Brand vs Retailer: What Each Should Actually Monitor

Two Jobs That Share a Name

Brands and retailers are both told to monitor competitor prices, as if that phrase described one job. It describes two, with different objects, different thresholds and different ways of being expensive.

A brand monitors its own products across other people's storefronts. The object is its own SKU on marketplaces and dealer sites it does not control, the question is who is selling it and at what advertised price, and the failure mode is channel erosion: authorised dealers watch an unauthorised seller undercut them, stop promoting the line, then stop carrying it.

A retailer monitors other people's products against its own assortment. The object is a competitor SKU matched to something it stocks, the question is where its price sits on the handful of items shoppers actually compare, and the failure mode is margin loss.

That is why a brand that buys retailer repricing tooling gets nothing from it: the tool answers a question the brand never asked. The same mismatch shows up in the build-or-buy decision, where the honest comparison depends on which of the two jobs you are actually doing, as the guide to scraping it yourself versus buying a monitoring API sets out.

The Side by Side

Dimension	Brand	Retailer
Object monitored	Own SKUs across third-party storefronts and marketplaces	Competitors' SKUs matched to own assortment
Unit of alert	Seller identity plus advertised price vs MAP	Competitor price vs own price on a matched SKU
Threshold	Any advertised price below MAP; any new unrecognised seller	Gap beyond a set band, on key value items only
Latency needed	Hours to days; evidence quality beats speed	Daily on most items, hourly on contested KVIs
Response	Evidence log, then a unilateral supply decision. Never a negotiation	Repricing decision inside the margin floor
Legal frame	Colgate and Leegin, unilateral policy discipline, Maryland per se	AB 325, shared-algorithm and competitor-data exposure
Cost of a missed signal	Channel erosion, dealers disengaging, Featured Offer suppressed	Margin loss on fast movers, or overpricing on KVIs
Cost of a false positive	Terminating a compliant dealer, antitrust discovery risk	Unnecessary markdown, self-inflicted price war

Unit of alert, threshold, latency and response are the rows that decide whether a tool built for one side is usable by the other.

The Unit of Alert Is Not the Same Unit

For a retailer, an alert is a number compared to a number; the seller behind the price is a constant, since you already know who Walmart is.

For a brand, the seller is the variable. The unit of alert is a seller identity attached to a price, and half the information sits in the identity. The same 22% discount means one thing from an authorised dealer running an approved promotion and something else from a storefront nobody on your channel team recognises. A setup that reports only "price dropped to $79" has discarded the half of the signal that determines the response.

Israeli, Anderson and Coughlan, studying MAP violation patterns across hundreds of products and hundreds of retailers, found that authorised retailers comply more often than unauthorised ones and that the two markets are largely separate, with violations in one only weakly associated with violations in the other (Marketing Science, 2016).

The operational edge of that finding: policing your authorised dealers does not fix your unauthorised problem. Dealing with unauthorised sellers undercutting your brand is a separate workstream from detecting MAP violations across your retailers, and merging the two queues buries the harder one.

A Band for One Role, a Line for the Other

A retailer's threshold is a percentage band applied only to key value items. Watching the whole catalogue at a 3% band produces an alert stream nobody reads by the second week. The band should reflect how much volume you lose per point of price gap, an elasticity question before it is a monitoring question.

A brand's threshold is a line. Any advertised price below MAP is a violation regardless of how far below, because the policy is binary by construction, and a brand that quietly tolerates a 2% breach has moved its MAP down 2%. The second brand-side trigger has no percentage in it at all: a new seller on a listing you did not authorise. MAP monitoring covers how to write the policy so the line holds.

Latency splits the same way. A retailer needs speed because the response is a price change it can push in minutes. A brand gains little from learning about a violation forty minutes sooner: its response takes days and depends on evidence quality, meaning timestamped captures, seller identity, advertised price, and proof the violation persisted rather than flickered.

The Mechanism That Links Both Roles

Amazon's Featured Offer eligibility takes account of a competitive external price, a price observed outside the Amazon store. Amazon's own guidance tells sellers to "select View reference and business prices to view the current competitive external price", and says the Pricing Health page's Pricing Opportunities tab lists "your offers that aren't currently eligible to become the Featured Offer", advising sellers to price at or below that external price (Amazon, April 2026).

