Launching a Product Without Burning the Budget

A launch is a chicken and egg problem stated in public. Nothing ranks without sales, sales need reviews, and reviews come from sales. Every launch tactic that has ever existed is an attempt to get through that opening.

What follows is the order that works, and the order matters more than any single tactic in it.

The listing has to be finished. Not adequate: finished. Every image, the full title, the bullets, the A+ content if you have Brand Registry, and the backend fields filled in.

This is not a preference. Advertising a half-finished page means paying for clicks that were never going to convert, and the conversion rate you record in the first fortnight is the one Amazon uses to decide whether to show you organically.

Stock cover matters as much. A launch that runs out in week three loses the ranking it just paid for, and getting it back costs more than holding the stock would have.

Running out in the middle of a launch undoes it: the rank the advertising bought falls away while the listing is unavailable, and the relaunch costs nearly as much as the first. The stock plan is part of the launch plan.

Enough for the launch and the gap after it. Plan for the units the launch is meant to sell over its first two or three months, plus the weeks a reorder takes to arrive and become sellable. A launch that succeeds faster than expected is the commonest way to run out.

Inside the capacity limit. Amazon sets each account's FBA capacity monthly, and a launch shipment has to fit within it alongside everything else the account sells. Check the limit before the order is placed with the factory, not when the shipment is ready. FBA capacity limits explains how it is set.

Not too thin, either. Amazon charges a low inventory level fee on products whose stock stays low against their sales, judged for each variation since January 2026. New products are exempt for a time, but a launch that keeps its stock tight to save cash can meet the fee once that period ends.

A reserve outside Amazon. Holding some stock at the seller's own or a third party warehouse, ready to send in, protects the launch from one late shipment without tying up capacity in Amazon's network.

Vine is the sanctioned route, available to registered brands. You give units to Amazon's reviewer programme and get honest reviews, which means a weak product gets honestly reviewed as a weak one. That is a feature: better to learn it now.

The request a review button on every order, which is the compliant way to ask. Automating it through an approved tool is fine; writing your own message asking for a positive review is not.

What does not work, and costs accounts: paid reviews, review groups, inserts offering anything in exchange for a review, and asking to be contacted before a negative one is left. All of it is against policy, all of it is detectable, and the penalty lands on the account rather than the product.

A new listing is the one place a seller can get the title right from the first day, and since July 2026 the rules for it are tighter than most launch guides still assume.

Seventy five characters. Since 27 July 2026, titles outside media are limited to 75 characters, spaces and brand included, and Amazon rewrites titles that run over. A launch listing written to the old, longer limit will be rewritten in Amazon's words, possibly in the middle of the launch.

What goes in. The brand, what the product is and the one attribute a buyer chooses by. The terms a launch most needs to rank for belong at the front, because the title is the field sellers widely find carries the most weight for relevance.

Item Highlights for the rest. The same change added Item Highlights, up to 125 characters that are searchable and shown with the title. The materials, use cases and secondary terms that no longer fit the title go there, so the launch listing is indexed for them from the first day.

Check it before advertising. Run the indexing test for the main terms once the listing is live and before the launch campaigns start, so the advertising is not paying to promote a listing that cannot be found for its own terms.

The detail is in Amazon's 75 character title limit and Item Highlights.

Start on exact match against the handful of terms the product genuinely is. Broad and automatic campaigns are for discovery later, when there is enough data for the discovery to mean something.

Expect the cost per sale to be bad. A launch buys ranking rather than profit, and treating the first fortnight's ACOS as a verdict on the product is the most common way a launch is killed before it starts.

What matters is the trend across weeks: cost per sale falling and organic sales appearing beside the paid ones. If neither happens by the end of the first month, the problem is the listing or the price rather than the bids.

Lower than the target, and not by so much that returning to the target kills the conversion rate. A launch price is an investment in early sales velocity, and the plan for coming off it should exist before it starts.

Coupons and a discount badge do more work than the same money taken off the price, because the badge is visible in search results and the price alone is not.

Before the stock lands. The listing finished and checked on a phone, since most shoppers will see it there. Vine enrolled if the product qualifies, so the first reviewers receive units as soon as they are available. Exact match campaigns built and paused, a coupon set up, and the plan for the launch price written down, including the date and the steps for coming off it.

Week one. Turn the campaigns on as the stock goes live and request a review on every order. Read sessions and conversion daily this week only, to catch the obvious faults: a suppressed listing, a broken variation, an image that did not load.

Week two. Read the first search term report. Add the irrelevant terms as negatives, and put more budget behind the terms that converted. Read every review as it arrives, and fix in the listing anything a reviewer had to find out by buying.

Week three. Check organic rank on the main term. If it is moving, the exact match campaigns are doing their job, and a broad or automatic campaign can start, at a low bid, to find the terms you have not thought of.

Week four. The verdict, from the trend rather than the totals. Cost per sale falling and organic sales appearing beside the paid ones means the launch is working and the price can start coming up. Neither happening means the listing or the price, and the bids are the last thing to change.

Months two and three. Holding what the first month bought: stock that never runs out, the price stepping up to its target, the terms the broad campaigns found moved into exact match with bids of their own, and every recurring point in the reviews answered by changing the listing or the product rather than the reply.

Sessions, conversion rate, buy box share and organic rank on your main term. Those four say which of the four possible problems you have: nobody arrives, they arrive and do not buy, they would buy and you do not have the button, or it is working and needs time.

The same four questions decide a launch on Walmart, where competition is thinner and the opening is often cheaper, which is worth knowing if the Amazon category you are entering is expensive.

Reading them in that order stops the usual mistake, which is changing bids when the conversion rate is the problem.

The launch price is only an investment if it ends, and the way it ends decides whether the rank it bought survives.

In steps, not at once. Two or three rises a couple of weeks apart, with conversion read after each. A single jump to the target price is the change most likely to show up as a falling conversion rate, and rank follows conversion.

One change at a time. Raising the price in the same week as cutting the advertising, or ending the coupon, leaves no way to tell which change did what. Keep the coupon running while the list price rises, and reduce it last.

Know when to pause. If conversion drops sharply after a step and stays down for a week, hold the price where it is and look at what the page is competing with at that price. A competitor sitting just below the new price is a reason to stop at the step before, not to give up the plan. Hold there for a fortnight, and try the step again once the page has more reviews behind it.

Not every launch works, and the expensive ones are those that carry on for months on hope. Deciding in advance what failure looks like makes it possible to stop, fix or change course while there is still budget to do so.

Set the limits before starting. A budget for the launch period, a date to review it, and the numbers that count as on track by then: sessions, conversion rate and organic rank on the main terms. Without them, every week looks like it might be the turning point.

Read which problem it is. Few sessions means the listing is not being found or shown: relevance, indexing or advertising. Plenty of sessions and few sales means the listing or the price is not persuading: images, reviews, the offer against competitors. Sales that turn into returns or poor reviews means the product itself.

Fix the cause, not the budget. More advertising helps the first problem, and makes the second and third more expensive. A listing that does not convert needs better images, a clearer offer or more reviews before more traffic.

Know when to stop. A product whose reviews keep naming the same fault, or whose market is held by competitors it cannot match on price or quality, is not rescued by persistence. Stopping early, clearing the stock and putting the budget behind the next product is a result too, and usually a cheaper one.

Launches are the part of Amazon account management where sequence matters most, because almost everything expensive about a bad launch is caused by doing the right things in the wrong order.

The photography and A+ content are usually the constraint rather than the advertising, which is why launch work and design work get planned together rather than one after the other.

If you have stock arriving and no plan for the first month, that is the conversation to have before it lands.