Written by: Bryan Grobstein, Vice President, Global Revenue, AnyRoad | Last updated: October 25, 2026
Key Takeaways
- Retail shopper marketing ROI measures incremental sales and profit caused by campaigns, not total sales during the campaign period.
- A credible ROI calculation requires a counterfactual baseline, incremental lift estimation, contribution margin conversion, full cost subtraction, and final ROI and ROMS computation.
- Common mistakes that distort ROI include counting all campaign sales as incremental, using revenue instead of contribution margin, ignoring cannibalization, and treating attribution as causation.
- Closing the activation-to-purchase gap requires receipt-verified purchase data that links experiential activations directly to retail sell-through.
- AnyRoad provides the technology to capture first-party consumer data at activations and verify purchases through receipt-based rebates without retailer POS integration.
See How Receipt-Verified Purchase Data Works
How To Measure Marketing Campaign ROI: The Seven-Step Sequence
A defensible retail shopper marketing ROI number follows a fixed sequence. Each step depends on the one before it, and skipping any step produces a number that will not survive scrutiny from a CFO or category buyer.
- Establish your baseline and counterfactual.
- Estimate incremental lift.
- Convert lift to profit using contribution margin.
- Build and subtract the full cost stack.
- Calculate ROI and ROMS with a worked example.
- Close the activation-to-purchase gap.
- Interpret the number.
The sections below work through each step with the data to pull, the method to use, and the mistakes to avoid.
Step 1: Establish The Baseline And The Counterfactual
Attribution shows which purchases were connected to a campaign under a defined rule and lookback window. Incrementality goes further: it asks whether those purchases would have happened without the marketing. The IAB and IAB Europe Guidelines for Incremental Measurement in Commerce Media, published November 3, 2025, draw this line explicitly: attributed sales assign credit to observed purchases, but do not demonstrate that removing the marketing would have prevented them. A method must satisfy three requirements to claim causality: establish a credible counterfactual, control or reduce bias, and separate signal from statistical noise.
Three counterfactual methods are available for in-store shopper marketing, and each suits a different tactic.
Test vs. control stores is the appropriate method for displays, sampling, and in-store activations where the treatment is applied at the store level. Circana's Liquid Testing measures causal incrementality by comparing test stores against matched control stores using store-level POS data, selecting comparable store groups based on historical sales patterns, demographics, distribution, pricing dynamics, and external factors such as seasonality. NIQ takes a complementary approach: it recommends designed experiments where a retail partner allows them, combined with an ensemble of econometric regression modeling and passive experimentation, identifying naturally occurring situations in which only one material variable changed.
Exposed vs. unexposed shoppers is the appropriate method for loyalty programs and digital coupons where a shopper-level identifier exists. Loyalty card data enables brands to compare purchase behavior among members who received an offer against those who did not, while controlling for pre-existing purchase frequency.
Modeled lift uses synthetic controls, propensity models, or marketing mix modeling when no clean control group exists. The IAB grades econometric methods as high on scope but moderate to weak on causality, with stated weaknesses including being backward-looking, lacking granularity, and not being predictive for one-off campaigns. Modeled lift works well for strategic budget allocation. It does not replace a designed experiment when decision stakes are high.
Platform-reported lift, the incrementality figure most retail media network dashboards surface, is classified by the IAB as a hybrid proxy with weak causal strength. Circana reports that practitioners frequently see a clear gap between attributed ROAS and incremental ROAS, with attributed ROAS suggesting a campaign is wildly successful while iROAS reveals a more modest but accurate incremental lift. Attribution and ROAS show what happened, while incrementality shows what the campaign caused.
Prove Incremental Retail Sales Impact From Your Experiences
Step 2: Estimate Incremental Lift Against The Counterfactual
Incremental lift quantifies the difference between what happened with the campaign and what would have happened without it. The counterfactual method you choose in Step 1 determines how you calculate that lift.
In a test vs. control design, lift equals the sales difference between treated and matched control stores over the campaign period. In an exposed vs. unexposed design, lift equals the difference in purchase behavior between shoppers who received the treatment and those who did not, after controlling for baseline behavior. In a modeled approach, lift comes from the model's predicted baseline compared with observed sales during the campaign window.
Lift should be expressed in both units and revenue. This dual view keeps the later profit and ROI calculations grounded in the actual volume change, not just topline dollars.
