How to monitor MAP violations, step by step

MAP monitoring is a loop, not a report. Every listing that advertises your product gets checked against the MAP in force for that SKU on that date. When a listing falls below the floor beyond your tolerance and stays there past your grace window, the page is captured, the seller is identified, a notice goes out, and a later check verifies the cure. The ten steps below are that loop, in the order you have to build it.

Before you monitor: the two lists everything depends on

Almost every stalled MAP program stalls in the same place, and it is not detection. It is that nobody can produce a trustworthy authorized seller list or a MAP price list with effective dates. Without the first, you cannot tell a partner who slipped from a diverter who was never yours. Without the second, you cannot prove which floor applied on the day of the capture.

• The authorized seller list. Every account permitted to advertise your products, with its legal entity name, its storefront names on every marketplace it operates, and its status date. Marketplace display names change; keep the aliases. This list is what turns a violation into either a channel conversation or a diversion investigation.
• The MAP price list. One row per SKU with the floor and the date it takes effect, versioned as its own document rather than buried inside the policy. Seasonal or promotional floors get their own dated rows. The MAP policy template covers how the policy should reference it.

How to monitor MAP violations: the ten-step procedure

1. 1

   1. Fix the authorized seller list, with aliases

   Reconcile your reseller accounts against every storefront name you can observe in the channel. One legal entity often trades under three or four marketplace names. Record each alias against the parent account so a violation attaches to the account you can actually act on, not to a storefront nobody in your building recognises.

2. 2

   2. Publish the MAP price list with effective dates

   One floor per SKU, one effective date per floor, kept in a file the monitoring reads. Include the currency and the condition the floor applies to. If you run a promotional MAP for a period, add it as a dated row rather than editing the existing one, so the historical floor is still recoverable when you enforce a violation from three weeks ago.

3. 3

   3. Choose the channels to watch, and rank them

   List every surface where your products are advertised: authorized dealer sites, marketplaces you sell on, marketplaces you do not, comparison shopping engines, and paid search. Rank by revenue exposure rather than by irritation. The largest single source of below-MAP advertising is usually one or two marketplaces, and the loudest complaints usually come from somewhere else.

4. 4

   4. Set the detection thresholds

   Three settings decide what counts: tolerance in percent or dollars, whether an on-page coupon counts toward the violation price, and whether an out-of-stock listing can violate. Set them to match the wording of your published policy, not to match your instincts. A rule you enforce that is not in the document is a rule a reseller can push back on.

5. 5

   5. Set the grace window

   A grace window is how long a listing must stay below the floor before a case opens. It filters pricing errors, feed glitches and five-minute experiments out of your queue. Twelve to twenty-four hours suits most brands. Set it to zero on your top SKUs during peak periods, when a short violation still does real damage.

6. 6

   6. Set the refresh cadence to the speed of the category

   Any violation shorter than your check interval can be missed entirely. Checking once a day means a promotion that runs from 08:00 to 22:00 between two checks leaves no trace. Match the interval to how fast the category moves, then accept that the interval is the floor on what you can see.

7. 7

   7. Capture evidence at the moment of detection

   A full-page screenshot of the listing, the capture timestamp, the source URL, the observed price, any crossed-out reference price and the seller identity as shown at that moment. Evidence gathered later is evidence of a different page. Capture on every subsequent check too, because a sequence of dated captures establishes duration, and duration is what moves a reseller.

8. 8

   8. Identify the seller and attach the case to the account

   Resolve the storefront to a legal entity using your alias map. Cases key to the seller, not the URL, so one seller undercutting on nine SKUs appears as one offender with nine listings. If the seller resolves to nothing on your list, it goes to the diversion workflow instead, which is a different problem with a different fix.

9. 9

   9. Send the notice and start the clock

   Notice goes to the address on file, states the SKU, the URL, the observed price, the MAP in force on that date and the correction window, with the capture attached. Record the send date against the case. The cure clock runs from the observation timestamp in the notice, not from the day the case reached your desk.

10. 10

    10. Verify the cure, then escalate or close

    Confirm the correction with your own fresh check rather than the seller's reply. If the listing is compliant, close the case and record the time to cure. If it is not, apply the next step on the published ladder without further discussion, because a ladder that pauses for negotiation is not a ladder. Every closed case adds to the seller's occurrence count for the lookback period.

Setting thresholds and grace windows without drowning the queue

Detection settings are a tuning problem. Too tight and the queue fills with rounding artefacts nobody will ever send a letter about. Too loose and you miss the 8 percent undercut that is quietly resetting the price floor in a category. The arithmetic is worth doing explicitly: a $199.00 MAP listed at $189.00 is $10.00 low, which is 5.0 percent, because 10.00 / 199.00 = 0.0503. A $199.00 MAP listed at $159.00 is 20.1 percent low. Those are different problems and should not share a queue.

How cadence bounds what you can detect

Check interval sets a hard ceiling on detection. If you check every N hours, a violation shorter than N hours can occur entirely between two checks and leave nothing behind, and for violations that do get caught the average delay from start to detection is about N divided by 2. That is not a vendor claim, it is arithmetic, and it is the single most useful number when deciding what cadence a category needs.

Choosing channels, and what changes on each one

The same violation looks different depending on where it appears, and the levers available to you differ too. Watch the channels where you have a lever first, then the ones where you only have information.

Capturing evidence that survives an argument

A notice that says our system shows you below MAP invites a conversation about your system. A notice with a dated image of the reseller's own page ends it. The difference in outcome is not subtle, and it is the reason evidence capture belongs at detection rather than at escalation.

• Full page, not a crop. The surrounding page carries the promotional wording, the badge and the shipping offer, all of which matter if the case turns on effective price.
• Timestamp on the capture, not on the export. The cure clock is measured from when the page was read.
• The URL as read, including any parameters. Resellers move pages; the URL at capture time is what the notice refers to.
• A capture per check, not per case. Three dated captures on consecutive days prove duration. One capture proves a moment, and a moment is easy to call a mistake.
• Seller identity at capture time. Marketplace offers change hands. Record who held the offer when the price was observed.

Everything on that list is what the MAP protection module produces automatically on each violation, and what an evidence packet contains when you export it for enforcement.

The weekly review that keeps the program alive

Detection running is not a program. A program is a scheduled review where somebody decides what happens next, and it takes about forty minutes a week once the queue is tuned.

1. 1Work the priority queue first. Everything over 15 percent below MAP, oldest first. These are the cases that reset category price perception.
2. 2Check cure verification on last week's notices. Any case still open past its window moves to the next rung of the ladder, without an email discussion about it.
3. 3Review new sellers. Any storefront advertising your products for the first time this week gets resolved to an account or routed to the diversion workflow.
4. 4Scan repeat offenders. Sellers with two or more occurrences in the lookback window are the ones the ladder exists for.
5. 5Read the coverage report. If a source stopped returning data, you have a blind spot rather than a clean week. Silence and compliance look identical in a violation count.

Month to month, four numbers tell you whether the program is working: compliance rate, median time to cure, repeat-offender share, and the share of violations coming from sellers not on your authorized list. The last one is the important one. If it is the majority, the pricing conversation is a distraction and the real work is upstream. For the wider context on what a policy can reach in the first place, start with what MAP pricing is.