Amazon Price Watch for Sellers: Tracking Competitor and Buy Box Price History

A seller’s Amazon price watch is not a shopping tool. A buyer watches one ASIN to catch a dip; a seller watches a competitive set to answer a different question — is this movement worth responding to, and what does responding cost me? Competitor price history comes from three sources with very different guarantees: third-party trackers that poll public pages, Amazon’s own seller-facing data, and your own collection through the Selling Partner API. Data checked 2026-08-11.

The consumer-side question — which tracker to install, how to read a Keepa chart before buying — is a separate page: Amazon price check tools. This one assumes live offers and a margin to defend.

What Seller-Side Price Watching Has to Answer

Buyer-side tracking has one output: buy now or wait. Seller-side watching has four, and they need different data:

• Should I move my price? Needs the Featured Offer price and your own floor, not the lowest offer on the page.
• Is a competitor discounting, or out of stock? Needs the offer count alongside the price — a price jump is usually an offer disappearing.
• Is this a price war or a one-off? Needs frequency of change, not level of price.
• Is my price defensible if Amazon looks at it? Needs the ASIN’s own recent history, because that is what Amazon’s models compare against.

Only the first is about price alone. Buy Box ownership logic — who is eligible, what beyond price decides it — belongs to Amazon Buy Box (Featured Offer); this page stays on the price-history line.

Three Sources of Competitor Price History

Third-party trackers are outside observers: they poll listings on a schedule and store what they saw. That makes them the only practical source of multi-year history — and the reason a flat line can mean “price never moved” or “nobody sampled it.” Keepa is the reference implementation; its paid data tier is widely reported at €29/month, but keepa.com returned a JavaScript shell with no pricing text and a 403 to automated requests on 2026-08-11, so treat that as third-party corroborated, not vendor-confirmed.

Amazon’s own data is authoritative but short-lived. The free Automate Pricing tool keeps what Amazon calls “Automate Pricing’s detailed 30-day history” (checked 2026-08-11) — enough for last month, nothing about seasonality.

Self-collection closes that gap. Amazon’s published Selling Partner API model for Pricing v2022-05-01 exposes two batch operations — getFeaturedOfferExpectedPriceBatch and getCompetitiveSummary. The first returns “the computed listing price at or below which a seller can expect to become the featured offer (before applicable promotions),” abbreviated FOEP, with the note that “Featured offer is not guaranteed because competing offers might change.” Poll it on your own schedule and you own a history nobody can revoke.

The Four Prices That Get Mixed Up in One Chart

Most seller price-watching mistakes are a units error. These are four different numbers:

Segmentation breaks the mental model of “the Buy Box price.” Amazon’s API model defines a SegmentedFeaturedOffer as “a product offer with segment information indicating where it’s featured,” and each segment carries a customerMembership field plus a glanceViewWeightPercentage — “the glance views for this segment as a percentage of total glance views across all segments for the ASIN.” Different customer groups can see different featured offers on the same ASIN at the same moment, and Amazon tells you how much traffic each segment is worth. A tracker plotting one Buy Box line is flattening that.

The listing-price definition is the second trap: comparing your landed cost against a competitor’s listingPrice measures it against a number that excludes shipping and promotions — see coupons vs. deals vs. price discounts for what price history never sees.

What No Price History Can Tell You

Four blind spots survive every source above:

• Offer disappearance looks like a price increase. A jump from $20 to $60 is usually the $20 seller running out. Watch offer count next to price.
• Condition mixing. Amazon returns lowest-priced offers “for the specified item condition and offer type” — a used-good offer undercutting your new offer is not the same customer.
• Promotions are not in the series. Coupons, Subscribe & Save and deal placements land after the listing price.
• Notification aggregation drops the interesting part, covered next.

Setting Watch Frequency and Alerts You Will Actually Read

Frequency is a budget problem, not a preference. Amazon publishes default rate limits per operation in its API model (checked 2026-08-11):

Amazon notes that “selling partners whose business demands require higher throughput may receive higher rate and burst values than those shown here,” so treat these as a floor. Either way, polling does not scale to a full catalogue at high frequency — tier your ASINs.

The alternative to polling is push. Amazon’s Notifications API model lists ANY_OFFER_CHANGED as a subscribable event filter type, carrying an aggregation setting whose only two values are FiveMinutes and TenMinutes. Amazon’s description of what aggregation does is worth reading twice: “if FiveMinutes is the value chosen, and 50 price updates occur for an ASIN within 5 minutes, Amazon will send only two notifications; one for the first event, and then a subsequent notification 5 minutes later with the final end state of the data. The 48 interim events will be dropped.”

That is the whole alert-design problem: aggregation keeps a notification firehose survivable, and erases the evidence of a fast repricing fight. Choose it per ASIN tier, not globally.

Three rules follow:

1. Alert on your position, not their price. “I lost the Featured Offer” is actionable; “a competitor changed price” usually is not.
2. Set thresholds in margin points, not percent of price. A 3% move on a $9 item and on a $400 item are different decisions.
3. Give every alert an owner and a default action, or it becomes noise within a week.

Reading a Price War vs. Ordinary Movement

A price war has a frequency signature, not a level one. Ordinary movement looks like step changes: a competitor drops once, holds for days, drops again at a promotion. A war looks like oscillation — the same two or three sellers leapfrogging within minutes, each move small, the floor ratcheting down over days.

1. Count changes per day before you look at the price. Two repricers pointed at each other produce dozens of changes a day on an ASIN that normally sees one or two. That count is visible in third-party history and in your own collected series, and largely invisible in aggregated notifications — which is why the aggregation choice above matters.

