Seller Fulfilled Prime Just Got Harder. You Have Until October 17 to Fix It.

Amazon raised the delivery speed bar for Seller Fulfilled Prime on July 6. If you run SFP and you have not looked at your delivery promise coverage since spring, you are probably already failing the new thresholds and do not know it yet.

The reason you do not know it yet is the grace period. Amazon is not enforcing on the new speed metric calculations until October 17. That is your runway, and it is shorter than it looks because peak planning eats September.

What actually changed

Amazon announced this on May 26 and it took effect July 6. Thresholds moved up on all three size tiers:

Standard size

• One-day delivery: 30% to 40% of Prime page views
• Two-day delivery: 70% to 75%
• Five-day delivery: stays at 90%

Oversize

• One-day delivery: 10% to 15%

Extra large

• Two-day delivery: 15% to 25%

Read the metric carefully, because sellers get this wrong constantly. These are percentages of Prime customer page views, not percentages of orders shipped. You are being measured on the delivery date Amazon displays to shoppers before they buy, weighted by where those shoppers are. A seller shipping fast from a single East Coast warehouse can hit their promises on every order and still fail the threshold, because the promise shown to a shopper in Phoenix was never a one-day promise in the first place.

The standard-size one-day jump from 30% to 40% is the one that will pull the badge. That is a 33% relative increase in required one-day coverage, and one-day coverage is the hardest and most expensive kind to buy.

The two ways out, and they cost very different amounts

There are only two levers that move page-view-weighted delivery speed: where your inventory sits and how late you can accept an order and still ship it same day.

Adding a second or third node is the durable fix and the expensive one. If you are a single-warehouse SFP seller on the East Coast, no amount of carrier upgrading gets you to 40% one-day coverage, because roughly a third of the US population is more than one ground day away from you. You need West Coast or Central inventory. That is a 3PL contract, a working capital hit, and a split-inventory forecasting problem.

Cut-off times are the cheap lever and most sellers have not touched theirs in years. Moving your same-day ship cut-off from noon to 4pm converts a meaningful slice of afternoon page views from two-day to one-day promises, with no new inventory and no new lease. Same for weekend pickup. If you are not shipping Saturday, every Friday afternoon and Saturday page view is showing a Monday-plus promise. Adding Saturday ship-out is usually the highest-ROI change on this entire list.

Upgrading everyone to air is the third option and it is almost always the wrong one. It fixes the metric and destroys the contribution margin you were protecting by staying out of FBA. If your only path to 40% is paying for expedited on 40% of units, run the math against FBA fees honestly. For a lot of standard-size SKUs, SFP stopped being the cheaper answer somewhere in the last two fee cycles.

The September tool is worth waiting for, but not worth waiting on

Amazon is shipping a zip-code-level delivery promise tool in September that lets you feed granular cut-off times and weekend availability directly into the delivery estimate calculation. That matters, because a lot of SFP sellers are being underrated today: your actual capability is better than what Amazon models, and the model is what shoppers see.

Do not treat that tool as the plan. It closes the gap between your real capability and your displayed promise. It does not create capability you do not have. If you genuinely cannot reach the West Coast in a day, better data will show that more accurately, not fix it.

Use August to fix the physical constraints. Use September to make sure Amazon’s model reflects them.

What to do before October 17

1. Pull your current delivery speed metrics in Seller Central and compare against the new thresholds, not the old ones. Know your gap in percentage points before you spend anything.
2. Segment by size tier. A seller who is fine on standard can still be failing extra large at the new 25% two-day bar. The tiers fail independently.
3. Move your ship cut-off later and add weekend ship-out. Cheapest points on the board. Do this first.
4. Model a second fulfillment node against your gap. If cut-off changes get you to 34% and you need 40%, you need geography, and a 3PL onboarding takes longer than the runway you have left.
5. Run the SFP-versus-FBA math per SKU, not per account. The answer is different for a 2lb standard-size item than for a 30lb oversize one. Some of your catalog should probably move.
6. Have the decision made by mid-September. Do not walk into Q4 planning with the badge in question.

The blunt read: Amazon has spent three years steadily raising the price of keeping SFP. For part of your catalog that is worth paying. For the rest, this is the cycle where the honest answer is to stop defending a badge that is costing more than it returns. Decide which SKUs are which before the grace period runs out, not after the badge is gone and your conversion rate tells you.

Fulfillment network decisions, cut-off optimization, and the SFP-versus-FBA call per SKU are the kind of operational work we run for the brands we operate. If you have SFP exposure heading into Q4 and want the analysis done before October 17, a strategy assessment is the fastest place to start.