Peak Week WBR: Which Five Fires to Fight

In peak, every brand on Amazon has more fires than people. So which five do you fight in peak week, and which do you let burn?

My first manager at Amazon answered that in my first week as a vendor manager: 10 to 20 high-visibility action items would land every week, and I would pick the 5 with the most impact on my P&L and category goals. The next week brought a new set plus last week's backlog, and I had to justify the cut to my manager. Peak raises the count of fires; it does not add people. The real job of a peak war room is deciding what not to fix, and the way to do it is a fixed rhythm on input metrics, the numbers your team can move directly.

Deadlines in the next 30 days

Dates were checked on 25 October 2026. The two earlier inbound cutoffs, 14 October for AWD bulk shipments and 21 October for FBA minimal splits, have already passed.

For any brand selling through FBA, Amazon's peak fulfillment surcharge raises the fulfillment cost behind every margin line in your review; a 1P vendor feels it only if it also sells as a 3P seller. FedEx and USPS Christmas ship-by dates are confirmed for 2026 and are next week's subject.

What belongs in a peak weekly business review?

Inputs before outputs, in a fixed format, with last year's peak on the same chart. A weekly business review (WBR) is the same deck of charts, read in the same order every week, so a change jumps out. Amazon's own leadership reviews run this way: public accounts describe a deck that lets leaders scan 400 to 500 metrics in a 60-minute meeting, which runs 90 minutes in holiday peak. That is Amazon's company review, not something Amazon requires of vendors, but the method copies well. At Amazon, I was taught it as a discipline: an owner for every line, every variance explained or escalated.

A peak WBR template to copy:

1. Order the metrics inputs first. This is a starting order to adapt, not an Amazon template:
   • Customer experience: review rating, return rate, in-stock rate, late shipments
   • Traffic and conversion: sessions, unit session percentage (units ordered per session), Buy Box percentage (how often your offer holds the Add to Cart button)
   • Demand outputs: units, ordered product sales
   • Advertising: spend, advertising cost of sales
   • Profitability: Net PPM for 1P, Amazon's measure of its own margin on your products
2. Never show an output without its inputs beside it. Revenue, profit and conversion are the scoreboard, and you cannot move them directly. In-stock rate, review volume and selection added are the drivers you can act on.
3. Draw every metric as a 6-12 chart. The trailing 6 weeks as weekly points beside the trailing 12 months as monthly points, last year's line faded behind, the target marked. Six weeks rather than four avoids month-end distortion. In peak, this is the chart that puts this week beside last year's peak.
4. Read variance one of two ways. Routine: the owner says there is nothing to see, and the meeting moves on. Exceptional: the owner gives the root cause, or says "we don't know, still investigating", and a follow-up is logged for next week. An explanation invented on the spot is never accepted. Repeated target misses escalate.
5. Keep strategy out of the room. The review spots variance. Rewriting the plan goes to a separate session, so the weekly review never turns into a planning meeting in peak.

A 1P vendor should add the availability inputs Amazon's own teams watched when I was on its side: PO confirmation rate, PO rejections, out-of-stock percentage and Retail Fast Track, an availability measure Amazon's retail teams track. Both the in-stock and vendor manager teams watched these closely, and every variance had to be explained: why it happened, and whether there was a plan to fix it.

Those four were on our review every week. What changed as peak got closer was the focus: the review leaned harder on deals and the metrics that affect whether a deal succeeds.

What should you look at every morning during peak?

The few input metrics that can break inside a day: stock, sales pace and ad spend. The method expects every metric owner to read their own numbers daily, and being surprised by your own metric in the weekly review counts as a failure. October Prime Day was the test of this rhythm: whatever surprised you in that review belongs on the morning check now.

On the 1P side, Vendor Central's working set is three dashboards: Retail Analytics, Operational Performance and Inventory and Forecasting. Vendor Central renames things often, so check the names in your own account. Three checks:

• Hourly sales, if your account shows it. Vendor Central reportedly shows hourly sales data for the past 48 hours. That comes from a practitioner, so confirm the view exists in your account before you build a morning check on it.
• Open purchase orders against sellable on-hand units. Rising open POs against flat sellable stock point at Amazon's receiving, not your shipping, and explain out-of-stocks that sales data alone cannot.
• Ad pacing. For any brand advertising on Amazon, 1P or 3P, Amazon Marketing Stream pushes hourly ad metrics, including budget usage, into your own systems, against 12 to 24 hours on standard reports, so with an Ads API integration you can pace ad budgets within the day.

On Amazon's side, the weekly review itself kept its normal cadence through peak. What shifted was the day-to-day work of the category teams, and on the day of the event itself, we were monitoring and checking everything live.

How do you decide which fires are worth fighting?

Rank each fire by its impact on the number you own, fix the top five, and write down what you are dropping so the cut can be defended. In my experience on Amazon's side, vendors were usually on top of their numbers in Q4, given the size of the opportunity and the pressure to hit annual targets. The hard part is choosing.

How many of my weekly items came from the event depended on how close it was: the nearer a Prime event or Q4 peak, the more of our attention shifted to it. All five of these kinds of fire competed for the five slots: a vendor short of stock on a deal item, a purchase order the vendor would not fully confirm, a deal that failed to launch, a price that was off, a listing that had gone down. The checks on each of them tightened as the event got closer.

It helps to know how Amazon's side ranks your fires. When I was a Senior Vendor Manager on Amazon's side, from 2016 to 2019, the vendor manager was goaled on far more than the P&L, including availability, programme participation, deals and marketing events and their results, and new selection. Deals, events and availability are all on that scorecard, so peak is when your problems and the vendor manager's goals overlap most.

There was no standard vendor cadence. The vendor manager decided who got weekly, monthly or quarterly meetings; the largest vendors were always met, and some only at the annual negotiation. Most 1P vendors are never chosen for a recurring review, so the rhythm has to be your own. What moved a vendor up the list mid-year was the ability to move the vendor manager's goals, or a big swing in key metrics in either direction. A crisis earned attention as reliably as momentum did.

When a recurring review did happen, it covered three things:

• Winners and losers in the metrics
• The programme pitch: Amazon's retail programmes, events and selection
• Friction from either side: chargebacks eating the marketing budget, suppressions on top ASINs, in-stock rates falling because of PO rejections

Vendors under-use the third. The cost of a slipped input was concrete: if a vendor down-confirmed a peak PO and the vendor manager worried about out-of-stocks, it could get the vendor's promotions cancelled.

Explaining a variance at the weekly review was the default. In peak, three kinds of problem were escalated straight away instead:

• A deal becoming ineligible because its price had swung
• A high-revenue ASIN trending toward an out-of-stock
• A content problem on an ASIN that drives the category

What does your vendor manager need from you in peak week?

One fire that moves their goals, with the evidence attached. An availability or event problem, with the numbers that show its size, fits their scorecard. Low-value problems they cannot fix, such as routine detail page content complaints, counted against the vendor when I was on Amazon's side. Bring a content problem only when it puts a deal or a top ASIN at risk.

The rule I worked to was simple: around peak, the requests that get a vendor manager's attention are the ones that will have the biggest impact on their P&L. Lead with that impact.

The volume leaves a trail. Amazon deductions land after peak, and a margin recovery audit is how a brand checks them once the season closes.

The decision for this week: name who owns each input metric on the review, and set when the review meets.

If you want a second pair of eyes on your peak review before Black Friday week, a 45-minute working session on your metrics, owners and morning checks is the fastest way to set it up. Get in touch to book one.