Efficient, but small
A high return on a modest spend. Genuinely good, and it will not move the category on its own.
Scale it
Increase the commitment where the mechanic is proven.
Promotional Margin Recovery
Which events actually earned their trade spend?
A governed workflow that connects POS, promotion master, vendor fund, cost and hierarchy data into ranked, confidence-tagged recommendations — routed to category managers, trade marketing and joint business planning teams.
What the workflow runs on
Feeds you already have
POSPromotion masterVendor fundsCostHierarchy
Governed decision logic
Store-SKU-weekBaseline decompositionFour-tier bandingConfidence
What reaches the team
Ranked eventsFund reallocationBuild size
The challenge
The second-largest line on the P&L
Trade promotion is the second-largest cost line on a CPG retailer’s P&L, yet the methodology to measure what each event truly earns stays fragmented across teams and systems. When that gap persists the consequences compound: zero-return events repeat year after year, inventory builds are sized to vendor overforecasts, and cannibalization between competing promotions stays invisible to every category manager reviewing events separately.
Builds follow the vendor’s forecast
Shoppers face out-of-stocks or markdowns because inventory is sized to a vendor projection rather than to independently measured lift.
No shared view of net margin
Category managers and trade marketing work from different numbers, and halo and cannibalization belong to neither of them.
Reporting, not decisioning
Promotion analytics runs after the event rather than inside the planning cycle, when the decision is still open.
Reallocation is manual and reactive
There is no ranked evidence tied to the JBP cycles where change is actually possible, so funds move by argument rather than by return.
And the retailer arrives at every joint business planning table without independent evidence of its own.
The leadership question
One question, asked in every review
“How do we get a view of which events earn their trade spend — with forecasts replenishment and finance can trust — without overhauling our planning systems all at once?”
The split
High return is not the same as high margin
Cutting on percentage alone quietly kills the events that carry the category. The workflow produces both a return ratio and an absolute margin figure, because a portfolio decision made on one without the other is a decision made half-blind.
Large, but thin
Substantial absolute margin on a heavy spend, at a return that looks unimpressive as a ratio.
Restructure, do not cut
Change the mechanic before you surrender the volume.
How the two axes resolve
Return against absolute margin
Every event lands in one of these corners. The banding exists precisely so that large-volume events are never over-cut by a portfolio decision made on percentage alone.
Return on trade spend →
Small · Strong returnScale
Efficient and under-committed. Increase the weight.
Large · Strong returnProtect
Earning its spend at scale. Defend it at JBP.
Small · Weak returnCut
Neither efficient nor material. Reallocate the funds.
Large · Weak returnRestructure
Too much volume to cut. Change the mechanic.
Absolute margin →
What is included
Six capabilities, six deliverables
Each one is a packaged capability — the data model, the scoring logic and the deliverable already designed around the decision it serves. They are built to run together, but each produces an artefact that stands on its own.
01
Promotional Measurement Model
Promotional Data Backbone
A single, normalized store of every sales, event, vendor fund, cost and product record the workflow touches — so every baseline, lift figure and return output traces back to one auditable, versioned source rather than a patchwork of category-level extracts assembled by hand each cycle.
02
Baseline Decomposition Engine
Baseline & Lift Estimation Engine
Time-series decomposition per store, SKU and week that isolates trend, seasonality, day-of-week and calendar effects, then resolves gross lift into true net incremental units after halo gains on related items, cannibalization losses on substitutes and pull-forward adjustments — producing numbers that survive vendor scrutiny.
03
Event Performance Scoring Framework
Event Return & Scoring Framework
Defined, transparent formulas combining net incremental margin against full event cost — vendor funds applied, retailer margin sacrifice and operational cost-to-serve — producing both a return ratio and an absolute margin, so large-volume events are never over-cut by portfolio decisions made on percentage alone.
04
Pre-Event Forecast Index
Pre-Event Forecast & Activation
A forward-looking forecast generated for every planned event before it runs — predicting lift, sizing the inventory build and attaching an independent return estimate, so replenishment and finance hold a retailer-owned reference before vendor commitments are made.
05
Reallocation Brief
Category & JBP Decision Outputs
Lift decomposition, fund efficiency, vendor forecast bias by event mechanic and event pairings, surfaced through category and vendor dashboards — routing exception queues to category managers and structured negotiation artefacts to trade marketing ahead of each joint business planning cycle.
06
Data Confidence Model
Governance & Data Confidence
Every recommendation carries a transparency rating — high, medium or low — based on the completeness of the data underneath it, with gaps surfaced explicitly so reviewers know what to trust, what to question and what needs enrichment before any action is taken.
The artefacts
What actually arrives, and who it arrives for
Six named deliverables rather than access to a tool. Each is versioned, traceable to the model that produced it, and routed to the team that acts on it.
Deliverable 01 · Promotional Data BackbonePromotional Measurement Model
A single, normalized store of every sales, event, vendor fund, cost and product record the workflow touches — so every baseline, lift figure and return output traces back to one auditable, versioned source rather than a patchwork of category-level extracts assembled by hand each cycle.
