Price elasticity of demand — the ratio of percentage change in volume to percentage change in price — is one of those concepts that gets invoked constantly in pricing discussions and measured rigorously almost nowhere. The standard response when someone asks a brand's pricing team what their price elasticity is: a confident-sounding number pulled from industry benchmarks or intuition, with no actual measurement behind it.
This matters because elasticity estimates are the foundation of pricing decisions. When you're evaluating whether to raise prices by 8%, the relevant question is: how much volume will we lose, and does the margin impact of the volume loss outweigh the margin gain from the price increase? Without an elasticity estimate, that question can't be answered quantitatively. You're guessing.
The good news is that practical elasticity measurement doesn't require econometrics expertise. It requires good data hygiene, methodological care, and the discipline to run structured price tests rather than waiting for observational data to magically produce clean estimates.
The elasticity ranges you should expect in consumer goods categories
Before discussing measurement, some calibration on what elasticity numbers typically look like in consumer goods contexts. A price elasticity of -1.0 means a 10% price increase produces a 10% volume decline (revenue-neutral). More negative values mean higher price sensitivity; less negative values mean lower sensitivity.
Broad patterns from published consumer goods pricing research suggest:
- Commodity or near-commodity consumables (paper products, cleaning supplies, basic personal care): elasticity often in the -1.5 to -2.5 range — relatively high price sensitivity
- Branded consumer packaged goods in established categories (snacks, beverages, household goods): typically -0.8 to -1.5 range — moderate sensitivity, with brand loyalty dampening the response to price changes
- Premium or specialty products with strong brand differentiation (premium beauty, specialty food, premium home goods): often -0.3 to -0.8 — lower price sensitivity, consumers are buying the brand as much as the product
- Luxury goods and high-trust health products: sometimes approaching price-inelastic territory, particularly for repeat buyers
These are ranges, not targets. The actual elasticity for any specific SKU in your catalog depends on your specific competitive landscape, your brand strength, and the degree to which your product is differentiated from alternatives. The only way to know your actual elasticity is to measure it.
The simplest approach: analyzing historical price changes
If you've changed prices on specific SKUs in the past — for any reason: cost increases passed through, competitive response, promotional pricing — you have a natural dataset for elasticity estimation. The approach:
- Identify price change events in your history: periods where a specific SKU's price changed by more than 5% and the change held for at least 3-4 weeks
- For each price change event, extract weekly unit volume for the 4 weeks before and 4 weeks after the change
- Calculate the percentage change in price and the percentage change in volume (using pre-change averages as the baseline)
- Divide: % volume change / % price change = your elasticity estimate for that event
- Repeat across multiple events and SKUs; average the estimates, weighting by volume significance
The caveats with this approach are real and important to acknowledge explicitly:
Omitted variable problem: Price changes often coincide with other factors — seasonal shifts, competitive activity, promotional changes — that also affect volume. A price increase in October that's followed by a volume decline may be partly reflecting the natural post-summer slowdown in your category rather than a price response. Controlling for these factors requires either careful timing of your analysis (isolating periods without confounding events) or more sophisticated statistical controls.
Small sample problem: With four weeks pre/post, you have limited statistical power. The estimate is directionally useful but has wide uncertainty bounds. Multiple events across multiple SKUs narrow this uncertainty — single-event estimates should be treated as rough signals, not precise measurements.
Despite these limitations, analyzing historical price changes is almost always worth doing before running controlled experiments, because it gives you a reasonable prior estimate and often reveals that elasticity varies substantially across your SKU portfolio — some SKUs very sensitive, others barely responding to price changes at all. That variation is actionable even before you have precise estimates.
Controlled price testing: the more reliable path
The gold standard for elasticity measurement is a controlled price test: holding price constant for a control group while varying price for a test group, then comparing volume outcomes. In a pure DTC environment, this is relatively straightforward — you can A/B test prices on your website using traffic splitting, with proper controls for customer segment and timing.
In a marketplace environment (Amazon, Walmart), true A/B price testing is more complex because you can't directly control which consumers see which price. The practical alternative is sequential testing: run price A for 4-6 weeks, then switch to price B for 4-6 weeks, controlling for seasonal effects. Sequential testing introduces more noise than simultaneous A/B, but it's actionable with careful methodology.
For sequential price testing to produce usable elasticity estimates:
- Test periods should be the same relative position in the seasonal calendar — comparing week 8 of Q3 to week 8 of Q3 the prior year is better than comparing Q3 to Q4 in the same year
- Price changes should be large enough to produce measurable volume effects — changes smaller than 8-10% often produce volume responses too small to distinguish from noise
- Test periods should be isolated from major competitive events, promotional activity, and supply disruptions
- Results should be looked at across 3-4 weeks minimum; single-week volume can be noisy enough to produce misleading signals
SKU-level elasticity heterogeneity: why averages mislead
One consistent finding across brands that do rigorous elasticity measurement: elasticity varies enormously across SKUs in the same brand portfolio. A premium product line within a brand may have elasticity of -0.4; the entry-level product in the same brand may be at -1.8. These are different businesses operating in the same catalog, and pricing them with the same strategy — same promotional depth, same response to competitive moves — is leaving margin on the table at the premium end while potentially causing unnecessary volume loss at the entry end.
This is why catalog-level elasticity estimates are usually insufficient for pricing decision-making. The actionable insight is the SKU-level distribution: which specific products are price-inelastic (where modest price increases capture margin without significant volume loss), which are elastic (where price reduction drives meaningful volume), and which cluster in the moderate range where competitive positioning is the dominant factor.
We're not saying you need a precise elasticity estimate for every SKU in your catalog. We are saying that having a rough, directional classification of your catalog into high/moderate/low elasticity tiers changes pricing decision quality significantly. It changes which SKUs you protect against competitive price pressure and which you allow to drift. It changes your promotional strategy — deep promotions on low-elasticity SKUs are low-return; the same promotional investment on high-elasticity SKUs drives substantially more volume.
Communicating elasticity estimates to non-technical stakeholders
Elasticity estimates are only useful if they influence decisions, which means they need to be communicated in terms that resonate with people who aren't thinking in elasticity coefficients.
The translation that works best: frame elasticity estimates as volume impact scenarios. Instead of "SKU 14's estimated elasticity is -1.3," say: "At the current price, if we raise SKU 14 by 10%, we expect to lose roughly 13% of unit volume — that's approximately 400 units per month at current run rate. At the higher price, gross margin per unit increases by $2.80. The math: we lose 400 units × current contribution of $4.60 = $1,840/month in volume contribution, and we gain 1,600 remaining units × $2.80 margin improvement = $4,480/month in margin improvement. Net: the price increase is expected to improve contribution by approximately $2,640/month, assuming the elasticity estimate holds."
That framing makes the decision concrete. Stakeholders who can't engage with an elasticity coefficient can absolutely engage with "we expect to lose X units but make Y more dollars in margin — here's the scenario analysis." That's where elasticity measurement actually earns its return.