What is repricing? How it works and where it stops
Repricing recalculates a selling price from defined signals, then applies it within limits set by your team. A dependable rule verifies the competing offer, full unit cost, and floor price. It can also hold the current price when no action is justified.
The essentials
- Repricing executes a pricing rule. It does not replace pricing strategy or your team’s responsibility for the outcome.
- A dependable cycle verifies the competing offer, full cost, floor price, and channel permission before publishing a price.
- Applying the same schedule across the catalog creates unnecessary reviews: at an electronics seller, about 30% of a catalog of roughly 1,000 products needed high-frequency repricing.
- A bad product match or incomplete cost can make the calculated price wrong before any race to the bottom begins.
At 9 a.m., a competing offer drops on a marketplace. Your price holds, then the Featured Offer moves to another seller. Should you follow immediately, wait for the next sale, or keep the price because the gap remains profitable? Repricing handles that repeated decision without removing the limits your team set.
What is repricing in ecommerce?
Repricing is the regular recalculation of a selling price under predefined rules. A competitor price may trigger it. So can another seller going out of stock, a cost change, inventory position, or demand. The calculated price may remain a recommendation for review, or the system may publish it to an enabled channel.
Repricing is not the whole pricing strategy. Strategy sets positioning, target margin, and priority categories. Price monitoring observes the market. Repricing executes the rule that applies to one SKU in one context.
That distinction prevents a costly shortcut: repricing does not mean becoming the cheapest seller. A rule may match an eligible offer, remain above it, raise the price when a competitor sells out, or hold. On a marketplace, price affects the Featured Offer, also known as the Buy Box. It cannot guarantee it because availability, fulfillment, and seller performance still matter.
| Job | Question | Output |
|---|---|---|
| Price monitoring | What is the market doing? | Observed data |
| Pricing strategy | Where should we compete? | Team decision |
| Repricing | Which price applies now? | Guarded calculation |
How does repricing work from signal to published price?
A useful cycle begins with an observation, not an automatic price cut. The system receives a signal and checks whether the observed offer matches the product and warrants a response. A SKU without a shared barcode may require matching on brand, dimensions, material, images, or trade-specific attributes. Uncertain matches should be routed to human review.
The engine then calculates an acceptable price. It starts with purchase cost and adds referral fees, fulfillment, expected returns, and the chosen margin. That calculation produces the floor price. The pricing rule works above this boundary. A ceiling can also prevent an implausible increase.
The system handles exceptions, requests review where needed, and publishes only to enabled channels. Every change should retain its trigger, previous price, and rule. Without that history, the team sees an outcome it cannot explain.
- 1Signal
Competitor price, stock, cost, or demand
- 2Eligible offer
Same product, relevant seller, available offer
- 3Floor price
Costs, referral fee, fulfillment, returns, margin
- 4Pricing rule
Match, stay above, raise, or hold
- 5Review
Approve, block, limit, or handle an exception
- 6Publish
Send the price to an enabled channel and log it
An uncertain match, a missing cost, or a price below the floor stops the cycle before publication.
When does repricing help?
Repricing is useful when several sellers carry a comparable offer and their stock or price changes often. Marketplace products, items sold through several channels, and categories with frequent Featured Offer changes fit this pattern. A rule can also raise a price after the cheapest eligible competitor sells out.
It is less useful for a private-label product with no direct equivalent. A competitor-only rule has no dependable reference there. Your team may use demand, sales velocity, or margin instead, but that is a different pricing method.
Scope matters as much as speed. At an electronics marketplace seller, about 30% of a catalog of roughly 1,000 SKUs required high-frequency repricing. The other 70% were stable, lightly contested, or slow-moving. This is one client case, not a market average.
A second catalog shows why the rule should change with the situation. Across 23,365 products at a gardening and pet supplies retailer, 45.4% had no identified competitor, 29.1% were already winning, 17.3% sat behind the market, and 8.2% were directly contested.
- No competitor45.4%
- Test value and margin, not an automatic reduction
- Already winning29.1%
- Hold the price while performance remains sound
- Behind the market17.3%
- Check the gap, floor, and non-price advantages
- Directly contested8.2%
- Use a guarded rule and monitor more often
Repricing does not require one frequency or one rule across the entire catalog.
Which pricing rules and limits should you set?
A complete rule names the signal, permitted response, and stop condition. It applies to a catalog segment, never to a vague goal such as stay competitive. An operator should understand it without reading the engine’s code.
| Element | Decision | Example limit | Set by |
|---|---|---|---|
| Reference offer | Is it comparable? | Verified match | Team |
| Floor price | Which costs count? | Minimum margin | Team |
| Movement | Raise, lower, or hold? | Maximum change | Rule |
| Frequency | When to recalculate? | Segment schedule | Team |
| Publishing | Where can it go? | Enabled channel | Team |
| Exception | When should it stop? | Locked price | Team |
Test the floor price on a real SKU. A missing referral fee, a zero return allowance, or an understated fulfillment cost corrupts every calculation that follows. The rule should block the price instead of compensating for absent data.
Frequency should follow competitive intensity. A contested SKU may need close monitoring. A stable range can use a slower schedule. Speeding up the entire catalog creates more decisions to review without improving every price.