Deliverable 02 · Baseline & Lift Estimation EngineBaseline Decomposition Engine
Time-series decomposition per store, SKU and week that isolates trend, seasonality, day-of-week and calendar effects, then resolves gross lift into true net incremental units after halo gains on related items, cannibalization losses on substitutes and pull-forward adjustments — producing numbers that survive vendor scrutiny.
Deliverable 03 · Event Return & Scoring FrameworkEvent Performance Scoring Framework
Defined, transparent formulas combining net incremental margin against full event cost — vendor funds applied, retailer margin sacrifice and operational cost-to-serve — producing both a return ratio and an absolute margin, so large-volume events are never over-cut by portfolio decisions made on percentage alone.
Deliverable 04 · Pre-Event Forecast & ActivationPre-Event Forecast Index
A forward-looking forecast generated for every planned event before it runs — predicting lift, sizing the inventory build and attaching an independent return estimate, so replenishment and finance hold a retailer-owned reference before vendor commitments are made.
Deliverable 05 · Category & JBP Decision OutputsReallocation Brief
Lift decomposition, fund efficiency, vendor forecast bias by event mechanic and event pairings, surfaced through category and vendor dashboards — routing exception queues to category managers and structured negotiation artefacts to trade marketing ahead of each joint business planning cycle.
Deliverable 06 · Governance & Data ConfidenceData Confidence Model
Every recommendation carries a transparency rating — high, medium or low — based on the completeness of the data underneath it, with gaps surfaced explicitly so reviewers know what to trust, what to question and what needs enrichment before any action is taken.
Previews are schematic. They carry no figures, because the numbers on them would be ours rather than yours.
The output
Every event lands in one of four bands
The bands are what let a category manager tell an event to scale from an event to cut without relitigating the methodology every quarter.
TIER 1Scale
Earning its spend with room to grow. Increase the commitment and repeat the mechanic.
TIER 2Protect
Carrying real margin at scale. Defend it with evidence at the next planning cycle.
TIER 3Restructure
Material volume at a weak return. Change the mechanic rather than surrender it.
TIER 4Cut
Neither efficient nor material. Reallocate the funds to something that earns them.
The workflow
From a scanned unit to a defensible number
Each step is explicit and inspectable. That matters because the output has to survive a vendor challenging it across the table, not just an internal review.
1
NormalizeSales, event, fund, cost and product records into one store
2
DecomposeTrend, seasonality, day-of-week and calendar effects isolated
3
AttributeGross lift resolved to true net incremental units
4
CostVendor funds, margin sacrifice and cost-to-serve applied
5
ScoreReturn ratio and absolute margin, both carried forward
6
RouteException queues to category, negotiation artefacts to trade
Every baseline, lift figure and return output traces back to one auditable, versioned source rather than a patchwork of category-level extracts.
What changes
Four things that stop being true
The workflow does not add another report. It replaces four sentences that get said in every review with something specific enough to act on.
“The event performed well.”
Net of halo, cannibalization and pull-forward — and against its full cost, not just its funds.
“We’ll size the build to the vendor forecast.”
Builds are sized to a retailer-owned pre-event forecast with a confidence bound.
“Cut the bottom decile by return.”
Ranking carries absolute margin too, so the category’s volume engines are not cut by ratio.
“We’ll take the vendor’s numbers into JBP.”
You arrive with independent evidence and a brief your whole team shares.
Who it is for
One workflow, four seats
The same governed output, presented for the decision each team actually makes. Nobody is asked to interpret somebody else’s view.
Category Managers
A clear separation between events to scale and events to cut, using pre-event confidence intervals rather than vendor estimates to size inventory builds.
Trade Marketing & JBP Leads
The Reallocation Brief — lift decomposition, vendor forecast bias by mechanic, and a success paired to every underperformer — ready for the negotiation.
Merchandising & Finance Leadership
Trade spend efficiency, the share of positive-return events and reallocatable fund value, mapped directly onto gross margin lines.
Replenishment & Inventory
Builds sized to the pre-event confidence upper bound, cutting in-event stockouts and the post-event excess that drives markdown exposure.
How it is governed
What makes it repeatable
Anyone can produce this read once. What makes it a capability is that the grain, the logic and the model version are fixed and stored, so the next cycle is measured the same way as the last one.
Grain
Store × SKU × week. Baseline, lift and return all resolve to it, so finance and category are never reconciling two different truths.
Transparency
The scoring formulas are defined and inspectable rather than proprietary. A vendor can challenge the number and be answered.
Confidence
Every recommendation is rated high, medium or low on the completeness of the data beneath it, with the gaps named.
Versioning
The measurement model is versioned, so a number produced this quarter can be reproduced next quarter and compared honestly.
Start with one category.
Bring a single category and one season of events. We’ll decompose the baseline, resolve the true net lift and show you the reallocation brief it produces — before you commit to anything wider.