolation requires evidence for those assumptions; a coverage ratio does not supply it.
How this looks in practice
Consider a hypothetical brand planning a campaign across its website, a marketplace, and specialist stores. The following data are available. This is an editorial illustration, not a customer result.
| Sales channel | Available records | Initial treatment |
|---|---|---|
| Own website | Consistently defined consumer orders | Assess for inclusion in the modeled sales outcome. |
| Marketplace | Consumer sales by period and product | Assess usable detail and align with the measurement scope. |
| Specialist stores | Shipments from the brand only | Keep outside a consumer-sales outcome until suitable data are available. |
The first model can help the team understand measured website and marketplace outcomes if those feeds pass the data assessment. It should not conclude that advertising has no effect in specialist stores.
If the launch is concentrated in those stores, the team still needs relevant consumer-sales evidence before judging the launch's total impact. The narrower model answers a useful part of the question, and the reporting makes the remaining gap explicit.
How Sellforte helps
Sellforte works with teams to define the sales channels and data that can support an MMM analysis. The measurement scope can expand as additional feeds become usable, with the outcome definitions and reporting boundaries kept explicit. Book a demo.
Authors
Lauri Potka is the Chief Operating Officer at Sellforte and has over 15 years of experience in Marketing Mix Modeling, marketing measurement, and media spend optimization. Before joining Sellforte, he worked as a management consultant at the Boston Consulting Group, advising some of the world's largest advertisers on data-driven marketing optimization. Follow Lauri on LinkedIn, where he is one of the leading voices in MMM and marketing measurement.