Run the first rules in recommendation mode. For each suggested move, have an operator record whether the offer match was valid, whether the floor included every channel cost, and whether the action fit the product’s role. Sample unchanged SKUs as well: a system that correctly holds a price deserves the same scrutiny as one that moves it. Enable publishing only after the team can explain both outcomes from the log.
What are the risks of repricing?
A bad product match is the first risk. The system compares different offers and turns a false signal into a new price. Incomplete cost is another: the floor appears protective while omitting a fee, return, or fulfillment expense. Feedback between rules creates a third risk when sellers repeatedly follow one another downward.
The evidence shows why human control still matters. The European Commission’s 2017 E-commerce Sector Inquiry reported that 53% of surveyed retailers monitored competitor prices online. Of that group, 67% used software. Among monitoring-software users, 8% made fully automatic adjustments. The inquiry describes 2017 practice; it does not measure rule quality or commercial results.
Chen, Mislove, and Wilson tracked the top 20 sellers of 1,641 Amazon best-seller products every 25 minutes across two crawls in 2014 and 2015. They identified 543 likely algorithmic sellers. This older Amazon sample documents the mechanism, not the right schedule for a current catalog.
The four rows use different denominators. They frame the scope of control; they do not combine into one market rate.
Control also protects consistency across channels. A marketplace price cut may conflict with the price on your own store, a scheduled campaign, or a supplier commitment. History, alerts, and a stop control are production functions rather than optional reporting.
How should an ecommerce team evaluate repricing?
Start with a SKU whose costs your team knows. Rebuild the floor price by hand, then compare it with the screen. Introduce a competing offer that is not an exact match. The system should reject it or send it for review. Finally, test a rule that reaches the floor and confirm that no lower price reaches the channel.
Ask to see the log for one change. A margin chart alone cannot replace that record.
| Test | Evidence to expect |
|---|---|
| Trigger | Timestamped signal |
| Qualified offer | Explained product match |
| Calculation | Rule and floor price |
| Publication | Channel and approval |
Review segmentation as well. The system should separate contested SKUs, current winners, products with no competitor, and items outside the scope. At the gardening retailer above, one schedule across 23,365 products would have mixed four incompatible decisions.
Market context cannot prove a tool’s effectiveness, but it shows the operating scale. The U.S. Census Bureau reported $340.2 billion in US retail ecommerce sales for Q2 2026, equal to 17.1% of total retail sales. Ecommerce grew 12.2% year over year, compared with 6.7% for total retail. These are national market figures, not repricing outcomes.
Returns belong in the same test. The National Retail Federation and Happy Returns estimated that 19.3% of online sales would be returned in 2025. A floor built with no return allowance may look precise and still be wrong. The report is an industry estimate, not a promise for any catalog.
Read next: definition, selection, or product?
To compare tool families, cost inputs, and rollout approaches, use the ecommerce pricing software selection guide. To assess Boostmyshop’s commercial capability, rules, and safeguards, see myPricing dynamic repricing software.
Sources and methodology
- European Commission, E-commerce Sector Inquiry, SWD(2017) 154, 2017.
- Chen, Mislove, and Wilson, An Empirical Analysis of Algorithmic Pricing on Amazon Marketplace, ACM, 2016.
- U.S. Census Bureau, Quarterly Retail E-Commerce Sales, 2nd Quarter 2026, August 2026.
- National Retail Federation and Happy Returns, 2025 Retail Returns Landscape, October 2025.
- Catalog data comes from the anonymized client cases linked in the article. The two configurations are not market benchmarks.
- The limits of each source are stated in the article.
Frequently asked questions
Repricing is the regular recalculation of a selling price under defined rules. It may produce a recommendation or publish the price to an enabled channel within limits set by your team.
No. A rule may lower, match, hold, or raise a price. It can also take no action when the competing offer is not comparable or the calculated price falls below the floor.
It adjusts price, one input to the Featured Offer decision. It cannot guarantee a win because stock, fulfillment, offer eligibility, and seller performance still matter.
The floor is the lowest price a rule may publish. Your team builds it from the full cost of the sale and its minimum acceptable margin.
No. Scope depends on competition, velocity, and comparability. In one client case of roughly 1,000 SKUs, only 30% needed high-frequency repricing. That result is not a market average.
Continue exploring
Stock availability, picking accuracy, and shipping times affect marketplace sales. Jonathan Coloma, COO of Repiauto, explains how his team went from 300 to 1,000 orders a day in two years. He describes the limits of paper-based work, ERPs built for different needs, a go-live estimate of fifteen days, and changes to his daily routine.
To hold up under peak volume, split the order queue into waves based on real operating constraints, then decide where barcode checks belong. At the packing bench, the operator verifies the parcel. Carrier rules select the service and print the label, so the next order can start without another decision.
Industrializing your purchasing means ordering on rules, not guesswork: at the right time (the reorder point), in the right quantity (your coverage period), from the right supplier, with a purchase order validated in one click. Available on replay.
See it in your own operations
Book a 20-minute call. We'll map your workflow and point out where it breaks. No slides.
Free 20-min call · No commitment