The $2.4M Decision That Looked Right
In Q2 2025, a mid-size beverage CPG made a pricing decision that was, by every internal metric, correct.
Their RGM team had modeled the move carefully. Elasticity coefficients checked out. Historical velocity data supported the price increase on their flagship SKU. The board had approved it. The trade team had communicated it to retail partners. Everything was aligned.
What the model did not know: in the six weeks leading up to that decision, diesel prices had risen 14%. Their primary competitor had just absorbed a significant freight cost increase across their network. Their competitor's margin was under acute pressure. Their competitor could not afford to respond to a price increase without compressing their own P&L further.
The pricing ceiling in that category had moved. And the CPG brand's model was still pointing at where it was ninety days ago.
They executed the price increase at a level the model said was appropriate. The competitor blinked. But the CPG brand had priced below the ceiling that now existed. Conservative estimate of the margin opportunity left on the table: $2.4 million.
Not from a bad decision. From a model that did not know what the market had just told it.
The Quarterly Review Problem
This story is not unusual. Versions of it play out every pricing cycle across the CPG industry.
Most CPG pricing models are built on historical data—and they are technically accurate about the past. The elasticity coefficients are real. The baseline volume estimates reflect what actually happened. The promotional lift factors are derived from actual events. The model is correct.
The problem is that commercial decisions happen in the present, in response to market conditions that change continuously. And the gap between when market signals shift and when they appear in a pricing model's inputs is, for most CPG companies, somewhere between 60 and 90 days.
That is not a data problem. The signals are available. Diesel prices are published daily. Commodity indices are updated weekly. Competitor promotional activity is visible through syndicated data and retailer circulars. Consumer confidence figures are released monthly. The information exists. The architecture to connect it to a pricing decision often does not.
The result is what we call the quarterly review problem: a pricing model that is, at the moment of every decision, already responding to a market that no longer exists. The organizations that have moved beyond this—that have connected live market signals to their pricing intelligence—are outperforming their peers by 6 to 8 gross margin points over 12 months. That gap is not explained by better products or stronger brands. It is explained by faster, better-connected decisions.
What External Market Signals Actually Means
The term can sound abstract. In practice, it refers to a specific set of inputs that most CPG pricing models are not ingesting—and that meaningfully change what the right pricing decision looks like.
Fuel and freight costs are the most direct. When diesel prices move, logistics costs move with them—not just for you, but for every competitor in your category. A fuel price increase does not just affect your cost of goods. It affects your competitor's capacity to hold price, their margin buffer, and their willingness to absorb a price decrease or match an increase. Modeling fuel price as an input to your own cost structure is table stakes. Modeling it as a signal about the competitive pricing ceiling is where the advantage lives.
Commodity input indices operate on a similar logic. Agricultural commodity prices, packaging material costs, and energy inputs drive cost-of-goods volatility across the category simultaneously. When input costs spike across an industry, the first brand to move on price often captures the window before competitors respond. Most quarterly review cycles miss it entirely.
Consumer sentiment and trade-down risk are leading indicators of price sensitivity that change faster than most pricing cycles can accommodate. A meaningful drop in consumer confidence can increase elasticity within a single quarter—meaning a price increase that was defensible at the start of a planning cycle may face measurable resistance by the time it executes. Brands that monitor these signals in real time can adjust their approach before executing, rather than discovering the problem in the next velocity report.
Competitor promotional velocity is perhaps the most underappreciated input. When a major competitor runs a deep promotional event in your category, your everyday price elasticity temporarily increases—consumers have an alternative. When they go quiet on promotion, your pricing ceiling rises. Most elasticity models are built on aggregate historical data that averages across promotional and non-promotional periods. They cannot distinguish between the two in real time, which means they cannot tell you when the competitive environment has just changed in your favor.
None of these signals require exotic data infrastructure. Most are available through existing syndicated data subscriptions, public economic data, or API-accessible price indices. The challenge is integration—connecting them to the pricing model in a way that the commercial team can act on before the window closes.
The Hidden Cost of Single-SKU Thinking
Even a pricing model that perfectly accounts for external market signals can produce systematically wrong recommendations if it treats each SKU in isolation.
Cross-portfolio cannibalization is one of the most common sources of invisible margin destruction in CPG pricing—and one of the least modeled. Here is how it typically unfolds.
A brand raises the price on its premium SKU. The model says the elasticity supports it. The move executes. Overall category revenue is flat or slightly positive. Six weeks later, the margin report shows that gross margin has declined.
What happened: volume shifted to the mid-tier SKU in the same portfolio. The premium price increase was accurate in isolation. What the model missed was the substitution effect—the degree to which consumers in this category, at this price gap, will trade down within the portfolio rather than absorbing the increase.
Three scenarios appear repeatedly in CPG pricing that look correct on a single-SKU model but fail on the portfolio P&L:
- A premium price increase that is defensible in isolation but accelerates trade-down to a lower-margin SKU in the same family.
- A promotional depth decision on a flagship that pulls volume from a higher-margin adjacency.
- A price-pack architecture change that improves per-unit revenue on the focal SKU but drives purchase behavior toward a configuration that is less profitable for the portfolio as a whole.
Building a cross-SKU elasticity model requires a unified product master—a single version of your SKU hierarchy that is consistent across your ERP, your trade systems, your syndicated data, and your BI layer. Most CPG brands do not have this. Product master fragmentation is the infrastructure problem that sits underneath most cross-SKU modeling failures. It is also one of the first things a governed analytics approach addresses.