2. Check whether the floor is moving or only the ceiling. In a genuine war the sustained low falls week over week. Peaks moving while the floor holds is stock rotation among offers, not a war.

3. Separate “undercut” from “hijack.” A new seller undercutting an ASIN you own may be an unauthorised offer — the response is a takedown, not a price cut. See Amazon listing hijacker removal; for competitive intelligence beyond price, competitor ad-spy tools.

The exit from a war is a floor set in advance. Repricer speed and floor protection are compared in Best Amazon Repricer 2026.

The Compliance Line Around Automated Repricing

Two published constraints bound what automation can do, and both are about high prices more than low ones.

The Fair Pricing Policy compares you against price history. The Amazon Marketplace Fair Pricing Policy document published on Amazon’s own media CDN (February 2025 revision, a tracked-changes redline, read 2026-08-11) states: “Amazon uses robust statistical models to detect significantly high and/or erroneous prices. These models use objective data, including historical offer prices, prices from other retailers and sales data, to evaluate a fair price that customers can expect to pay.” Listed among the practices it covers is “setting a price on a product or service that is significantly higher than recent prices offered on or off Amazon,” with no numeric threshold published for “significantly.” The enforcement ladder, the Pricing Health dashboard and the off-Amazon competitive price are covered in Amazon price match and Fair Pricing Policy.

The operational reading: a maximum price in your repricer is a compliance control, not a nicety. Amazon’s Automate Pricing page treats both bounds as standard — “manage your margins by setting a minimum price and an optional maximum price” — and its competitive rules “compare against the Featured Offer, lowest price, or lowest external price,” so your own off-Amazon channel can move the target (checked 2026-08-11).

Reference prices are validated separately. Automating a strike-through is a distinct risk from automating a selling price. Amazon “validates List Prices against sales in our store and external competitor prices,” and says an inflated one gets the display disabled and may flag the account for a policy violation. The same page defines the Typical Price as “automatically computed using the median price customers have paid for a product in the Amazon store over the past 90 days, not including prices paid during any limited-time promotions” (checked 2026-08-11, Amazon List Price) — a first-party 90-day price statistic on the detail page whether or not you track anything.

Two timing facts: existing rules reprice in “less than 15 minutes,” but a new or edited rule “can take up to one hour” to propagate — do not judge one in its first hour. And automation does not buy placement. Asked whether automating price guarantees the Featured Offer, Amazon answers “No. Lowering your price can help you become the Featured Offer, but you also need to meet performance-based requirements in areas like inventory availability, fulfillment, and customer service” (checked 2026-08-11).

A Competitor Price Watch Setup Checklist

• Define the competitive set per ASIN — the two or three offers that actually take the Featured Offer from you, not every offer on the page.
• Pick one authoritative price field and use it everywhere: Featured Offer price for decisions, lowest offer as a secondary signal only.
• Record offer count with every price point, or you cannot separate discounting from stockouts.
• Tier ASINs by watch frequency — hot list on push notifications, mid tier hourly, tail daily.
• Choose aggregation per tier, knowing FiveMinutes discards interim events.
• Set both a floor and a ceiling in whatever tool changes prices.
• Keep more than 30 days — collect your own series before you need the seasonal comparison.
• For spot checks, the free AMZBase Chrome extension shows ASIN, Best Sellers Rank, lowest FBA offer and FBA seller count on search and detail pages, with CamelCamelCamel price history and Keepa rank history on hover. Read-only — it does not change prices.

Frequently Asked Questions

How do I track a competitor’s price history on Amazon?

Three routes: a third-party tracker such as Keepa for long public history, Amazon’s free Automate Pricing tool for a 30-day record tied to your own rules, or the Selling Partner API’s getCompetitiveSummary and featured-offer-expected-price operations to collect and keep your own series (SP-API model, checked 2026-08-11).

Is the Buy Box price the same as the lowest price?

No. Amazon’s data model returns lowest-priced offers per item condition and offer type, while the featured offer is segmented by customer membership — so different shoppers can see different featured offers on the same ASIN.

How far back does Amazon’s own price data go?

Automate Pricing surfaces what Amazon calls a “detailed 30-day history” (checked 2026-08-11). Anything longer comes from a third-party tracker or from data you collected yourself.

How often can I poll competitor prices through the API?

Default rates are published per operation — 0.033 per second for the v2022-05-01 batch operations, 0.5 for v0 getItemOffers — and higher values may be granted to sellers whose volume requires it (checked 2026-08-11). Push notifications via ANY_OFFER_CHANGED avoid polling entirely.

Can automated repricing get my listing suppressed?

The Fair Pricing Policy targets prices “significantly higher than recent prices offered on or off Amazon,” so an unbounded rule that raises price is the riskier configuration. A maximum price alongside your floor is the direct control; see Fair Pricing Policy for the enforcement ladder.

Conclusion

A seller’s price watch rests on three decisions, none of which is which tracker to install. Decide which price you are reading — Featured Offer, lowest offer, or raw listing price — and remember the first is segmented. Decide watch frequency per ASIN tier, with eyes open about what notification aggregation throws away. Decide your floor and your ceiling before the first alert fires, because Amazon’s published policy compares your price against the ASIN’s own recent history. Everything else is plumbing.

API and policy details were read from Amazon’s published Selling Partner API models and Amazon-hosted seller pages on 2026-08-11. Third-party tool pricing is marked where it could not be confirmed on the vendor’s own page. Amazon revises rate limits and pricing policies without notice.