Harpy Guide

Sponsored Brands, Done Properly

14 min read · Harpy Media

The complete Sponsored Brands system: format choices, modified broad match, single-creative ad groups, keyword-dense headlines, video hooks, and placement bidding after the 2026 rule change.

Part 1

The three formats

Sponsored Brands are the most expensive ad type on Amazon — and in most accounts we audit, they're either bleeding money with terrible creative or barely spending because months of poor performance squeezed the bids toward zero. Same root cause both times: nobody taught the seller how to actually use them. They are your storefront billboard on the most competitive retail platform on earth; treat them accordingly.

FormatWhat it isVerdict
Product CollectionCustom lifestyle image + up to 3 featured products with stars, review counts, Prime badges. Drives to store or landing pageThe workhorse — most of your budget lives here
VideoVideo ad to a product page or store, horizontal or verticalHighest CTR potential — and the most dangerous when done badly
Store SpotlightFeatures three store sub-pagesConsistently underperforms Product Collection — no product-level reviews visible. Use selectively, for brand-building
💡 Tip. If you could run one format forever: Product Collection. It combines brand storytelling (lifestyle image, headline) with the conversion signals shoppers need (reviews, ratings, Prime badge), at manageable CPCs. Video is powerful but can price you out.

Where the ads actually show

The primary placement is top of search — the banner above all the sponsored product and organic results. That's the endcap of the aisle. Sponsored Brands also appear in rest-of-search, on product detail pages (including inside the reviews section), and experimentally on the homepage. The critical mechanic: non-top-of-search placements compress your creative — your lifestyle image and headline shrink to a tiny logo and brand name. Top of search is the only placement that renders your creative as designed, which is why it carries the strategy.

For video specifically: choose the Store as landing page to compete for the top-of-search video slot; choose a product detail page and the video runs in the in-search placement mid-page. Run both at once to saturate one results page with your brand.

Amazon now supports up to ten creatives per campaign and nudges you with a “creative strength” indicator. Use the capacity for testing — but as Part 3 explains, real testing needs isolation, not volume.

Part 2

Targeting rules

Sponsored Brands targeting mirrors Sponsored Products — keywords, ASIN/product targeting, broad/phrase/exact — with two critical differences that cost real money if missed.

Difference 1 — broad match is wider here

Sponsored Brands broad match opens onto synonyms, translations, and loosely related terms — broad enough that “red colored pens” can match “blue writing utensils.” The fix is modified broad match: a plus sign before each word (+red +colored +pens) requires every word to appear in the search, in any order.

⚠️ Watch out. Never run standard broad match in Sponsored Brands. The match type is too wide, the CPCs are the most expensive on the platform, and the waste is enormous. Modified broad, phrase, or exact — nothing else.

Difference 2 — the trap option

Amazon offers “keywords related to your brand” as a targeting option. Avoid it completely. It behaves like an auto campaign for the most expensive ad type — tests of it have produced the worst conversion rates and ACOS levels several times beyond salvage. Modified broad research with your own guaranteed minimum relevancy beats it every time.

Part 3

Structure — one creative per ad group

The foundational structural decision for Sponsored Brands: single creative ad groups (SCAGs). One campaign per keyword theme, multiple ad groups, one creative per ad group.

1 campaign (keyword theme) → many ad groups → 1 creative eachThe creative = headline + image + product selection — test them in isolation

Why isolation matters: put multiple creatives in one ad group and Amazon picks a favorite almost immediately. It doesn't rotate fairly or A/B test — it assigns nearly all impressions to one creative, often not the best one, and you're left with no data on the rest and no way to compare. Separate ad groups give you even impression distribution (each ad group competes independently for the same keyword), keyword-to-creative correlation, manual control to pause and promote, and clean attribution when performance moves.

Change one variable at a time

The three modifiable variables: headline (including the CTA), featured image, and ASIN selection. A valid test changes exactly one. If Creative A runs “Discover” with a close-up image and Creative B runs “Shop” with a wide-angle, and B wins — you've learned nothing about which change mattered.

💡 Tip. Name ad groups with their variables: “Discover | Close-up | B0xxx B0yyy B0zzz” versus “Shop Now | Close-up | same ASINs.” Across dozens of ad groups, the test you're running should be readable from the name alone.
🔑 Key insight. Most sellers and agencies are not testing their Sponsored Brands creatives at all. That's the edge. As CPCs climb, the advertisers winning top-of-search at reasonable cost are the ones whose tested creatives earn higher CTR and conversion — which Amazon rewards with a better quality score, lower CPCs, and more impressions. Testing is how you buy discount clicks.

Part 4

Headlines that match the search

The manager often can't control the imagery — you work with the brand's assets. The headline is yours. It's also one of the highest-leverage changes available: headline changes alone routinely move CTR by 30–50% and pull CPCs down with it.

Two approaches — use both at once

Creative headlines (wit, emotion, story) capture attention; keyword-dense headlines match the shopper's exact search. The best headline does both. The single-keyword-theme strategy: tailor the headline to the search itself. Search “fasting electrolytes,” see “Don't Break Your Fast — Zero Calorie Electrolytes.” Search “hangover hydration packets,” see “Last Night Was a Mistake. Hydrate Next Time.” When the headline mirrors the query, both the algorithm (which appears to index headlines for relevancy) and the human (who instantly recognizes “exactly what I was looking for”) reward you.

These are keyword theme campaigns, not single keyword campaigns — you target a theme with phrase and modified broad, and craft everything (headline, image, landing page) around it.

💡 Tip. Always include a call to action. “iPhone Cases from PhonePros” is a label; “Discover iPhone Cases from PhonePros” is an invitation. In every test, CTA headlines outperform CTA-less ones — the action verb tells the shopper what happens next.

