TL;DR
Five rules, in the order you should build them: a profit guard, a Buy Box chaser, a competitor filter, a stock dial, and a ceiling hunt. Build the guard first, because every other rule runs inside it. Build the ceiling hunt last, because it’s the one that makes you money rather than saving it. Each rule below is written as a condition and an action you can copy straight into a rule builder, with the thresholds to start from and an honest note on when not to use it.
Most articles about repricing rules describe rules. This one gives you the rules.
Each is written the same way: the condition, the action, what to set it to, which SKUs it belongs on, and when it’ll hurt you. Build them in order. Rule 1 is the foundation and rule 5 is the payoff, and skipping to 5 is how people lose money.
New to this? Start with what repricing is and the setup walkthrough, then come back with your costs in hand.
Rule 1: The Profit Guard
Build this first. Everything else runs inside it.
The rule
IF any rule would price below (landed cost + referral fee + fulfilment cost + returns provision + ad allocation + target margin) THEN hold at that floor. Do not price below it. Ever.
Set it to
Your floor is a calculation, not a number you type. Work out the all-in cost per unit, then add your target margin. As a starting point: 15% to 20% on high-velocity, low-margin lines; 25% to 30% on mid-velocity; more on slow or specialist stock where you’re not making it back on volume.
The critical part is that the floor recalculates when the inputs move. A typed £24.99 doesn’t know Amazon’s fees changed in January. Our net margin guide covers what belongs in the calculation, and minimum price floors enforce it.
Use it on: everything. No exceptions.
When it hurts: it doesn’t. If a floor is losing you sales, the floor isn’t wrong, the product is. That’s a sourcing conversation.
Watch: review it whenever Amazon announces a fee change. Fee rises silently turn a good floor into a bad one, and nothing on your dashboard tells you.
Rule 2: The Buy Box Chaser
Now you can chase, because the guard means you can’t chase off a cliff.
The rule
IF a competitor holds the Buy Box and their price is above my floor THEN price just under the Buy Box holder, not the lowest seller on the listing UNLESS that would breach my floor, in which case hold at floor
Set it to
Undercut the box holder by the smallest increment that actually wins: 1% or a few pence. Not 5%. You’re trying to rotate into the box, not buy it outright.
The important word is holder. Pricing against the cheapest seller on the listing is the most common configuration error there is, because that seller often isn’t winning anything, and you’re burning margin racing someone who isn’t in the race. The Buy Box algorithm guide covers why the cheapest offer frequently loses.
Use it on: your fastest-moving 20%, where extra rotation is worth the margin you give up.
When it hurts: when a dominant seller holds the box on metrics you can’t match, or when Amazon itself is on the listing. You’ll undercut all the way to your floor and still lose. Recognise it and stop.
Watch: Buy Box percentage and profit per unit together. If share climbs and profit per unit falls, this rule is too aggressive.
Rule 3: The Competitor Filter
Twenty minutes to set up. Saves more margin than rules 2, 4 and 5 combined.
The rule
IF a competitor matches any of: feedback below your threshold, a different item condition, no Buy Box eligibility, or a known dumper THEN ignore them entirely. Do not price against them.
Set it to
Start by excluding: sellers under roughly 90% feedback, anyone listing used or refurbished when you’re selling new, anyone not currently Buy Box eligible, and any seller you’ve watched liquidate stock below cost before.
Then add to the list as you spot them. This rule gets better with age.
Use it on: every contested listing. It’s the difference between a repricer and a liability.
When it hurts: if you over-filter, you may ignore someone who is genuinely taking your rotation. Review the exclusion list quarterly and check your Buy Box percentage hasn’t quietly slipped on those ASINs.
Why it matters: nearly every price war starts with somebody matching a competitor they should have ignored. Our price war guide covers how the spiral starts and how to step out.
Rule 4: The Stock Dial
Price as a velocity control, not just a competitive response.
The rule
IF projected days of cover < my restock lead time THEN raise price to slow velocity IF projected days of cover > 60 (or a storage fee window is approaching) THEN lower price to clear, down to the floor
Set it to
Tiered, and tuned to your lead time rather than a generic number. A rough starting ladder:
| Days of cover | Action |
| Under 7 | Raise 10% to 15%, protect availability |
| 8 to 30 | Standard competitive rules apply |
| 31 to 60 | Hold, watch |
| Over 60 | Reduce 5% to 10% to accelerate |
| Over 90 | Clear aggressively, respecting the floor |
The key input is your lead time, not the calendar. A private label seller with a 90-day restock should start raising prices far earlier than a wholesaler who can reorder on Thursday. Set the “under” trigger to your actual lead time, not to seven days because an article said so.
Use it on: seasonal stock, long lead times, anything heading toward a storage fee. FBA inventory management and the seasonal repricing guide cover the wider levers.
When it hurts: on fast-restock lines with steady demand, this adds noise for nothing. If you can reorder in a week, you don’t need to ration.
The real prize: avoiding the stockout. Going to zero costs you the ranking, and rebuilding that takes longer than the stockout lasted.
