Hourly Repricing Is Table Stakes. Here Is Why You Are Still Losing.

1,000 SKUs. One dashboard. Hourly updates. That is the advertised ceiling of entry-level Amazon repricing tools in 2026. If your catalog sits under that threshold, the tool handles itself. You set a floor, you set a ceiling, you walk away. That is exactly the problem.

The Floor Price Trap

A floor price calculated six months ago is not protecting your margin today. It is reflecting your landed cost from six months ago. Freight rates shift. Supplier invoices change. Your floor stays put. The result: you are repricing competitively against a number that has no relationship to your current cost structure. You are not defending margin. You are automating margin erosion. Pull your top 50 ASINs by revenue. Calculate current landed cost against your active floor price for each one. For most brands running passive repricers, at least one decile of the catalog is already underwater and the dashboard shows green.

Velocity Windows Are Not Hourly Events

Hourly repricing catches price changes. It does not catch demand spikes. A competitor goes out of stock at 11:43 AM. Your repricer checks prices at noon. You won the window at cost, not at the margin the demand signal justified. Top-decile operators are building a second logic layer on top of their repricer. When a competing ASIN drops to zero units available, that is not a pricing event. It is a sell-through opportunity. Your price should move up, not match down. The repricer cannot make that call on its own. You need to define the trigger. Competitor inventory depletion is a ceiling-lift signal. Treat it that way.

NetPPM Is the Metric Your Repricer Ignores

Repricers optimize for win rate. Win rate is the wrong optimization target if you are selling through the SP-API with full FBA cost load. A price that wins the Buy Box at a 4% NetPPM is a worse outcome than a price that holds second position at 18% NetPPM on a high-velocity ASIN. Your repricer does not know your referral fee tier. It does not know your FBA dimensional weight surcharge. It definitely does not know your co-op commitments or your AMS spend against that ASIN. You know those numbers. Build them into your floor logic. NetPPM by ASIN cohort, not average margin across the catalog, is the input that makes a repricer functional rather than decorative.

The Setup Protocol That Actually Works

Start with a cycle count of your landed cost data. Every ASIN, current numbers, not last quarter. Then segment your catalog into three cohorts: margin-healthy, margin-thin, and margin-negative. Margin-negative ASINs should not be in an active repricing rule at all. They need a pricing decision, not a repricing schedule. For margin-healthy ASINs, set your floor at landed cost plus your minimum acceptable NetPPM. Then set a velocity ceiling trigger. If a top-three competitor on that ASIN drops below 10 units available, your ceiling lifts by a defined percentage. You decide the percentage based on historical price elasticity for that ASIN. For margin-thin ASINs, the rule is simpler: hold price, do not chase. Let the competitor win the Buy Box. Protect your unit economics. Chasing a margin-thin ASIN down to the floor to win win rate is volume for its own sake.

Three Questions to Pressure-Test Your Repricing Setup

First: when did you last reconcile your floor prices against a current landed cost calculation for each ASIN individually? If the answer is a quarter ago or longer, your floors are fiction. Second: does your repricing logic treat a competitor stockout as a signal to drop price, hold price, or raise ceiling? If you do not have an explicit rule, the repricer defaults to matching down. Third: can you name the three ASINs in your catalog where winning the Buy Box is actually diluting your NetPPM? If you cannot name them in 60 seconds, your catalog cohort analysis is overdue. Pull those numbers before your next repricing rule change. Then change the rule.