Step 3: Build The Full Cost Stack
An incomplete cost stack is the most common reason a shopper marketing ROI number collapses under scrutiny. The full cost stack for an in-store activation includes six categories. Production costs cover displays, POSM, and signage. Agency and creative fees pay for the design work. Retailer co-op and slotting fees buy the placement. Activation labor and ambassador wages pay the people running the activation. Samples and logistics cover the product and its movement. Incremental discounts or rebates pay for trial and redemption.
The double-counting trap is structural. In most CPG organizations, the shopper marketing budget is split between marketing and sales: trade spend held by sales buys the placement from the retailer, while the shopper marketing budget held by marketing buys what occupies that placement, and because two functions co-fund a single activation, neither is clearly accountable for the full cost. Costs funded by a different department are still costs. A program that omits them will post an inflated ROI that does not survive a line-by-line review.
Forrester documented that 36% of retail media activities are funded through existing trade marketing budgets and 26% through shopper activation budgets, which illustrates how fragmented budget ownership makes the full cost stack genuinely difficult to assemble. Assembling that full picture is the practitioner's job before presenting a number to leadership. Once the cost stack is complete, the next decision is which performance signal to measure for each tactic.
Step 4: Map Each Tactic To Its KPI
Different tactics produce different signals, and measuring a display program with a coupon redemption metric, or vice versa, produces a number that is technically accurate and practically useless. For full KPI definitions by tactic, see Shopper Marketing KPIs: The Complete Measurement Guide. The primary measures by tactic are:
- Endcaps and displays: incremental sales lift, display compliance rate, and units per store per week.
- Coupons: redemption rate and incrementality over baseline, plus new-to-brand rate among redeemers.
- Sampling: trial-to-purchase conversion, cost per trial, and repeat purchase rate at 30, 60, and 90 days.
- Retail media: incremental ROAS measured against a holdout, with attributed ROAS reported separately and labeled as such.
- Loyalty and CRM: incremental basket size and repeat purchase rate among exposed members versus unexposed members.
- Co-marketing: incremental category lift and halo effect on adjacent SKUs in the same portfolio.
For alcohol and CPG brands running ambassador tastings inside third-party retail accounts or on-premise accounts, the historical default measure is the depletion report: cases moved from distributor to account. A depletion report proves execution, not causation, and that gap is addressed in Step 6.
Step 5: Calculate ROI And ROMS With A Worked Example
The formula for calculating ROI in FMCG uses contribution margin, not revenue. Revenue-based ROI flatters unprofitable promotions by ignoring the cost of goods and variable costs embedded in every unit sold.
- Incremental gross profit = incremental sales × contribution margin
- Net return = incremental gross profit − total campaign cost
- ROI (%) = net return ÷ total campaign cost × 100
- ROMS (return on marketing spend) = incremental sales ÷ total campaign cost
The following example shows why contribution margin, not revenue, determines whether a campaign is profitable. Consider a $50,000 campaign that drives $150,000 in incremental sales at a 40% contribution margin. The campaign generates $60,000 in pre-media incremental gross profit ($150,000 × 40%). After subtracting the $50,000 campaign cost, net contribution profit is $10,000. For the same $50,000 campaign generating $150,000 in incremental revenue, ROI is 200% (($150,000 – $50,000) / $50,000 × 100). The revenue-to-spend ratio is 3.0x, which is a 300% ROAS, not a 20% ROI and 3.0x ROMS.
ROI and ROMS are not interchangeable because ROMS ignores margin. A 3.0x ROMS on a 40% margin product produces a positive ROI, but the same 3.0x ROMS on a 20% margin product produces a loss, since the revenue is identical but the profit left after cost of goods is not. Reporting ROMS without margin context is as incomplete as reporting revenue without cost of goods. For the formula fundamentals, see How To Calculate Shopper Marketing ROI.
Step 6: Close The Loop From Activation To Receipt-Verified Purchase
For in-store activations, even a well-designed test leaves one gap unaddressed: linking a specific consumer interaction to a specific purchase. Ambassador tastings and in-store activations have historically produced only a depletion report. Cases moved from distributor to account confirm that product left the warehouse. They do not confirm that a specific consumer, reached at a specific activation, bought a bottle because of that interaction. That is the activation-to-purchase gap. Closing it requires a mechanic that works without a POS integration from the retailer or bar, which generic ROI frameworks do not provide.
