You have been paying for a repricing tool for a year. The dashboard shows healthy Featured Offer share and a reassuring count of price changes made. You still cannot answer the only question that matters: is this thing making me more money than not having it?
The reason you cannot answer it is structural. Repricing tools report on their own activity, not on your profit. Price changes made is a measure of effort. Featured Offer share is a measure of position. Neither is a measure of outcome, and both can improve while your contribution per unit declines.
Here is how to get the actual answer.
Contents
Why the Obvious Measurement Does Not Work
The instinct is to compare before and after: profit in the months before automation versus the months since. It fails for three reasons.
Your catalogue changed. Different SKUs, different mix, different sourcing. You are comparing two different businesses.
Your costs changed. Amazon’s 2026 US fee changes took effect for most sellers on 15 January 2026, and a 3.5% fuel and logistics surcharge was added to FBA fulfilment fees in the US and Canada from 17 April 2026. Any before-and-after spanning those dates is measuring fee changes as much as pricing decisions.
Your market changed. Competitors arrived and left. Demand moved. The counterfactual — what would have happened without automation — is not observable in historical data.
So the answer has to come from a comparison that runs at the same time, not a comparison across time.
The Measurement That Works: A Held-Out Group
Pick two matched groups of SKUs and run them differently, concurrently.
Building the groups. Take 30 to 40 SKUs where repricing plausibly matters — contested ASINs, competitors who move, meaningful volume. Split them into two groups matched on: units per week, contribution per unit, number of competing offers, and fulfilment method. Matching matters more than group size; two badly matched groups of fifty tell you nothing.
Running the test. Group A stays on automation as configured. Group B goes to a fixed price — your current price, held — or to manual review at a set cadence you can actually sustain. Run for four weeks minimum, six if your sales cycle is slow. Change nothing else: no advertising shifts, no promotions, no listing edits.
Reading the result. Compare, per group, per week:
- units sold
- revenue
- contribution per unit
- total contribution
- Featured Offer share
- days spent at the floor price
Total contribution is the answer. Everything else is explanation.
The Four Outcomes and What Each Means
Automation group wins on total contribution. The tool is earning its keep. Now check whether it earns more than its subscription, which is a separate and smaller question — and check whether the gain is concentrated in a handful of SKUs, because it usually is. That tells you where to apply automation and where it is overhead.
Automation group wins on share, loses on contribution. The most common result, and the one sellers least expect. You are buying position with margin. The tool is working; the configuration is wrong — almost always a floor built on gross margin rather than contribution, or a reference set to the lowest offer on the listing rather than the Featured Offer price.
This outcome is good news. It means the problem is fixable without changing tools.
Groups are indistinguishable. Your market does not move fast enough for automation to add value on these SKUs. Worth knowing before another renewal. Check whether you selected genuinely contested SKUs for the test — if you did, the honest conclusion is that this is not where your money should go.
Automation group loses on both. Something is misconfigured badly: a reference pulling you toward an outlier, a floor that lets you undercut profitably-held positions, or rules fighting each other across overlapping SKU groups. Diagnose rather than cancel.
The Number You Should have been Tracking All Along
One metric predicts most of this and almost nobody records it: days spent at the minimum price, per SKU.
A SKU parked at its floor is a SKU where automation has run out of room and is selling at the lowest price you authorised. If that number is high and your floors are built on gross margin, you have found your profit leak without needing a test at all.
Add it to your weekly review. A rise in floor-days across the catalogue is an early signal of either a cost change you have not absorbed or a competitor who has gone aggressive.
Costing the Tool Honestly
Once you have a contribution difference, compare it against the real cost of the tool, which is larger than the invoice.
- Subscription, including volume-based tiers. Bills that scale with sales are easy to stop noticing.
- Time spent configuring and supervising it. Real hours.
- The cost of its mistakes — any period where a misconfiguration produced losses, which belongs on this side of the ledger rather than being treated as your error.
If the measured contribution gain exceeds that total, keep it and spend your attention on configuration. If it does not, you have a specific, defensible reason to change something, which is a much better position than a vague suspicion at renewal time.
What to Fix Before You Conclude Anything
If your test shows share up and contribution down, four changes usually close the gap, and they are worth making before you judge the tool:
- Rebuild floors on contribution, with current fee inputs including the surcharge. Not gross margin.
- Reference the Featured Offer price, not the lowest offer on the listing.
- Add upward movement, so price recovers when competition thins. A rule that only descends ratchets your margin down permanently.
- Split rules by economics, so thin-margin SKUs are not governed by logic written for healthy ones.
Then re-run the test. A tool that fails the first test and passes the second was never the problem.
The Habit Worth Keeping
Run the held-out comparison once a year, or whenever you change tools, pricing strategy, or cost base. It is the only method that produces an answer rather than an impression, and it takes four weeks of mild inconvenience.
PricingMag’s research team has documented a step-by-step protocol to tell if your repricer is making you money, including how to match the two groups and what each of the four outcomes should lead you to change.
Their wider Amazon repricing coverage on PricingMag covers the configuration fixes behind the most common result.
A dashboard reports activity. A held-out group reports truth. Only one of them is worth a renewal decision.
