Cross-elasticity, cannibalization, halo effect
.avif)
A price change results in both cannibalization (of substitute products) and the halo effect (on complementary products). Ignoring this mechanism is equivalent to measuring only half of the actual impact of a pricing decision.
On average, 22% of the increase in sales of a product on promotion comes from a simple shift from other SKUs in the same product line.
Lowering the price of a product doesn’t just affect its own sales—it shifts demand toward (or away from) other items in your catalog. This shift has a name—cross-elasticity—and two sides to it: cannibalization, which destroys value internally, andthe halo effect, which creates it. Ignoring this mechanism is like setting a price while assuming it exists in a vacuum on the shelf.
Three effects, a single phenomenon
Classical price elasticity answers a simple question: If I change the price of product A, how do sales of product A change?Cross-price elasticity poses a more troubling question: If I change the price of product A, how do sales of product B change?
The answer has a sign, and that sign changes everything:
- Positive sign → substitutes. Two plain yogurts from different brands. Lowering the price of one automatically causes sales of the other to drop: this is called cannibalization.
- Negative sign → complementary. A coffee machine and its capsules. Lowering the price of the machine can boost capsule sales: this isthe halo effect.
The same price change therefore simultaneously produces cannibalization on certain references and a halo effect on others. Managing a price without considering these two effects means measuring only half of the real impact of a decision.
Cannibalization: When One Price Eats Into Another's Sales
Cannibalization is not a rare occurrence; it is the default behavior of a poorly isolated promotion. A study published in an academic journal (ScienceDirect) on the effects of promotions in mass retail shows that, on average, 22% of the increase in sales generated by a promoted product comes from other versions of the same product—not from new customers, but simply from a shift in volume within the same product portfolio.
On average, the increase in sales of a product on promotion comes from other sizes of the same product—not from new sales, but from a shift in volume—and does not create new value (ScienceDirect study on retail promotions).
In categories with high substitution rates (yogurt, laundry detergent, bottled water), the phenomenon goes even further. According to revenue growth management experts, a cannibalization rate of 15 to 35 percent is common in fast-moving consumer goods, and when it exceeds 50 percent, promotions merely shift volume among the company’s own products—at a loss—without generating a single euro in new demand.
The catch: this figure never appears in a standard sales report. A 12% increase in sales of the promoted product looks like a success—until you look at the two comparable products, which each saw an 8% decline during the same period.
The Halo Effect: When a Price Drop Benefits Others
The halo effect is the positive counterpart to cannibalization. It occurs between complementary products, where the purchase of one makes the other more likely or more necessary:
- A printer and its ink cartridges.
- A barbecue and charcoal.
- A capsule coffee machine and the capsules themselves.
Lowering the loss-leader price (such as a printer or a machine) can generate more margin on the complementary product (cartridges or capsules) than the price reduction itself cost. This is one of the rare cases where aggressive pricing is justified by data, rather than just marketing intuition.
Find the complete definition ofthe halo effect in our pricing glossary.
How to Read a Cross-Price Elasticity Coefficient
The reading principle is simple once you have the sign and the order of magnitude:
- High, positive coefficient: strong substitution. A price drop for Product A will erode a significant portion of Product B's sales; this should be treated as a trade-off, not as a net gain.
- Positive and low coefficient: moderate substitution. The two references coexist to a large extent; the cross-impact remains manageable.
- Negative coefficient: complementarity. This is a signal to use when developing offers or setting an introductory price, rather than one that needs to be corrected.
- Coefficient close to zero: the two products lead virtually independent lives—a rare occurrence in reality, except for categories that are very far apart in the catalog.
The most common mistake isn't misreading a coefficient—it's failing to calculate it at all, and treating each SKU as if it were the only one in the catalog. See also our article on how to interpret price elasticity.
How to spot these effects in your own catalog
Map high-risk pairs
Identify references that share an aisle, a use case, or a customer need (such as similar yogurts, or machine + consumable) even before looking at the data.
