Pricing Elasticity and Competitive Context
Elasticity models trained on your own history answer a different question than the one you're asking. Why competitive context belongs in the model — and how to put it there honestly.
Pricing 3 min read
The Role of Price Monitoring in Competitive Research
A competitor price change is a message — but most teams decode it with incomplete context. A working framework for turning price observations into evidence you can defend in a pricing review.
Somewhere in your market, right now, a competitor has changed a price. By the time the change reaches your pricing analyst — via a screenshot, a sales rep, or a customer asking for a match — the observation has already lost most of its evidentiary value. What matters is not the number; it's the context: since when, across which SKUs, at what availability, and consistent with what pattern.
Why screenshots don't scale
Manual spot-checks produce three systematic errors. Selection bias — you check what you remember to check, usually where you recently lost a deal. Timing blindness — a single observation can't distinguish a weekend flash sale from a repositioning. Matching drift — the "equivalent product" being compared quietly stops being equivalent as catalogs evolve.
Continuous price monitoring exists to remove those three errors. It observes a defined competitor set on a fixed cadence, stores every observation with its context, and applies matching rules that are versioned and auditable. The result isn't just fresher numbers — it's a different epistemic category: observations you can defend in a pricing review.
The matching problem is the real problem
Ask anyone who has run a monitoring program: the scraping is the easy part. Matching is where integrity lives. A comparison between your product and the wrong competitor SKU doesn't just mislead — it actively corrupts downstream models that trust your input.
- Start with deterministic anchors: manufacturer part numbers, GTINs, model codes.
- Layer attribute similarity on top — brand, spec, variant, packaging — with explicit confidence scores.
- Route low-confidence matches to human audit, and log the decision so the same question never gets answered twice.
- Version the rules. When a match improves, you want last quarter's comparison rebuilt under the *old* rules for honest trend lines.
| Observation type | Likely meaning | Recommended response window |
|---|---|---|
| Single-SKU dip, 1 retailer, promo flag | Tactical clearance or retailer promotion | Monitor; no action unless pattern repeats |
| Broad dip across retailers, sustained 2+ weeks | Possible list-price repositioning | Verify cost structure; schedule pricing review |
| Price rise with restock after absence | Supply-driven correction | Update competitive indices; watch elasticity |
| Strikethrough framing without net change | Perceived-value play | Note in positioning report; no price action |
Availability is part of price
A price on an out-of-stock listing is not a competitive price — but it is information. Availability patterns reveal supply constraints, channel strategy and demand shifts. Tracking stock states alongside price turns a flat price feed into a much richer signal, which is why our monitoring treats availability as a first-class field, not an annotation.
Distribution of observed price-change magnitudes by competitor (last 90 days)
From feed to policy
The endpoint of price monitoring is not a dashboard; it's policy: rules your team has pre-committed to. If competitor X sustains an undercut beyond N% for M days, the response is a defined review — not an improvisation. Programs with pre-agreed thresholds react in days and argue less, because the argument happened before the data arrived.
That's the quiet promise of monitoring done well: fewer exciting mornings, better-priced quarters.
Related reading
Continue here
Yusuke Tanaka
The observation-type table is going straight into our pricing playbook. The response windows are realistic rather than panic-driven.
Samuel Osei
Pre-committed thresholds really do reduce the arguing. We call ours ‘fireside rules’ — agreed calmly, applied during the fire.
3 comments
Versioned matching rules for honest trend lines — we learned this the hard way after silently ‘improving’ matches mid-quarter and breaking every index.