From the retailer's chair that is ordinary Buy Box competition: somebody outside Amazon is cheaper, match or lose eligibility.

From the brand's chair it is something else. Your own off-Amazon channel price is an input to your own Featured Offer eligibility. A deep promotion on your direct-to-consumer store, or a dealer's aggressive weekend sale, can land in that external field and suppress the offer you are trying to win. The brand competes with itself through a mechanism its monitoring is not pointed at, aimed as it is at MAP violators rather than its own promotional calendar. This is the sharp end of marketplace versus own-store price competition, and the platform mechanics sit in the Amazon repricing and Buy Box guide.

Same field, same data, two entirely different jobs.

The Legal Frame Splits Too

Both roles collect competitor pricing data. What they may lawfully do with it diverges hard, and the exposure runs in opposite directions.

Brand side: the data may only feed a one-way decision

Leegin Creative Leather Products v. PSKS (2007) moved minimum resale price maintenance from per se illegality to the rule of reason under federal law, and Maryland is the one state that restored per se treatment under its own antitrust act. The doctrine that governs daily practice is older: United States v. Colgate & Co. (1919) lets a manufacturer announce its terms unilaterally and refuse to deal with those who ignore them, because a refusal to deal is not an agreement.

The boundary is narrower than most brand teams assume. A brand may announce the policy, may terminate or stop supplying violators, and has wide discretion over what counts as non-compliance. It may not negotiate price with the dealer, accept assurances of future compliance, or let communications skirt the line of becoming an agreement. Accepting a promise to comply converts a lawful unilateral policy into an unlawful agreement (K&L Gates, June 2019).

Your monitoring data is evidence, and under a unilateral policy it may feed exactly one kind of decision: supply, or do not supply. The moment it feeds a conversation about what price a dealer will agree to charge, the same data becomes documentation of an agreement. Run the policy, the notification templates and the termination process past antitrust counsel before any of it goes live.

Retailer side: California AB 325

AB 325 took effect on 1 January 2026, amending the Cartwright Act. It defines a "common pricing algorithm" as any methodology used by two or more persons that uses competitor data to recommend, align, set or influence a price or commercial term, and prohibits using or distributing one as part of a restraint of trade. There is no public-data exception: the statute reaches competitor data whether it is public or not, which defeats the standard "we only look at publicly posted prices" answer. Corporate criminal fines run up to USD 6 million (Morgan Lewis, January 2026).

Federal enforcement so far has turned on something narrower. The DOJ's RealPage litigation and the Greystar settlement produced a final judgment barring Greystar from using revenue-management software that relies on competitors' nonpublic competitively sensitive data (Wilson Sonsini). That judgment binds Greystar, not the market, but it shows where federal enforcement has drawn the line: pooled nonpublic data. California's statute drops the nonpublic limit.

The practical line, subject to counsel reviewing your specifics: a single-firm tool using your own data and your own observation of public prices, producing a recommendation only you see, is the defensible posture. A shared tool that hands the same recommendation to your competitors is the pattern the statute targets.

Hybrids Are Most of the Real Market

The vendor pages currently ranking for this topic tend to address one side without saying which. Most real businesses run both jobs at once, on different SKU sets, and that is where configuration goes wrong.

Brand with a direct-to-consumer store. You are your own dealer. If your site promotes below your published MAP, the policy is unenforceable in practice and starts to look pretextual. Put your own store in the dealer table as a row like any other, and have antitrust counsel look at any policy your own channel is breaching.

Retailer with a private label. MAP-style logic on your own label wherever it sells through marketplaces, competitor logic on the national brands you resell. Two rulebooks, never one blended threshold: a single 5% band across both fires alerts that mean opposite things depending on the SKU.

Brand selling 1P and 3P on Amazon. Amazon as a vendor sets its own retail price and is not bound by your MAP. Your third-party sellers are. Expect your worst apparent violator to be Amazon, permanently, and build the exclusion into the alert rules instead of re-triaging it weekly.

Authorised dealer leaking to unauthorised resellers. The case brands most often presume and least often verify. The 2016 finding that the two markets are largely separate suggests the leak is rarer than assumed, so investigate with purchase-trail evidence rather than terminating on a suspicious pattern. Terminating a compliant dealer is the expensive false positive in the table above.