Analyze historical cross-promotional data
When reviewing past promotions, look at what happened with similar products at the same time, not just the promoted product.
Isolate the true signal from noise
Neutralize seasonality, out-of-stock events, and weather effects before concluding that a sales movement stems from a cross-elasticity effect.
Document continuously, don't just calculate once
Product relationships evolve alongside the assortment. A coefficient measured a year ago may no longer reflect the reality of the current catalog.
A real-world example: one of our clients in the food industry
For one of our clients in the food industry, the platform we implemented combines sales forecasts, simple and cross-price elasticities, and cannibalization detection into a single pricing recommendation system, rather than treating each SKU in isolation. The stated goal was not merely to set the right price, but to achieve a threefold objective simultaneously: margin, competitiveness, and price image. A price that increases margin but cannibalizes two similar SKUs at a loss does not meet this objective, even if it appears to do so in an isolated report.
Before validating a price reduction
- Have I identified the substitutable references that could lose sales?
- Have I identified the complementary references that could gain sales?
- Is the net impact (direct gain − cannibalization + halo effect) positive, rather than just the direct gain?
- Has this coefficient been recalculated recently, or has it been sitting unused in a spreadsheet for a year?
FAQ
The questions we are most frequently asked before getting started.
Simple price elasticity measures the effect of a product's price on its own sales: if I lower the price of A, how many more units of A will I sell?Cross-price elasticity asks a different question: if I change the price of A, how do sales of product B react?
The sign of the cross-coefficient completely changes how the results are interpreted. A positive sign indicates substitutes (two plain yogurts from different brands): lowering the price of one automatically reduces sales of the other—this is cannibalization. A negative sign indicates complements (a coffee machine and its capsules): this is the halo effect.
Setting a price based solely on simple price elasticity means measuring only half of the decision’s actual impact: the increase in sales of the product whose price is being lowered may be partially—or even largely—offset by a loss on a similar product. We explain the definition in detail in our article , “What Is Price Elasticity?”
For a pricing team, ignoring cross-price elasticity is equivalent to approving price cuts that appear profitable when viewed in isolation, but are no longer profitable once the effect on the rest of the product lineup is taken into account.
The sign of the cross-elasticity coefficient provides a clear indication: if it is positive, the two products are substitutes, with a risk of cannibalization; if it is negative, they are complements, with the potential for a halo effect. We explain this calculation method in detail in our article on how to calculate price elasticity using data.
The magnitude of the coefficient then refines the interpretation: a high positive coefficient indicates a strong substitution effect, which should be treated as arbitrage rather than a net gain, whereas a low positive coefficient suggests that the two products can coexist quite well.
In practice, the pairs that should be tested first are those that share a shelf category, a use, or a customer need (such as two yogurts in the same segment, or a machine and its consumable) before even analyzing any sales figures.
A coefficient close to zero, on the other hand, indicates two products that lead nearly independent lives: a rare occurrence in reality, except between categories that are very far apart in the catalog, as this article points out.
No, not necessarily. Using a low-margin product to subsidize a high-margin one may be exactly the result a sales campaign is aiming for; it’s only a problem when this isn’t recognized and skews the assessment of a promotion’s success.
The real risk is statistical blindness: according to a study published in an academic journal (ScienceDirect) on retail promotions, approximately 22% of the increase in sales generated by a promoted product variant comes, on average, from other variants of the same product—a shift in volume, not value creation—which never appears in a standard sales report. We detail the mechanisms that mitigate this risk in our article on the 7 promotional pricing strategies.
In categories with high substitution rates (yogurt, laundry detergent, bottled water), a cannibalization rate of 15 to 35 percent is common, according to revenue growth management experts; once it exceeds 50 percent, promotions merely shift volume among the company’s own products, resulting in a loss.
The right approach, therefore, is not to avoid any cannibalization, but to measure it systematically in order to distinguish between a deliberate margin trade-off and a promotion that masquerades as a success even though it does not generate any new demand.
