Amazon repricing software: how automated pricing actually works.

The Buy Box changes hands constantly, often several times an hour on competitive ASINs, and price is the single largest factor Amazon's algorithm weighs. A seller checking prices manually twice a day is competing against software that checks every few minutes. That gap is the entire reason repricing tools exist.

The two repricing models sellers actually choose between

Every repricer on the market falls into one of two categories, and the difference matters more than any individual feature list.

Rule based repricers follow a fixed instruction: beat the lowest price by a set amount, match the Buy Box price, or stay a fixed percentage above your floor. Tools like RepricerExpress and BQool built their early reputation on this model. It's simple to set up and easy to audit, but it reacts to competitors rather than to your own margin or sell through rate.

Algorithmic repricers weigh more than the lowest price: your sales velocity, current margin, stock level, and how long you've held or lost the Buy Box. Seller Snap and Informed.co are built around this approach. They ask for more setup work up front, in exchange for pricing decisions that account for whether winning the Buy Box at a given price is actually profitable.

What the software is checking before every price change

1. Your minimum and maximum price bounds. Every reprice happens inside a floor and a ceiling you set per SKU or per rule group.
2. The competing offers on that listing. Same condition, same fulfillment method (FBA versus merchant fulfilled), filtered so you're not repricing against a listing you can't actually match.
3. Buy Box eligibility signals. Price alone doesn't win the Buy Box. Account health, fulfillment method, and shipping time all factor in, which is why two sellers at the same price can get different results.
4. How often to check. Most tools poll every few minutes; some offer near real time repricing for high velocity ASINs.

Where a repricer quietly becomes an account risk

The floor price is the setting that causes the most damage when it's wrong. A blank or misconfigured minimum can let the software chase a competitor down toward the cost of goods, or in rare misconfigurations, toward a fraction of a cent. Selling below cost for an extended run doesn't just hurt margin. Sudden, sharp price swings across a catalog are one of the patterns Amazon's Fair Pricing Policy is built to catch, and a fair pricing violation shows up on the account exactly like any other policy strike.

The second common failure is a stale competitor set: repricing against an offer that's actually out of stock, a different condition, or a different fulfillment channel. The result is a price that technically follows the rule but doesn't reflect the real market, which erodes margin without winning anything back in return.

Before you turn one on

Run the repricer on a small, low risk slice of the catalog first, not the whole account on day one. Check the results after 48 hours: is the price actually landing above your floor most of the time, or sitting at the floor constantly. A tool sitting at its floor isn't optimizing, it's failing quietly, and it's worth catching that in a test batch of twenty SKUs rather than across a thousand.

If a repricing misfire has already triggered a fair pricing warning, a high pricing notice, or a broader account suspension, treat it the same way you'd treat any other suspension: pull your repricer's price history logs and your own settings before you write a response. A Plan of Action that names the specific configuration error and shows it's been corrected reads very differently to Amazon's review team than one that just promises to "monitor pricing more closely."