Amazon SEO vs PPC: where sellers should spend first

Why this is a loop, not a choice

On Google, organic and paid are genuinely separate systems. Ranking well brings traffic; what happens after the click is largely your problem, and your ad performance does not move your organic position.

Amazon does not work that way, and the difference is the whole article.

Conversion feeds ranking. Amazon's job is to show shoppers what they will buy, so a listing that converts well on a term climbs on that term. That is true whether the sale came from an ad or from organic placement. Advertising therefore does not just buy traffic, it buys ranking inputs.

Advertising produces the data SEO needs. The search term report tells you the literal strings shoppers typed before clicking. Not a model, not a panel estimate — the words, with spend and conversions attached. That is the single best keyword source available to any seller, and it only exists if you are running ads.

SEO reduces what advertising costs. Each term you hold organically is a term you stop paying for, or pay less to defend. A mature account funds a smaller share of its revenue through ads not because its bidding improved but because its listings earned position.

So the sequence runs: listing quality → ads find the converting terms → those terms go into the listing → organic rank rises → ad dependence falls → budget frees up for the next product. Starting anywhere other than listing quality breaks the loop at the first step.

Which is why the usual framing is wrong

Most comparisons present this as a trade-off: SEO is slow and cheap, PPC is fast and expensive, pick according to your patience and your bank balance. That framing is imported from Google, where the two systems really are independent, and on Amazon it produces bad decisions in both directions.

A seller who "chooses SEO" because the budget is tight ends up with an optimised listing nobody sees, because without early sales there is no rank and without rank there are no sales. A seller who "chooses PPC" because they want speed ends up renting a position indefinitely, paying the same acquisition cost in year three as in month one, with nothing accumulating underneath.

Neither is a strategy. They are the two failure modes of treating a loop as a fork.

What each one actually does

The row worth staring at is the last one. Both disciplines have a characteristic way of looking healthy while doing nothing, and they are different failures. SEO work that nobody measures produces a nicer-reading listing and no rank movement. Advertising that nobody segments produces a respectable blended ACoS carried entirely by people searching your brand name, while the campaigns meant to acquire new customers quietly lose money underneath.

There is more on that second failure in our piece on ACoS, TACoS and what a good one looks like, and on the first in the keyword research guide.

Why the listing genuinely comes first

This is the part sellers most want to skip, because ads feel like action and listing work feels like admin. The arithmetic is unkind to that instinct.

Take a $30 product at a 10% conversion rate and a $1.20 cost per click. Ten clicks produce one sale, so you spent $12 to make $30. Now take the same product converting at 15% after the listing is fixed. Under seven clicks produce a sale, so you spent about $8. Nothing about the campaign changed. Your cost per sale fell by a third because the page got better at its job.

And that improvement does not stop at advertising. The same conversion rate feeds organic ranking, so the listing work pays twice: once in cheaper ads, once in position you stop paying for.

Run it the other way and the damage compounds the same way. Ads to a weak listing buy expensive bounces, and those bounces are a signal to Amazon that shoppers searching that term do not want your product — which suppresses the organic position you were trying to build.

What "fix the listing" means in 2026

Specifically, and in this order:

• Main image. The single biggest lever on click-through rate, and the cheapest to fix. It is what decides whether your ad spend buys a click at all.
• Title, under the new limit. Amazon cut titles to 75 characters in all categories except media on 27 July 2026, with a new 125-character Item Highlights field displaying beneath it from 10 August. If your title has not been rewritten since, Amazon may have rewritten it for you — brand owners had 14 days to review AI-generated recommendations. Check what yours says before you plan anything.
• Price against the page. Not against your costs. If you are the most expensive result with the fewest reviews, no amount of bidding fixes that.
• Bullets and Item Highlights. Objections answered in sentences, not keyword strings.
• A+ content. A conversion asset rather than a ranking one, and worth doing after the above.

Our listing work runs in that dependency order for this reason, and the free listing checker will score a live ASIN against most of it in about a minute.

The split, by stage

Concrete numbers, because "do both" is not a plan. These are allocation ratios rather than absolute budgets — scale them to what you have.

Two notes on reading that table.

The launch row inverts on purpose. For thirty days you are buying sales velocity and relevance, not profit, and the listing work should already be done before you start — which is why pre-launch is 100% the other way. There is more on that sequence in our launch playbook.

The stalled row is the one most accounts are actually in, and the one most often handled by increasing ad spend. If your advertising is up and your total revenue is not, more bidding buys more of the same. Something on the page or in the price is the constraint.

Splitting a first $1,000

If you have one product and a thousand dollars, spend roughly $600 on getting the listing right — photography first, then copy under the 75-character limit, then a proper keyword set — and put $400 into exact-match Sponsored Products across fifteen to twenty terms you believe in. Three or four weeks later you will know which of those terms convert. That is worth more than either half alone, because the $400 has bought data as well as sales and the $600 has made the data cheaper to collect.

How to actually run the loop

The handover between the two disciplines is where most accounts leak, so here is the mechanism in detail.

Weekly: harvest. Open the search term report. Any term with conversions at an acceptable cost moves into an exact-match campaign of its own, where you can bid it properly. Any term with meaningful spend and no conversions becomes a negative, in the campaign and at the ad group above it. Fifteen minutes, done properly, is worth more than most optimisation.

Monthly: promote. Terms that have converted consistently for four weeks have earned a place in the listing. This is the moment PPC data becomes SEO work, and it is the step almost nobody schedules. Put them in the right field — the one or two that describe what the product fundamentally is compete for title characters, attributes go to Item Highlights, synonyms and misspellings to the backend.

Monthly: check the rank you bought. For each term you promoted last month, check organic position. If organic rank is climbing on terms you advertise heavily, the loop is working and you can start reducing bids on those terms specifically. If it is not, the listing is not converting well enough on them and advertising harder will not fix it.

Quarterly: recalculate. Break-even moves as fees and cost of goods move, and the whole allocation above is downstream of it. Amazon raised FBA fees by an average of $0.08 per unit from 15 January 2026, which is small on a $40 product and material on a $9 one.

The three mistakes that break the loop

Pausing keywords you rank for organically. A term where you hold organic position and also advertise looks like paying for traffic you would have had anyway. Sometimes it is. But pausing it can cost the organic position too, because the total sales velocity on that term falls. Reduce the bid, watch rank for a fortnight, then decide.

Judging organic work on ACoS. ACoS cannot see organic sales at all — they are not in its denominator. Successful SEO work will often make ACoS look flat or worse while total revenue grows. Watch TACoS instead.

Running ads with no negatives. Automatic campaigns drift into adjacent, poorly-converting queries within weeks. An unattended auto campaign is the most reliable way to pay Amazon for traffic that damages your ranking.

Which one is your actual problem?

Before allocating anything, work out which half of the funnel is broken. The diagnosis takes ten minutes in Search Query Performance and it decides everything downstream.

That report shows, for each query your ASINs appeared on, the whole funnel: impressions, clicks, adds to cart, purchases, and your share at every step. Read it by looking for where the drop happens.

Only the last row is an advertising problem. The first three are all listing or keyword work, and in every one of them extra ad spend makes the economics worse rather than better — you buy more impressions that do not click, or more clicks that do not convert, at exactly the point where conversion is already the constraint.

The pattern behind most stalled accounts

A seller notices flat revenue, increases the advertising budget, sees sales rise slightly and ACoS rise more, concludes the category has got harder, and increases the budget again. Three months later spend is up 40%, revenue is up 8%, and the conclusion is that Amazon is expensive.

Usually the second row of that table was true the whole time. Click share was low because a competitor refreshed their main image or dropped their price, and no amount of bidding changes what happens on a results page where your thumbnail is the least appealing option. The fix cost a photography day and would have improved the ad economics and the organic position simultaneously.

The general rule: advertising is a multiplier on the listing, not a substitute for it. Multiply a bad number by a bigger budget and you get a bigger bad number.

The straight answers

To the questions this actually comes down to.

Can you rank on Amazon without PPC? Yes, and it is slow. Products in thin categories with a strong differentiator and some external traffic do it. In any competitive category, launching without ads means waiting for organic sales that will not come, because you have no rank, and you have no rank because you have no sales. Advertising is how you break that circle.

Does PPC directly improve organic ranking? Not directly, and anyone who says otherwise is overclaiming. What happens is indirect and real: ads produce sales, sales produce velocity and conversion history on those terms, and those are ranking inputs. The distinction matters because it tells you that ads on a listing that does not convert will not build rank — they will do the opposite.

How long does Amazon SEO take? Indexing within 24 to 72 hours of a change. Meaningful rank movement in two to six weeks, if the listing converts. If nothing has moved in eight weeks, the problem is usually relevance or conversion rather than patience.

Should I pause ads once I rank organically? Reduce rather than pause, and do it one term at a time while watching organic position. Competitors bid on terms you own, and a paid slot above your organic listing is often cheap insurance.

Should I bid on my own brand name? Usually yes, and measure it separately from everything else. The clicks are cheap and convert well, and the alternative is a competitor occupying the top of your own branded search. But a blended ACoS carried by branded traffic hides what your acquisition campaigns are really doing, so never report the two together.

What if I can only do one? Then do the listing. An excellent listing with no ads will sell something. Ads pointed at a poor listing will sell very little and cost you money doing it. The asymmetry is the whole argument: listing work improves organic position and advertising economics simultaneously, while advertising improves neither if the page is the constraint.

If you want an outside read on which stage of that table your account is actually in, a free written audit comes back within 48 hours, and it will say plainly whether your problem is the page or the campaigns.

Questions people also ask