Peak Season Monitoring FAQ
Published on October 27, 2026 by Niccolò
The Questions Sellers Ask in the Last Week of October
This is for the person who set up competitor monitoring in September, has watched October's cuts land, and now has about four weeks before Black Friday on 27 November and Cyber Monday on 30 November.
What is left to decide is narrow and mostly about settings rather than strategy. Eight questions cover almost all of it. Each answer gives the short version, then points at the post carrying the full argument. The Black Friday price monitoring playbook is the spine the rest of it hangs off.
One caveat runs through all eight: every answer assumes a price history that starts before the discounting did. Without a September baseline these questions have no answer, only a guess, and the right move is to start recording now.
How Often Should I Check Competitor Prices During Peak Week?
More often than in October, with the ceiling set by how fast you can respond rather than by how fast a tool can poll.
Detection is not the expensive step. Deciding, changing a price or a bid, and waiting for the change to take effect is where the hours go, and marketplaces add lag of their own: Walmart's repricer documentation notes that a strategy-driven price change can take up to four hours to surface on the item page (Walmart Developer, April 2026). Detection faster than your own response loop buys nothing except cost and alert volume.
So hourly is the shortest cadence with a defensible rationale, and it belongs to the handful of genuinely contested products where an hour of staleness changes a decision. Mid-tier products want every few hours. The long tail stays daily. Nothing in the published research supports a sub-hour reprice window.
Raise your alert threshold in the same sitting. Tightening frequency while leaving a 5% threshold in place during a week when whole categories run past 20% off is the fastest route to a feed nobody reads. The frequency framework by product tier applies as written.
Should I Turn My Repricer Off During Black Friday?
No, provided three guardrails are in place. If they are not, yes, turn it off.
Turning it off trades one risk for another: you stop the runaway loop and you also stop responding for the busiest week of the year. The three guardrails are a cap on net movement per SKU per 24 hours, typically 5 to 10 percent with anything larger requiring a human to release it; a cooldown of 30 to 60 minutes after every downward move; and an explicit instruction for what the rule does when data is missing, stale or implausible.
The third one is load-bearing and almost nobody writes it down. On 12 December 2014, a fault in the RepricerExpress service dropped Amazon UK listings to GBP 0.01, affecting around 2,000 sellers in roughly an hour, with one merchant reporting GBP 100,000 of inventory lost before it was contained (Washington Post, December 2014). When the software lost confidence in its own state, it priced to the bottom; a rule holding its last known price would have produced a boring hour of stale prices.
With those three in place the worst case is a price that stands still. Without them the worst case has no floor. The full set of guardrails, the order they have to run in, and the edge cases worth deciding in advance are in repricing rules that survive peak season.
How Do I Tell a Real Competitor Price Cut From a Glitch?
Wait a few hours. Very little that deserves a response deserves it inside the hour, and time separates the two cases better than any instant test.
Then apply three checks in order. Does it hold? A price that recovers within hours was a test or an error, while one that holds for a day or two is a strategic move, and only the second kind is worth matching. Is it buyable? A rock-bottom price on an out-of-stock listing or a disabled add-to-cart button is not competition. Is it plausible against history? That check needs two reference points: the range that competitor has run before, and what the category is doing. Against category averages in the mid-teens, roughly where the seventh answer below puts them, an observed 60% cut in electronics is far more likely to be something broken in the feed, the page or the parse, so re-fetch it by a second method before believing it.
The full triage, including the artefacts your own collection method creates and the real prices shown only to some shoppers, is in the guide to telling a real price cut from a glitch.
What Does a Competitor Stockout During Peak Week Actually Get Me?
Usually close to nothing, occasionally quite a lot, and the difference is checkable in about ten seconds.
Most of the demand a stockout releases never leaves the competitor's page. Researchers analysing roughly 1.5 million records across 217 stores found that 51.7% of a stockout's unmet demand moves to an adjacent size of the same style (Li, Lu, Lu and Huang, Manufacturing & Service Operations Management, 2023). The alert was accurate and the event was commercially internal to them.