Nobody Has Measured Either Job

There is no credible figure for how many storefronts a typical brand's product appears on, and none for how many competitors a typical retailer needs to track. The numbers in circulation come from vendor blogs with no stated methodology. Both roles are being sold tools sized by a number nobody has measured.

The brand-side scale figures that do exist are self-reported by the platform doing the enforcing. Amazon's Trustworthy Shopping Experience Report says it identified and seized more than 15 million counterfeit products worldwide in 2025 (Amazon, April 2026). Read that as a measure of Amazon's activity rather than of residual harm to you. Amazon is grading its own homework, and none of it touches MAP, which is not an intellectual property violation and is not something Amazon enforces on a brand's behalf.

Start narrow, on the SKUs where money is visibly moving, and let observed data tell you how wide the problem is.

Configuring the Two Jobs Differently

A brand tracks a product URL for each storefront where its SKU appears, its own direct-to-consumer listing included, with alerts per variation. A breach on one size or colour is still a breach, and parent-level tracking averages it away. The alert itself fires on a price or stock change, not on who is selling: seller identity comes from which storefront each tracker points at, which is why one tracker per storefront beats one tracker per product. Tracker history is what turns those alerts into an evidence log rather than a screenshot somebody took once, and only if your plan retains enough of it. A 7-day window, which is what the free plan keeps, will not carry a termination file.

A retailer tracks competitor URLs for its key value items and watches the gap against its own price. Respot runs both configurations from one place, because both reduce to the same object, a tracked URL with price and stock alerts, and its 5 free trackers are enough to test the shape of your own problem before sizing anything.

Decide Which Job You Are Doing First

Before comparing tools, write one sentence: whose products am I watching, and what decision changes when the number moves?

If the answer is your own products on storefronts you do not control, and the decision is whether to keep supplying somebody, you are running the brand job. Alerts key on seller identity, the threshold is a line, and the data may only ever feed a one-way supply decision reviewed by counsel.

If the answer is somebody else's products matched to yours, and the decision is whether to move your price, you are running the retailer job. Alerts key on a gap, the threshold is a band on the items that drive comparison, and your exposure sits in whether the recommendation you act on reaches the firms you compete against.

If both are true, refuse to merge them into one threshold. Start with the SKUs where you already suspect money is leaking and let the data size each job for you.

Frequently Asked Questions

Is MAP enforcement legal in the United States?

A unilaterally announced minimum advertised price policy is generally lawful under federal law. Leegin (2007) moved minimum resale price maintenance to the rule of reason, and the Colgate doctrine (1919) lets a manufacturer announce terms and refuse to deal with non-compliers. Maryland is the only state with an explicit Leegin repealer. Anything beyond announcing and refusing to supply belongs with antitrust counsel.

Can a brand terminate a retailer for selling below MAP?

Under a properly run unilateral policy, yes. A brand may announce the policy and stop supplying a violator, with wide discretion over what counts as non-compliance. It may not negotiate price, accept a promise of future compliance, or let the exchange drift toward an agreement: accepting an assurance of compliance converts a lawful unilateral policy into an unlawful agreement. Have counsel review the termination process first.

Does California AB 325 make competitor price tracking illegal?

No. AB 325 targets a common pricing algorithm, meaning a methodology used by two or more persons that relies on competitor data to set or influence price. Observing public prices yourself and feeding them into your own single-firm decision is outside that definition. The exposure appears when a shared tool hands the same recommendation to your competitors. The statute reaches competitor data whether it is public or not, so the usual public-data defence does not apply.

Does my price on my own website affect my Amazon Buy Box?

It can. Amazon evaluates Featured Offer eligibility against a competitive external price, a price observed outside the Amazon store, and tells sellers to view it under reference and business prices and to price at or below it. A deep promotion on your own direct-to-consumer store can sit in that field and suppress your own Featured Offer.

Should a brand and a retailer use the same price monitoring tool?

The same tool, yes. The same configuration, no. A brand watches its own SKUs across third-party storefronts with alerts keyed to seller identity and any advertised price below MAP. A retailer watches matched competitor SKUs with a percentage band on key value items. A hybrid business keeps two separate rulebooks rather than one blended threshold.