The Dashboard That Changes the Conversation
There is a distinction that matters enormously in how CPG pricing decisions actually get made, and it rarely appears in discussions of pricing analytics.
The distinction is between a model that tells you what happened and a model that tells you what to do.
Most pricing analytics in CPG today are descriptive. They produce reports that accurately characterize the past: what the lift factor was, what the elasticity coefficient measured over the last twelve months, what baseline volume looked like before and after a promotional event. This information is valuable. It is not sufficient for a real-time commercial decision.
Predictive pricing intelligence is different. It takes the governed historical data, connects it to live market signals, runs it through a cross-SKU model that accounts for substitution effects and promotional interactions, and produces scenario outputs that the commercial team can act on before the decision is made—not after.
What that looks like in practice: a pricing team that can, in the week before a pricing cycle executes, run a scenario showing the expected volume and margin impact of their proposed move across the full portfolio, stress-tested against different assumptions about competitor response and against the current consumer confidence environment. A scenario that shows them not just what the model predicts, but why—with the market signal inputs and the elasticity coefficients visible and traceable.
This is what a pricing intelligence dashboard built for commercial decision-making surfaces. Not a static report that arrives after the quarter closes. A live, scenario-capable environment where the team can stress-test their assumptions, see the cross-portfolio implications, and walk into the pricing review with a defensible position based on current market conditions.
For the CFO and the board, the output is equally important: not just a recommended price move, but a traceable chain from market signal to model input to commercial recommendation. When a pricing decision needs to be defended—in a retailer conversation, in a board review, in a trade partner negotiation—the ability to show the data and the logic behind the move is the difference between a confident commercial team and one that is relying on a number they cannot fully explain.
The Governance Problem Nobody Talks About
There is a reason most CPG analytics teams have built technically sound elasticity models that the commercial team does not use.
It is not a modeling problem. It is a governance problem.
When the RGM team's pricing model produces a recommendation that is different from the number the CFO's finance model produces, something breaks. The commercial team does not trust the output. They run a parallel spreadsheet. The pricing decision gets made on the spreadsheet—because at least they know where those numbers came from.
This happens because the data feeding the two models is defined differently. "Baseline volume" means one thing in the analytics pipeline and something subtly different in the finance system. The SKU hierarchy in the trade promotion system does not map cleanly to the hierarchy in the POS data. The promotional lift factor is calculated on a different time window than the one the CFO's team uses.
The model is technically correct. The governance of the underlying data is not sufficient for the commercial team to trust the output. And a model that is technically correct but commercially unused is, from a margin improvement standpoint, indistinguishable from a model that was never built.
Governed analytics means something specific in this context. It means a single, consistent definition of every metric that matters—baseline volume, promotional lift, net elasticity, price-pack contribution—across every system that touches those numbers. It means a data pipeline with lineage tracking, so every output can be traced back to its source. It means a product master that is resolved across ERP, trade, and POS data, so cross-SKU modeling produces outputs that match the commercial team's understanding of the portfolio.
This is the infrastructure that makes predictive pricing intelligence commercially usable rather than analytically interesting. It is also the work that most analytics teams underestimate—and that most off-the-shelf BI tools do not address.
What This Looks Like in Practice
CPG companies that have moved to market-factor-integrated pricing intelligence share some common characteristics in how they operate differently.
They run pricing scenarios before the quarter starts, not after. Instead of executing a pricing decision and analyzing its impact in the next period's report, they scenario-model the decision against current market conditions before it executes. The scenario modeling is not a separate workstream—it is integrated into the pricing review meeting itself.
Their pricing and finance teams are operating from the same number. The output of the pricing model matches the number the CFO sees. When there is a discrepancy, it is caught in the governed data layer before it becomes a trust problem at the commercial level.
They can respond to market signal changes on a weekly cadence rather than a quarterly one. When diesel prices move, when a competitor goes quiet on promotion, when consumer confidence shifts—these signals hit the model immediately, not in the next quarterly data pull. The commercial team can decide whether and how to respond before the window closes.
The result, across companies that have made this transition, is a meaningful and durable margin advantage. Not from a single pricing decision, but from a compounding effect of consistently faster, better-connected decisions over time. Companies using AI-governed promotional analytics report 20 to 40 percent improvement in trade ROI within one planning cycle, with gross margin improvement of 6 to 8 points over 12 months.
The Starting Point
For most CPG companies, the path to market-factor-integrated pricing intelligence does not require replacing existing technology. The Snowflake instances, the dbt models, the syndicated data subscriptions—the infrastructure is usually already present.
What is missing is the integration layer: the governed semantic layer that makes the pricing model's outputs consistent with the finance team's numbers; the external signal pipeline that connects live market data to the elasticity model; the commercial interface that makes scenario modeling accessible to the RGM team without requiring an analytics engineer in the room.
The question worth asking before the next pricing cycle: what market signals changed in the last six weeks that your current model does not know about? What is your competitor's cost structure doing right now, and how does that change the ceiling for your next price move? How would your proposed price change look under different scenarios for the next 90 days of consumer sentiment?
If those questions are difficult to answer before the decision executes, that is where the margin opportunity lives.
Get the Framework
We have put together the CPG Pricing Intelligence Playbook—a practical guide covering the seven external market signals most pricing models ignore, a framework for connecting them to your elasticity model, and a scenario planning template you can use in your next pricing cycle.
Download the Pricing Intelligence Playbook
Or, if you would like to see what this looks like on your existing stack:
Book a 20-minute pricing analytics assessment
No prep required. We come with a view of what the gaps look like for your specific portfolio and category.