So the first check is whether the sold-out variant has a close substitute on their own listing. If the medium is gone but the small and large are live, the released demand has somewhere to go that is not you. If the product exists in one configuration, or the whole line is out, that path is closed and the alert is worth reading properly.
The second check catches the response that does real damage. Raising your price on a rival's stockout only works if you already held a comparable price before it happened. A peak-week stockout is often the terminal state of a promotion, so the shoppers now arriving at your listing were shopping the rival's discounted number rather than your everyday one. Raise against that expectation and you convert a demand transfer into a bounce. The other two conditions are in what a competitor stockout is actually worth.
Why Can I Not Discount as Deeply as I Planned on Black Friday?
Because Amazon evaluates deal pricing backward, and your September and October prices have already set the ceiling.
Two mechanics do the damage. The first fires on a single promotion: a promotional price has to beat your own recent record, so a plain price discount run through Seller Central in October sits inside the window a November deal is measured against. The mechanic decides whether it hurts, not the depth: a price submitted through Prime Big Deal Days is excluded by Amazon's own wording, while the identical price entered as a manual discount is not. That exclusion is no help this season. The submission window ran from 8 July to 8 September and has closed, and Amazon has not announced dates for an autumn Prime event, so nothing you run now can be routed through it.
The second is slower and costs considerably more. From 18 May 2026, if more than half the days in a product's 90-day price history sit below the non-promotional median, Amazon recalculates Typical Price using all sales including promotional ones (Amazon Seller Forums, April 2026). More than half of 90 is 46 days. Cross it and the strike-through collapses to the discounted level, so the same advertised badge shows the customer a smaller-looking cut while costing you real margin per unit.
Be careful with the percentage gates circulating on this topic. The "minimum 15% off" and "at least 5% below the 30-day low" figures repeated across agency blogs are not traceable to any Amazon-published source, and the description of the floor as the lowest price customers actually paid comes from a seller-forum answer rather than an Amazon moderator. If this year's honest answer is "less deeply than planned", count your discounted days now and decide next year's number in September. The arithmetic is in the 30-day lookback explainer.
What Should I Be Recording Now That I Cannot Reconstruct Later?
Four things, none of which any dashboard will hand you in December.
The pre-promotional price, per competitor, per variation. This one has a deadline attached, because once everyone is discounting, the "before" number exists only if somebody wrote it down. A watch list built in mid-November records the already-discounted price as its baseline, which makes every later comparison meaningless.
Who moved first, and when. Timing is the intelligence and depth is only the headline. A competitor who cuts in the first week of October is inventory-heavy and wants volume early; one holding list price into late November is either supply-constrained or betting on late shoppers. Record separately which rivals quietly reset list price upward beforehand, so a November badge lands near their August price.
Stock state alongside every price, at variation level. A price recorded without a stock flag cannot later be told apart from a price nobody could transact at, which poisons the post-mortem and any rule set you replay against it.
Your own discounted-day count per SKU, since that is what the deal-eligibility question above depends on and it is invisible in the seller dashboards.
Two process notes. Freeze the watch list before peak week opens, because anything added later has no pre-promotional price. And set thresholds and cadence before Thanksgiving, since changing a setting while prices are moving leaves you unable to tell a calm market from a broken threshold.
Do Competitor Prices Go Back Up After Cyber Monday?
Partly, slowly, and less than the discount headlines lead you to expect, because most of the cut was already in place before the event and does not unwind with it.
DataWeave's tracking of roughly 80,000 US SKUs across major retailers makes the shape clear. Consumer electronics carried an average 14.6% discount before Black Friday week and gained 2.6 percentage points during it. Home and furniture ran 13.9% before plus 1.7 points. Apparel ended highest at 17.6%, from 13.8% before plus 3.8 points (DataWeave, December 2025). The event-week increment is the part that comes back off. The baseline underneath it is a standing posture that started in October and tends to hold into December.
