Last updated: June 17, 2026 by the Trellis team

Amazon Prime Day advertising rewards preparation, not reflexes. Prime Day 2026 runs June 23 to 26, a 96-hour window that opens at 12:01 a.m. PDT on the 23rd, and most operators are asking the same question: how much should I spend?

The standard answer is a multiplier. Two to three times your daily budget, or 300 to 500 percent if you read enough guides. It is easy to act on, which is why it spreads, and it is the wrong instrument for the decision in front of you. A blanket multiplier treats every product in your catalog as if it has the same job, and it pushes more money into the auction at the exact moment the auction gets most expensive.

Everything below you can run yourself in Seller Central and the ads console, by hand or with Amazon’s own tools. We will note where a platform earns its keep, but the tactics are yours to execute. Start with the budget, then keep the surge from draining it.

Why the blanket multiplier is the wrong instrument

There is no safe universal multiplier for Prime Day. Instead of scaling every campaign two or three times, size headroom by each SKU’s job during the event, the margin it carries after the deal discount, and where your conversion concentrates. A few products and search terms usually deserve most of the increase.

Think of your catalog the way a portfolio manager thinks about positions. A new launch needs velocity even at a high cost. A mature hero SKU needs efficient harvest. A defensive keyword needs position held even when returns compress. Telling the whole portfolio to “hit one multiplier” is like telling a fund to target one return across growth stocks, defensive plays, and new issues alike. The average might land. The individual decisions will be wrong.

That failure is easy to watch in real time. Triple every budget and the auto campaign on a thin-margin SKU spends fastest, because it has the widest targeting and the loosest intent. The launch you needed to fund for rank hits its cap by noon and goes dark. The hero SKU on a deep deal, the position you most wanted to own, competes for the same headroom as a product you would rather protect. Same multiplier, four different jobs, and only one of them was served.

How to build your Prime Day advertising budget bottom-up

A defensible Prime Day PPC budget is built from three inputs, not one. Size each campaign by the job its product is doing during the event, by the margin you have to spend into, and by where conversion concentrates. The multiplier is the output of those three decisions, not your starting point.

Start with the job. Sort your advertised SKUs into a few postures and fund them differently. Each posture maps to settings you already have in the ads console.

Then check the job against margin. The 2026 cost stack is heavier than it was a year ago, so a campaign that was profitable at a given ACoS in 2024 may not be today. Set the ceiling on each campaign against the margin the product carries after fees and the deal discount, not against a number you remember being comfortable.

Finally, follow the concentration. Most accounts find that a small share of SKUs and search terms drives most of the Prime Day revenue. Fund those first and hard. Spreading the same headroom evenly across the catalog feels safe and wastes the budget on targets that will not convert during a four-day rush.

The three phases of Prime Day advertising, sized not guessed

A Prime Day advertising strategy runs in three phases, and the work that matters most happens before the 23rd. The budget split between those phases is a decision, not a default.

Lead-up (now through June 22). This is audience and rank building, not direct response. Move proven hero ASINs and high-converting exact-match terms into position, and raise budgets in steps a few days out so campaigns are not learning from a cold start on day one. Amazon recommends ramping ahead for exactly this reason, and you can schedule the increase with a budget rule or do it by hand in 20 to 30 percent steps. With the event six days away, deal submissions and inventory deadlines have passed for most sellers, so the lever you still fully control is the ad budget.

The event (June 23 to 26). Four days behave like one long peak with two crests. Front-weight Day 1, when demand opens, and Day 4, when closing-surge behavior kicks in, and run leaner through the middle. This is also where most budgets break, for reasons that have nothing to do with the multiplier you chose.

Lead-out (June 27 onward). Keep advertising for at least a week. Retarget shoppers who viewed but did not buy, and read the new-to-brand data the event generated. The rank and the audiences you built are the asset. The hard stop on June 27 is one of the most common ways operators give that asset back.

