A 90-Day Plan to Launch a Product on Amazon

Most disappointing launches share three traits. The listing was not ready when the traffic arrived, the ad budget came from hope instead of arithmetic, and nobody wrote down when the spending would stop. A new product has no reviews, no sales history and no organic rank, so it needs help, but that help should be a planned, capped investment. This plan splits the first 90 days into three phases, with a readiness check before day one and a clear question to answer at the end of each phase.

Before day one: make the listing convert

Traffic is expensive and a new listing wastes it faster than an established one, because it has no reviews to reassure anyone. So the first job is a page that does as much persuading as it can without them.

Start with the product itself: stock received and sellable, not merely shipped. If you are eligible, make sure Brand Registry is in place, since several of the tools below depend on it. Then look at the listing as a shopper would, on a phone first. Does the main image stand out in a row of search results and meet Amazon's image rules for your category? Do the title and bullets use the words shoppers actually search with? Do the other images show scale, use, and what makes this product different? Is the price defensible next to the nearest alternatives?

If any box is empty, advertising can wait a few days. A click costs the same whether the page is ready or not, but it only pays back if the page converts.

Set the launch budget from margin, with an end date

Launch advertising usually runs at a higher ACoS than you can sustain, and that is acceptable if it is bounded. The question to answer is how much you are willing to lose in order to buy early sales, reviews and data.

The arithmetic is short. Suppose the price is $30 and you earn $9 per unit before ads, a 30% margin. You accept an ACoS of 45% during launch, so each ad-driven unit costs $13.50 in advertising, which is $4.50 more than the unit earns. If you want 300 units from ads, the planned investment is 300 × $4.50, or $1,350.

Treat that figure as a cap, not a forecast. Hold the line on the dollars, not on the ACoS. If auctions are pricier than expected, you get fewer units for the same cap, and you review. Write down the cap and the date it ends, whichever comes first. The cap also ignores the organic sales a good launch can produce, which is the whole reason to spend it, so its purpose is protection, not prediction.

Days 1 to 30: collect evidence cheaply

The first month is for learning what the market tells you about this product. Run a broad automatic campaign so Amazon can match you to queries you did not think of, plus a small number of manual campaigns on the handful of terms that describe the product most directly. Keep bids steady rather than chasing position; the purpose is data.

Watch three things. Are you getting impressions for the terms that matter? Does the main image earn clicks? Do the clicks turn into orders? A weak click-through rate points at the main image, title or price. A weak conversion rate, after enough clicks to be meaningful, points at the detail page. If both are healthy but spend is high, it is a bidding problem.

Reviews matter in this window too. Use Amazon's own Request a Review feature on eligible orders, and if the product qualifies, look into Amazon Vine. Never offer money, free product or any other incentive in return for a review. Amazon prohibits it, and the consequences can reach your whole account, which is far larger than any listing.

Days 31 to 60: tighten and expand

By now the search term report should show which real queries produce orders. Move proven winners into tightly controlled exact-match campaigns where you can set their bids directly, and add the clear losers as negatives. Raise bids only on terms that are earning, and only while total spend remains inside the launch cap.

This is also the phase to improve the page. If you are a brand-registered seller and have enough traffic, test one change at a time, starting with the main image or the title. Add Sponsored Brands creative or A+ content if you have not already, since both help shoppers who are comparing you with an established competitor.

Start tracking organic movement. Pick five or so queries that describe the product and check, weekly, where you appear for them. Note the share of orders that did not come from ads. You are looking for the first sign that sales are continuing at a lower relative spend.

Days 61 to 90: step down to a steady state

In the last phase, the launch budget winds down and profit targets take over. Aim ACoS toward your break-even point and then below it, using the margin figure you calculated before day one. Compare total sales per week with ad spend per week. If total sales are holding or rising while ad spend falls, the launch worked. If sales drop as the spend does, the product is still renting its visibility and you need to decide whether that rental is worth it.

Keep an eye on stock too. A successful launch burns through inventory faster than forecast, and a stockout in week ten undoes work from weeks one to nine. At the end of day 90, compare the actual spend with the cap and decide what happens next: continue, hold, or stop.

Mistakes that muddy the data

A launch is an experiment, and an experiment only teaches you something if you can tell what caused what.

• Changing the price, the main image and the title in the same week, so you cannot tell which change moved the result.
• Judging a campaign after two days. Orders can be credited to a click for some time after it happens, and early clicks are too few to say much.
• Letting promotions sit outside the cap. A coupon or a discount is money out, just like ad spend, and belongs in the same total.
• Adding dozens of keywords at once, which spreads the budget so thin that no term collects enough clicks to judge.
• Scaling spend because a single day went well.

Where to start

• Work out your profit before ads per unit, then calculate a launch cap with the arithmetic above.
• Write the end date next to the cap, and put a review on the calendar for days 30, 60 and 90.
• Complete the readiness checklist before any campaign goes live.
• Pick three numbers to watch in each phase and ignore the rest.
• Use only Amazon's own tools to earn reviews, and never offer anything in exchange.