AnyRoad closes that loop. The mechanic works as follows: a brand ambassador uses AnyRoad Live to capture consumer data via QR code at the activation. The consumer receives an SMS with a cashback rebate redeemable anywhere the product is sold. They buy the bottle, photograph the receipt, and text it back. AnyRoad's AI reads the receipt, confirms the eligible SKU, and pays the rebate via Venmo or PayPal. No retailer POS integration is required. The same mechanic can run as a sweepstakes, with an entry to win a prize rather than cash back, and AnyRoad handles regulatory compliance, terms and conditions, and privacy policies.
In a one-month ambassador tasting pilot with a single unnamed craft brand on a small dataset, 56% of the records collected converted to a bottle purchase. This was an early, small-sample pilot and should not be read as a platform benchmark. It illustrates the category of data the mechanic produces: a first-party record tied to a receipt-verified purchase, which connects experiential and shopper marketing spend to retail sell-through. For more on how the rebate mechanic works in practice, see Shopper Marketing Rebate Software: Link Activations to Sales.
See How AnyRoad Closes The Activation-To-Purchase Gap
Step 7: How To Interpret The Number
What counts as a good ROI depends on margin, category, and tactic. A 20% ROI on a low-margin promoted item can represent less absolute profit than a smaller percentage return on a high-margin one. The defensible test is whether incremental gross profit clears the full cost stack and the brand's hurdle rate, rather than whether the percentage looks impressive in a recap deck.
ROI and ROMS tell different stories about the same campaign. A program can post a healthy ROI and a weak ROMS, or the reverse, depending on margin structure and the ratio of incremental sales to campaign cost. Both numbers belong in a leadership presentation, labeled clearly and calculated from the same incremental gross profit figure.
A lift estimate that sits inside the statistical noise floor is not a result. The IAB guidelines state that a lift estimate sitting inside the noise floor is unusable regardless of how good the design was, and recommend confidence intervals that exclude zero, bootstrapping, falsification tests, and sensitivity analyses as the standard for a reportable number.
Frequently Asked Questions
What Is The Formula For Calculating ROI In FMCG?
As shown in Step 5, the formula uses contribution margin, not revenue. Incremental gross profit equals incremental sales multiplied by contribution margin. Net return equals incremental gross profit minus total campaign cost. ROI equals net return divided by total campaign cost, multiplied by 100. Using revenue instead of contribution margin overstates the return on any promotion where variable costs are significant.
How Do You Measure Marketing Campaign ROI?
Establish a credible counterfactual, such as test vs. control stores, exposed vs. unexposed shoppers, or a modeled baseline, to separate incremental sales from sales that would have happened anyway. Estimate incremental lift against that counterfactual. Convert incremental units to incremental gross profit using contribution margin. Subtract the full cost stack, including production, agency fees, slotting, labor, samples, logistics, and incremental discounts. Calculate ROI and ROMS from the resulting net return.
Is 20% ROI High?
It depends on margin, category, and tactic. As Step 7 explains, the percentage alone does not indicate value, and margin and absolute profit matter more. The relevant question is whether incremental gross profit exceeded the full cost stack and cleared the brand's internal hurdle rate for the tactic in question.
What Is A Good Marketing ROI Ratio?
There is no universal threshold. The defensible standard is whether incremental gross profit, calculated at contribution margin rather than revenue, clears the full cost stack and the brand's hurdle rate for the specific tactic. A retail media campaign and a sampling program have different cost structures, margin profiles, and repeat-purchase dynamics, so a single benchmark ratio applied across tactics will misallocate budget.
What Is Considered A Bad ROI?
Any ROI where the cost stack is incomplete, the baseline is campaign-period sales rather than a counterfactual, or the lift estimate sits inside the statistical noise floor fails the credibility test, regardless of the percentage. A number that looks positive because costs were omitted or because organic sales were counted as incremental is not a measurement result. It is a reporting artifact that will not survive a line-by-line review.
Conclusion: Measure What The Campaign Caused
The structural takeaways from this playbook are three. First, measure what the campaign caused, not what happened while it ran, since campaign-period sales require a counterfactual before they become an incrementality number. Second, use contribution margin and the full cost stack, because revenue-based ROI and incomplete cost stacks produce numbers that flatter unprofitable programs. Third, close the loop from activation to receipt-verified purchase so the team can connect experiential spend directly to retail sell-through.
For the formula fundamentals and a step-by-step calculation walkthrough, see How To Calculate Shopper Marketing ROI.
See How AnyRoad Connects Activations To Retail Sales