Peak-day rates also run deeper than period averages, which makes the answer look more confusing than it is. A competitor who looked far deeper on the Friday morning than on the following Tuesday may not have raised anything; you observed two different hours.
Practically: the deepest prices vanish within days of Cyber Monday, the standing discount persists through December, and the first is not a rival retreating. Revert your own peak thresholds and cadence in the first week of December, since tighter settings left running into a quiet January burn checks and hide real signals.
Is My Peak-Season Data Any Use for Setting Normal Prices?
Mostly no, and treating it as though it were is one of the more expensive mistakes available in January.
Peak week is the single largest confound in your calendar. Sales are elevated, your ad spend is up, category traffic is up, and a lift you observed after a price move would very likely have appeared anyway. Since you cannot observe the counterfactual store where you did nothing, a price and volume pair from that week tells you very little about how demand responds. Useful price elasticity work needs observations where price is close to the only thing that changed, and peak week is the furthest thing from that.
The demand mix is different too. Shoppers were selected by a discount, a larger share of the purchases are gifts, and the return rate lands weeks after the revenue does. NRF and Happy Returns put returns at 19.3% of online sales across the year, with retailers expecting 17% of holiday sales back, from a survey of 2,006 consumers and 358 ecommerce professionals at merchants above $500 million (NRF, October 2025). That 17% is a retailer expectation rather than a measurement of what came back, and the sample covers large merchants, so treat it as a placeholder until your own January reconciliation produces a real rate.
Peak data is genuinely good for three things: competitor behaviour, a durable trait covering who cuts first, who holds and who resets list price upward beforehand; your own operational limits, meaning where fulfilment or the site started to strain; and the discounted-day count that decides next year's deal eligibility. Everything about your own price sensitivity waits for a quiet month.
Recording the Part You Cannot Reconstruct
Most of the answers above reduce to the same requirement: a record of price and stock, per variation, that starts before the discounting did and is honest about what was actually buyable.
Respot handles that layer. You paste a competitor product URL, it detects the product details, then tracks price and stock per variation rather than at parent-product level, which is what makes the stockout check in the fourth answer possible at all. Stock changes alert separately from price changes, so a sellout is not buried under promotional noise.
Two limits matter before you plan around them. Check cadence is adaptive rather than an interval you set, with plan floors of two hours on starter, one hour on pro and thirty minutes on agency. History depth is the other: the free plan keeps 7 days across 5 trackers, enough to confirm your contested products are trackable but not to count 90 days against a median.
The One Setting to Fix This Week
If you change a single thing, make it the answer to one question: what do your monitoring and your repricing do when an observation fails to arrive, arrives stale, or arrives implausible?
Most setups have no answer, which means they have a default nobody chose, and defaults chosen by accident are how a repricer sells inventory for a penny. Write the failure behaviour down, then test it by withholding observations on one product while the stakes are still a rehearsal. Everything else here is a refinement; this one decides whether the worst hour of your peak week costs you a little or costs you the day.
Testing it needs a baseline that already exists, which is the one thing you cannot add in November. If you have not started recording, put your contested products under monitoring now and let the four weeks before Black Friday build the history the rest of these answers assume.
13 min read
Catch competitor price cuts before they cost you sales
Respot monitors any competitor product page and emails you within minutes of a price or stock change. Free plan, no credit card.
Catch competitor price cutsRelated articles
What to Do With Your Pricing Data After Black Friday
The post-Black-Friday analysis only works if you recorded the right things beforehand. What to capture before peak week, and what to run in December.
Price Anomalies: Telling a Real Cut From a Glitch
How to triage a sudden competitor price drop in two minutes: eight checks that separate a real cut from a scraping artifact, a geo price, or a pricing error.
Brand vs Retailer: What Each Should Actually Monitor
Brands monitor their own products on other people's storefronts. Retailers monitor other people's products. Different objects, thresholds and legal risk.