The halo effect is leveraged by identifying pairs of loss leaders and complementary products—such as a printer and its ink cartridges, a grill and charcoal, or a capsule machine and the capsules themselves—and then setting the price of the loss leader not to maximize its own margin, but to maximize the combined margin of both products. This type of pricing strategy aligns with the approaches we describe in our article on the 5 data-driven retail pricing strategies.
In practical terms, lowering the price of a machine or printer can generate more profit margin on the complementary product than the price reduction itself cost: this is one of the few cases where an aggressive introductory price is justified by the numbers, and not just by marketing intuition.
This approach requires moving away from a reference-by-reference management approach: the introductory price is never evaluated in isolation; rather, it is evaluated in conjunction with all the additional sales it generates over the period in question.
For a retailer, formalizing these "loss leader/complementary product" pairs within its pricing strategy makes it possible to transform a one-off marketing intuition into a reproducible and measurable loss leader pricing policy.
No. Calculatingcross-elasticity across an entire catalog is rarely a wise investment: it’s better to focus first on high-stakes pairs—those that combine high volume with close proximity in terms of usage or shelf location. We discuss this level of detail in our article onprice elasticity by product, store, and cluster.
The method recommended in this article follows four steps: identify at-risk pairs before even looking at the numbers, analyze cross-referenced promotion histories for neighboring SKUs, isolate the specific effect of noise (seasonality, stockouts, weather), and then document the result rather than calculating it just once.
This last point is often overlooked: the relationships between products change as the product lineup evolves, and a coefficient measured a year ago may no longer reflect the current reality of the catalog, which is why it’s important to recalculate regularly rather than letting a figure sit idle in a spreadsheet.
This targeted approach delivers the bulk of the value for a reasonable amount of effort, without tying up a pricing team in an exhaustive calculation—since, in any case, the majority of product pairs have no significant cross-impact.
Advanced Pricing Optimization Software tools can indeed model simple and cross-elasticities based on sales history, but the quality of the results depends directly on the quality and depth of the data available upstream. We list these types of features in our article on the essential features of pricing software.
This is exactly the type of system that one of our clients in the food industry has implemented: a platform that combines sales forecasts, simple and cross-price elasticities, and cannibalization detection within a single pricing recommendation engine, rather than treating each SKU in isolation.
The goal of this system is not only to set an appropriate reference price for each product, but also to simultaneously ensure three outcomes: margin, competitiveness, and price image. A price that increases the margin but cannibalizes two similar products at a loss does not achieve this goal, even if it appears to do so in an isolated report.
For a retailer considering this type of tool, the key consideration is therefore less the sophistication of the model than the quality of the data that feeds it: without its own sufficiently extensive sales history, even the best algorithm will produce unreliable coefficients.
Measure cross-elasticity: how much a product’s sales change when the price of another product changes. If a price reduction on one product takes sales away from another in the same product family, the net gain is less than the apparent gain. Base your decision on the family’s margin, not that of the individual product.
Sources: "Estimating Cannibalizing Effects of Sales Promotions," Journal of Retailing and Consumer Services (ScienceDirect), sciencedirect.com · RGM Academy, "Cannibalization Rate," rgmacademy.app · Business Case Booper (internal client case study, national food retailer).
Lowering a price almost always leads to higher sales; that’s never the issue. The real question is whether the additional volume generates enough margin to offset the margin lost on each unit already sold. The answer depends on two figures that rarely align: the product’s markup rate and its actual price elasticity.
A price change results in both cannibalization (of substitute products) and the halo effect (on complementary products). Ignoring this mechanism is equivalent to measuring only half of the actual impact of a pricing decision.
On average, 22% of the increase in sales of a product on promotion comes from a simple shift from other SKUs in the same product line.
Price elasticity varies significantly from one store to another within the same chain. According to an NBER study, uniform pricing results in a 7 to 9 percent loss in profit compared to store-specific pricing.
The store cluster offers the best balance between analytical depth and operational manageability, while remaining consistent with the product lineup as perceived by the customer.