Repricing is the process of updating product prices — usually automatically and on a schedule or trigger — in response to competitor prices, inventory levels, marketplace position, or other signals. It is the execution arm of competitive pricing: where monitoring observes and dynamic pricing decides, repricing acts.
Rule-based versus algorithmic repricing
- Rule-based repricing — explicit "if-then" logic the seller controls. If a competitor's price drops, set mine €0.50 below, but never under my margin floor or below MAP. Transparent and predictable.
- Algorithmic repricing — machine-learning models that optimise for an objective (winning the buy box, maximising profit) by weighing many variables. More powerful, less transparent, and harder to constrain.
Most sellers start with rule-based repricing because they can reason about and trust the behaviour.
Where repricing is most visible
Amazon's Buy Box is the canonical battleground: with multiple sellers offering the same item, price (alongside fulfilment and seller rating) heavily influences who wins the featured offer, and the Box can change hands many times a day. Repricers exist largely to compete for it. Beyond marketplaces, direct-to-consumer stores reprice to stay competitive on their most-compared SKUs.
The critical role of guardrails
The defining feature of good repricing is what stops it, not what triggers it. Essential constraints:
- Margin floor — never price below a minimum profit.
- MAP floor — never advertise below a manufacturer's Minimum Advertised Price.
- Cost floor — never price below landed cost.
- Frequency limit — avoid thrashing the price dozens of times an hour.
- Ceiling — avoid runaway increases when you are the only seller in stock.
Without floors, two "always be cheapest" repricers can ratchet each other to zero margin — the classic price-war failure mode.
A concrete e-commerce example
A seller competes for the Buy Box on a popular kitchen appliance. Their repricer is set to undercut the current lowest FBA offer by €0.10, with a hard floor at a 12% margin and a cap at MAP. When a rival drops their price, the repricer responds within minutes and reclaims the Box; when the rival hits their own floor and stops, the war ends with the seller still profitable rather than racing to zero.
How it connects to monitoring
Repricing is only as good as its input data. A repricer acting on stale or wrong competitor prices will make confidently wrong moves. This is why repricing sits downstream of competitor-price-monitoring: tools like RivalScraper supply the continuous, accurate rival-price feed that a repricer needs to act on reality rather than yesterday's snapshot.
Common repricing pitfalls
Even with guardrails, repricing goes wrong in predictable ways. Reacting to a temporary competitor stockout (their price vanishes, yours over-corrects) is a frequent one — good repricers ignore out-of-stock rivals rather than treating absence as a zero price. Matching a competitor who is themselves mispriced (a fat-finger error, a currency mix-up) propagates the mistake across the market. And repricing too aggressively on low-margin SKUs can win the sale while losing money on every unit. The defence is the same discipline that governs the whole stack: trustworthy input data, hard floors, and sanity checks before a computed price ever goes live.
Marketplace versus owned-store repricing
The mechanics differ by channel. On marketplaces like Amazon, repricing is a contest for a shared listing (the Buy Box), so the reference point is other sellers of the identical item and the cadence is fast. On an owned direct-to-consumer store, repricing compares comparable products across rival domains, the matching is fuzzier, and the cadence is usually slower and more deliberate. The same engine and guardrails apply, but the signal — exact-match offers versus competitor catalogues — comes from different monitoring strategies.
Repricing in the wider stack
Think of the chain as: scrape (gather prices) → monitor (track and alert) → decide (dynamic-pricing logic and elasticity) → reprice (execute the change). Repricing is the last mile — the point where competitive intelligence finally changes the number a customer sees.