Keep the surge from draining your budget

The danger is not the budget you set. It is what the auction does to it. Cost per click typically climbs 20 to 40 percent during Prime Day as more advertisers bid into the same auction. Amazon’s up-and-down dynamic bidding can raise a single bid by as much as 100 percent on top placements. And Amazon’s own budget logic can spend past your stated daily budget on a high-traffic day, then reconcile across the month, which means your cap is softer than it looks during a surge. Add a 3x multiplier on top of all that and the budget can be gone by mid-morning on a target set that was never going to convert.

The fix is a few controls you set on purpose, all of them native to the ads console:

• Choose your bidding strategy deliberately. Dynamic bids, down-only is the event safety setting: Amazon can lower your bid but never raise it above your maximum, so the surge cannot inflate your CPC beyond what you decided. Reserve up-and-down for the terms you are confident convert, knowing Amazon can double the bid.
• Adjust bids by placement. Concentrate your placement boost on top of search for hero and deal SKUs, where Prime Day intent is highest, and pull back product-page placement where it does not earn.
• Schedule the ramp and the revert. A budget rule can raise budgets for the event dates and reset on its own. If you do it by hand, raise in steps a few days out and set a reminder to bring them back down. Manual changes do not revert themselves, and a boost left running after the 26th is a quiet leak.
• Hold a reserve. Do not pre-allocate all of your headroom. Keep 15 to 20 percent back and release it onto the campaigns proving out during the event, not evenly across the catalog. That is the difference between funding a surprise winner and robbing your planned ones to chase it.

All of this is doable by hand. Knowing what to do is the easy part. The hard part is doing it continuously, across thousands of targets, in the 96 hours that matter most. By hand you can set a bidding strategy once. You cannot truly pace by the hour, hold spend inside a floor and a ceiling instead of a single guess, or make every change revert itself when the window closes. That is the gap a platform like Trellis closes, and event week is where the manual approach shows its limits.

The cash and margin reality nobody budgets for

A Prime Day budget sized only against GMV will look great and feel terrible. Three parts of the 2026 cash and cost picture deserve a line in your plan.

Seller payouts now arrive seven days after delivery, so the cash you spend front-loading campaigns on June 23 may not return until well after the event closes. Hold a cash reserve, not just a budget reserve, so a surprise demand spike is an opportunity rather than a cash-flow problem. FBA fees rose earlier this year and a fuel and inflation surcharge was added in April, and elevated tariffs continue to compress margins across many categories. The practical effect is that the ACoS you can afford on a discounted unit is lower than it was, so size the budget against margin and cash flow, not against the top-line revenue the event will generate.

Keep pricing in the same line of sight. When your deal price and the bids behind it live in separate tabs, it is easy to fund a campaign into a price that no longer carries the margin to support it. You do not need to automate repricing to fix this. You need a guardrail: decide the floor price you will not cross, set the ad ceiling against the real discounted margin, and check both before you raise a budget.

Then size your scoreboard before the event, not after. Attributed ROAS in the ads console flatters during a surge, because the event drives sales your ads would have captured anyway. Read TACoS, total ad spend against total sales, alongside it, and reconcile the ads console against your Business Reports in Seller Central. A number that reconciles is a number that survives the first question your boss or your client asks.

What to do this week

With Prime Day 2026 six days out, your PPC budget is the lever still fully in your hands. Before the 23rd:

• Sort your advertised SKUs by the job each one does during the event, and fund them by job, not by one multiplier.
• Set each campaign’s bidding strategy on purpose, with down-only where you want a hard cap on CPC.
• Concentrate headroom and your top-of-search placement boost on the SKUs and terms that drive your revenue.
• Schedule your budget ramp, and plan the revert so a boost does not run past the event.
• Hold a reserve, in budget and in cash, for surprise demand.
• Plan the lead-out now, so retargeting and rank protection are funded before you are tired on June 27.

Amazon Prime Day advertising comes down to one discipline: knowing your number and protecting it. A multiplier gives you neither. A budget built by job, margin, and concentration gives you both, and you can see it run on your own account if you would rather not pace 96 hours by hand.