Rule 5: The Ceiling Hunt
The rule almost nobody configures, and the only one on this list that makes money rather than protecting it.
The rule
IF I’ve held the Buy Box for more than X hours AND sales velocity is at or above target THEN raise price in small increments until either the box is lost or velocity drops THEN step back to the last price that held both
Set it to
Increments of 1% to 2%, every few hours, with a ceiling cap so it can’t run away. Give it a fortnight before judging it.
You’re looking for the price where you still hold the box and still convert. It’s almost always higher than where you set it, because you set it while nervous.
Use it on: anything you hold the box on comfortably. Private label especially, where there’s often nobody to lose it to.
When it hurts: on knife-edge contested listings where you’re barely holding rotation. Don’t hunt a ceiling you’re already scraping.
Why it matters: if every rule you’ve built moves price in one direction, you’ve automated your losses and left your gains manual. That’s the single most common shape of a badly configured repricer.
Build order, and why
Layer them. Don’t switch all five on at once.
| Order | Rule | Why here |
| 1 | Profit Guard | Everything else runs inside it |
| 2 | Competitor Filter | Cheap, fast, prevents the worst outcome |
| 3 | Buy Box Chaser | Now safe to be aggressive |
| 4 | Stock Dial | Add once the basics behave |
| 5 | Ceiling Hunt | The payoff, once you trust the rest |
Note the filter sits at 2, not 3. The source of most repricing damage is matching someone you shouldn’t, so it’s worth twenty minutes before you switch on anything aggressive.
Test on 20 to 50 representative SKUs first, and give each change a fortnight. Changing three rules at once and watching the number move teaches you nothing about which one worked. The rule-based vs AI breakdown covers which segments to hand to the algorithm instead of writing rules by hand, and repricing strategies covers the wider patterns.
Worth testing later: time-of-day rules
Not in the main five, deliberately.
Demand does vary by hour and by day, and pricing slightly higher during peak windows is a real tactic. But it’s a refinement, not a foundation, and sellers reach for it before their floors are right, which is like tuning a stereo in a car with no wheels.
If you want to test it: try 3% to 5% firmer on your peak days, 2% to 3% softer on your slowest, and measure for a month before doing anything more elaborate. Get rules 1 to 5 working first.
Measuring whether the rules work
Weekly, four numbers:
- Buy Box percentage, alongside profit per unit. Never alone. The win rate tracking guide covers reading it properly.
- Profit per unit. The one that decides whether any of this worked.
- Price change frequency. If it’s thrashing, two rules are fighting.
- Days of cover trend. Tells you if rule 4 is doing its job.
Analytics and reporting pairs win rate against profit deliberately, because either on its own will mislead you. If configuring this across thousands of SKUs is the job that never gets done, managed setup hands it to someone who does it daily, and Repricer’s rule and AI engine runs both approaches side by side.
FAQ
What repricing rule should I set up first? A profit guard, before anything else. It’s a floor calculated from landed cost plus every fee plus your target margin, and it’s the boundary every other rule operates inside. Sellers who start with Buy Box rules and add profit protection later usually discover they’ve been winning unprofitable sales for a month.
Should I price against the lowest seller or the Buy Box holder? The Buy Box holder, almost always. The cheapest seller on a listing often isn’t winning anything, so pricing against them burns margin in a race nobody’s running. Amazon weighs fulfilment and seller metrics alongside price, which is why an offer that isn’t the cheapest routinely holds the Featured Offer.
How do I stop my repricing rules starting a price war? Two things: a floor calculated from real costs that the rules cannot cross, and a competitor filter that excludes sellers you shouldn’t be matching. Most spirals begin when someone’s rules respond to a seller with poor metrics or a liquidation sale, and everyone else’s rules follow. Filtering takes twenty minutes and prevents most of it.
Can I use different repricing rules for different products? Yes, and you should. Contested wholesale lines, slow movers with storage exposure, and uncontested private label SKUs all want different logic. Segment by margin, competition and velocity, then apply rules per segment. One rule across the whole catalogue is the fastest setup and the weakest result.
How do I know if my rules are too aggressive? Read Buy Box percentage and profit per unit together. Share climbing while profit per unit falls means you’re buying rotation with margin, which isn’t a win. Strong margins with share drifting down means you’re leaving rotation on the table. Neither number tells you anything on its own.
Should FBA and FBM inventory use different rules? Yes. FBA offers carry Prime eligibility and delivery advantages, so they often hold the Buy Box without being cheapest, meaning FBA rules can afford to be less aggressive. FBM usually has to work harder on price, but still inside a floor that accounts for your own fulfilment costs. Our FBM repricing strategies guide covers competing without the badge.
Where to start
Build rule 1 today, on ten SKUs. Just the guard.
Work out the true all-in cost on those ten, set the floor from that number, and compare it to whatever floor you’re currently running. For most sellers the gap is uncomfortable, and it explains a lot about last quarter.
Then add the filter. Then the rest.
If you want the whole thing configured against your own catalogue rather than built rule by rule:
