Sponsored Products is where the first Amazon ad dollar goes, because it converts search demand that already exists, at the highest purchase intent and with the clearest attribution of the three ad types. Sponsored Brands and Sponsored Display are expansion plays: they earn budget after Sponsored Products is efficient, and each has to be matched to a specific job (brand defense, multi-product discovery, retargeting, conquesting), not switched on to feel diversified. That is the whole allocation answer, and the rest of this guide is the reasoning and the sequencing. It also carries a warning I end up giving on a lot of diagnosis calls: a leaking Sponsored Products account does not get fixed by adding two more ad types. It becomes three leaks. Here is what each ad type actually does, when the second and third earn their place, and how I allocate across them by phase.
The three Amazon ad types, compared
Amazon sellers have three ad types. Sponsored Products places keyword-targeted and product-targeted ads in search results and on detail pages, and any professional seller can run it. Sponsored Brands places headline banners and video at the top of search and requires Brand Registry. Sponsored Display targets audiences and detail pages, on and off Amazon, and also requires Brand Registry.
| Ad type | Where it shows | The job it does | Funnel stage | Who can run it | Attribution clarity |
|---|---|---|---|---|---|
| Sponsored Products | Search results and product detail pages, styled like organic listings | Converts shoppers already searching for what you sell | Bottom of funnel | All professional sellers | Clearest: click to order, term by term |
| Sponsored Brands | Headline banners and video, most prominently at the top of search, driving to a Store or product collection | Brand defense, headline real estate, multi-product discovery | Mid funnel, plus defense at the bottom | Brand Registered sellers | Murkier: new-to-brand framings flatter the read |
| Sponsored Display | Detail pages and audience placements on and off Amazon | Retargeting warm shoppers, conquesting competitor pages, defending your own | Mid funnel and re-engagement | Brand Registered sellers | Loosest: view-through credit inflates apparent results |
Two columns in that table do most of the deciding. The funnel-stage column tells you which job each type can actually be hired for. The attribution column tells you how skeptically to read each type's own scoreboard, and it is no accident that the ad type with the clearest attribution is also the one I fund first.
Why Sponsored Products earns the first ad dollar
Sponsored Products earns the first dollar because the demand it converts already exists. A shopper typed a search, which is the strongest purchase signal Amazon has; your ad answers that search; and the order that follows is attributed to a click you can see in a report. No other ad type combines that intent with that clarity.
Break that into its three parts. Intent: a Sponsored Products impression only exists because someone is actively shopping the query or the detail page it appears on, so you are paying to be present at the moment of decision, not to create interest. Attribution: the click-to-order chain feeds the search term report, and that report is the raw material for every optimization that matters: negatives, harvests, bid repricing. When attribution is clean, the account teaches you something every week. Availability: it is the one ad type every seller can run from day one, no Brand Registry required.
This is also why nearly every other guide in this cluster is about Sponsored Products. The discovery-to-performance structure, the negative workflow, the bid formula: they all live inside SP, because SP is where the measurable, repeatable money is. You do not build awareness campaigns before you can profitably convert a search that describes your product. Awareness for a product that cannot close a high-intent click is just expensive rehearsal.
When Sponsored Brands earns its place
Sponsored Brands earns budget once two conditions hold: your Sponsored Products spend is converting at your phase target, and you have Brand Registry. Then it gets hired for named jobs: defending your branded search from competitor headlines, taking video real estate on core terms, and driving shoppers into a Store for multi-product discovery.
Brand defense first. The headline slot above your own branded search results exists whether or not you buy it, and a competitor with Brand Registry can buy it the day you do not. A shopper who typed your brand name is the cheapest conversion you will ever get; a rival banner sitting above your results is a toll gate on your own traffic. Owning that slot is insurance more than growth, and I treat the spend that way. It carries the same nuance as any branded bidding: the ad is partly redundant, because many of those shoppers would have found you anyway, and partly defensive, because the slot does not disappear when you vacate it. I walk through that redundant-versus-defensive test in my guide on what happens when you turn ads off, and it applies to Sponsored Brands defense dollar for dollar.
Then discovery, if the catalog supports it. Sponsored Brands is the only ad type that sends a click to a Store or a product collection instead of a single listing, which matters when you sell a range: one click can shop the catalog. The video format buys attention in the middle of the results page in a way a standard tile cannot. Both are legitimate jobs, and both should clear the same bar as any other spend: orders against dollars, per term. And if you are Brand Registered at launch, a Sponsored Brands flight on the handful of terms you must own can add early impressions while your Sponsored Products campaigns build history. Useful, but optional.
The honest caveat: the metrics flatter. Sponsored Brands reporting leans on new-to-brand percentages and brand-halo framings that make the spend look better than its incremental order count. In the accounts I audit, an SB campaign that looks great in its own dashboard and mediocre when you count plain orders against plain spend is a common finding. Read SB with the same cold arithmetic you apply to everything else.
When Sponsored Display earns its place
Sponsored Display earns budget for three jobs: retargeting shoppers who viewed your listing but did not buy, conquesting competitor detail pages (placing your ad directly on a rival's listing), and defending your own detail pages from competitors doing the same to you. It requires Brand Registry, and it demands the tightest monitoring of the three types.
Retargeting is the strongest case. Most detail-page views do not end in an order on that visit, and a shopper who viewed your listing and left is the warmest audience you can buy: they already found you, already considered you, and simply did not close. Sponsored Display is the only one of the three ad types that can follow that shopper off Amazon and bring them back.
Conquesting should follow evidence, not ambition. Your auto campaigns already run a free version of this experiment: the substitutes targeting group places you on competitor detail pages and reports which ones actually send converting traffic. Sponsored Display conquesting is the deliberate, scaled version of that placement, so let the substitutes data nominate the targets before you pay to sit on a competitor's page on purpose. The mirror image is defense: the ad slots on your own detail page will carry someone's product, and Sponsored Display is how you make it yours instead of a rival's.
The honest caveat: view-through attribution. Sponsored Display can claim credit for orders from shoppers who merely saw the ad and never clicked it. That is the loosest attribution standard of the three ad types, and it reliably makes SD look better than it is: some of those shoppers were coming back anyway. My rule is to judge Sponsored Display on click-attributed orders and treat view-through credit as decoration. If a Sponsored Display campaign is only profitable with view-through included, it is not profitable.
How to allocate across the three ad types by phase
Allocation follows phase and job, not a fixed percentage split. In every phase, Sponsored Products takes the majority of spend because it converts existing demand. Sponsored Brands and Sponsored Display scale up only as their jobs become real: there is nobody to retarget at launch, and there is no brand to defend before shoppers search your name.
| Phase | Sponsored Products | Sponsored Brands | Sponsored Display |
|---|---|---|---|
| Awareness (launch) | Carries the launch: discovery campaigns plus exact match on the terms you must own | Optional launch-velocity flight if Brand Registered; skip otherwise | Rarely: there is no warm audience to retarget yet |
| Market share | Core of spend: scale the proven exact match winners | Brand defense on your branded search as your name starts getting searched | Begin retargeting as viewer volume builds; conquest only where substitutes data points |
| Growth | Still the majority, funded to its productive appetite | Defense holds; add Store-driving and video where the catalog supports discovery | Retargeting the growing viewer pool; conquest where click-attributed data earns it |
| Profitability | Efficient and pruned, converting at the phase target | Selective: defense stays, discovery spend clears the same ACOS bar as everything else | Disciplined and small, judged on click-attributed orders only |
The phase definitions and the ACOS targets that go with them are the subject of my phase benchmarks guide; the point here is that the mix shifts with the account's job, not with the calendar. The sequencing rule I run is four steps:
- Get Sponsored Products efficient first. Phase-target ACOS, zero-order terms negated, bleed fixed, bids repriced. This is the foundation everything else stands on.
- Add Sponsored Brands for defense once Brand Registered. Branded-search defense first; Store-driving and video second, where the catalog justifies them.
- Add Sponsored Display for retargeting, then conquest. Retargeting is the strongest case; conquest follows the substitutes evidence. Click-attributed orders are the scoreboard.
- Rebalance by phase. Hold SP as the majority, keep the defense, and make every discovery dollar in SB and SD clear the same ACOS bar as the rest of the account.
The diversification trap: three ad types, three leaks
Do not spread budget across three ad types to feel diversified before Sponsored Products is optimized. Ad types are not uncorrelated assets: they draw on the same budget, chase the same shoppers, and inherit the same listing. Adding Sponsored Brands and Sponsored Display to a leaking Sponsored Products account produces three leaks, not a portfolio.
The pattern shows up in audits often enough that I can describe it from memory. A seller reads that a diversified ad mix is best practice, switches on all three types, and six months later the Sponsored Products campaigns still carry zero-order spend nobody negated while the SB and SD campaigns quietly underperform on top of it. The account did not get more resilient. It got harder to read, because now the waste is spread across three consoles instead of one.
The order of operations is the whole cure. Fix Sponsored Products first: run the Wasted Spend Finder on a 60-day file to see the zero-order spend, work the five structural levers for lowering ACOS, and apply the raise-or-fix diagnostic to any campaign hitting its cap. Then expansion funds itself: the dollars you reclaim from SP waste become the Sponsored Brands defense budget and the Sponsored Display retargeting test, no new money required. If you want to see what that SP-first read looks like on real rows, the audit demo walks it on a real anonymized account. Diversification is not a strategy. It is what you get to do after the core works.
Frequently asked questions
Should I run Sponsored Brands or just Sponsored Products?
Run Sponsored Brands only after Sponsored Products is efficient, and only with a named job for it: defending your branded search, or driving multi-product discovery through your Store. Sponsored Products converts existing search demand at the highest intent, so it carries the account either way. If your Sponsored Products spend still leaks, adding Sponsored Brands spreads budget without fixing anything.
Is Sponsored Display worth it?
It can be, for two specific jobs: retargeting shoppers who viewed your listing but did not buy, and placing your ad on competitor detail pages. The catch is attribution. Sponsored Display can take credit for views, not just clicks, which inflates its apparent performance. Judge it on click-attributed orders, and if it only looks profitable with view-through credit included, it is not earning its budget.
Sponsored Products vs Sponsored Brands vs Sponsored Display: which should I start with?
Sponsored Products, without exception. It is the only one of the three every seller can run, it converts shoppers who are already searching for what you sell, and its click-to-order attribution gives you clean data to optimize from. Sponsored Brands and Sponsored Display both require Brand Registry, and both do their best work on top of a Sponsored Products foundation that already converts profitably.
Do I need Brand Registry to run Amazon ads?
No for Sponsored Products, which any seller with a professional account can run. Yes for Sponsored Brands and Sponsored Display, which are both restricted to Brand Registered sellers. In practice this settles the question for unregistered sellers: run Sponsored Products, and lose nothing important by waiting, since that is where the first ad dollar belongs anyway. Enroll in Brand Registry when you can; it also unlocks your Store and A+ content.
Should I bid on my own brand name with Sponsored Brands?
A branded Sponsored Brands ad is partly redundant and partly defensive, and no report splits the two for you. Redundant, because a shopper searching your brand name was mostly going to find you anyway. Defensive, because the headline slot above your own results exists, and a competitor can buy it the day you leave it open. If competitors are actively bidding on your brand, defend the slot. If nobody is contesting it, test a taper and judge total sales, not ad sales.
How should I split my budget across the three ad types?
There is no universal percentage split, and anyone quoting one is guessing about your account. Allocate by job and phase instead. Sponsored Products takes the large majority of spend in every phase, because it converts existing demand. Sponsored Brands gets what brand defense and Store discovery genuinely need. Sponsored Display gets a tightly monitored allocation for retargeting and conquesting, judged on click-attributed orders. Rebalance from your own performance data as the account moves between phases.
Grade the foundation before you expand
Every expansion decision in this guide starts from the same question: is Sponsored Products actually efficient yet? You can answer it in about a minute. Upload a 60-day bulk file to the free Account Health Snapshot and it grades the SP foundation across nine metrics: in your browser, no email, no account. The Audit Dashboard then shows the specific rows behind any failing grade. And if you want a second set of eyes on the sequencing, book the free 30-minute diagnosis call: I will tell you where your next dollar's best job is, whether that is a Sponsored Brands defense campaign or a fix inside Sponsored Products first, whether or not you ever hire me.