Your customers write better copy than you

Conversion copywriters have found it repeatedly: headlines pulled from customer review language outperform headlines written from scratch, even by professionals — because review language comes with the passion, specificity, and emotional honesty of real experience. Mine your reviews and your competitors' reviews for phrases that capture the value in the customer's own words. If a review says “so sharp I don't need shaving cream,” the headline writes itself.

Single-keyword landing pages

Take the theme idea one step further: build dedicated store sub-pages as landing pages per theme. Someone who clicks “Head Shavers for Bald Men” should land on a page where the imagery, messaging, and product selection are all about that intent — not your generic store homepage. Hide the sub-pages from store navigation; they exist purely as ad destinations. When one product serves multiple contexts, build a page per context and let shoppers see the product in their own room.

Part 5

Branding psychology and lifestyle images

Every competitive category on Amazon is commoditizing: dozens of functionally identical products at similar prices with similar reviews. The winners in a commoditized market aren't the best specs — they're the strongest identity. Water is the most commoditized product on earth, and one brand built a multi-billion-dollar business on it by innovating the positioning, not the water.

Know the customer before writing a word

The biggest creative mistake is writing about how great your product is. Customers don't care about your product — they care about their problem. Understand the emotional driver behind the search, and make the end user the hero of the ad, not the product. Research sources: your reviews (the most honest feedback loop), social spaces where your customers actually talk, competitor reviews (complaints and wishes are your differentiation), and AI tools for drafting customer avatars worth verifying.

Pick your lane

DifferentiatorExample
PriceHard for small brands to defend — usually the wrong lane
Design / beautyProducts that look and feel premium
Technology / innovationA genuinely novel feature or approach
QualityBuilt to last, premium materials
ConvenienceEasy to buy, use, maintain
SustainabilityMaterials and shipping that align with values
Community / identityA tribe customers belong to by buying
🔑 Key insight. Lean hard into one or two differentiators and customers forgive weaknesses in the others — that's why premium design beats premium pricing arguments, and why values-aligned buyers accept performance trade-offs. Your Sponsored Brands creative should reinforce your chosen lane in every element: headline, image, product selection, landing page.

Lifestyle images and the eye-direction trick

Always use lifestyle imagery — tested extensively, it consistently outperforms product-on-white in CTR. And a specific mechanic worth knowing: the subject's gaze steers the shopper's gaze. Eye-tracking studies show a model looking at the camera draws eyes to the model; a model looking at the product or headline draws eyes there. Point your model's attention where you want the shopper's.

Part 6

Video that actually converts

The uncomfortable truth about Amazon video ads: most are terrible — recycled listing images in a slideshow, stock footage, bullet points with transitions. Animated brochures. The bar is on the floor, which means modest effort produces outsized results. A video is defined by performance, not production: a shaky phone video that converts beats a cinematic piece nobody clicks.

Match the video to the funnel stage

StageThe video's jobWhere it belongs
AwarenessBrand story, lifestyle, entertainmentOff-Amazon platforms — not in search placements
ConsiderationValue proposition, key features, hook fastSponsored Brands video in search results
ConversionTestimonials, detailed demos, final pushProduct page video
⚠️ Watch out. The most common video mistake: one beautiful awareness-style video dumped into every placement — search, product pages, SB video. That's a Super Bowl commercial on a shelf tag. Search placements need the value proposition in the first two seconds; product pages need the closing argument.

The first two seconds

Shoppers scrolling results give your autoplaying video about two seconds. Bad hooks: product shot with logo, slow-motion unboxing, “easy to install.” Good hooks: the dramatic result, the problem being solved visually, a bold text overlay matching the search intent — “Replace Your Jumper Cables. No Help Needed.” Lead with the result the shopper came for, not the introduction.

Vertical video — and how to get it on CPC

Vertical video fills the mobile screen natively, and tests show CTR at multiples of horizontal — partly format, partly low adoption. The catch: vertical is associated with the vCPM cost model. The workaround: select Store as the landing page when creating the campaign — that unlocks the top-of-search placement and, in the creative section, an orientation dropdown where you can choose Vertical on CPC bidding. Run horizontal and vertical in separate ad groups and let them compete.

💡 Tip. Every SB video autoplays muted. Most shoppers watch the whole thing with no sound, which makes voiceover invisible. Treat video as silent by default: subtitles and on-screen text carry the selling points and the CTA. This one change meaningfully lifts engagement.

Part 7

Placements and bidding

Sponsored Products let you raise bids per placement. Sponsored Brands historically worked in reverse — your base bid was your top-of-search bid, with negative modifiers to discount everything else. That era is over: as of February 2026, Amazon deprecated negative placement adjustments for Sponsored Brands. You can only bid up from base, exactly like Sponsored Products now.

Base bid = what your weakest placement can afford  →  positive modifiers bid up where performance justifiesThe post-February-2026 Sponsored Brands bidding logic

So the approach flips: set the base bid conservatively — aligned to your lowest-value placement — then increase for top of search based on how far its revenue-per-click outruns the rest. A common landing zone is a 50–100% top-of-search modifier from real placement data, refined weekly. And for granular control, the bulk file (multi-ad-group format) still lets you set separate positive adjustments per placement instead of one blunt slider.

⚠️ Watch out. By default, Amazon enables “automatically optimize bids for placements” on Sponsored Brands — handing placement control to Amazon on your most expensive ad type. Turn it off immediately (in bulk sheets, set the bid optimization flag to false) and set the modifiers yourself from your own placement data.

Part 8

Cost model and the complete strategy

CPC vs. vCPM

Sponsored Brands can bill per click (CPC) or per thousand viewable impressions (vCPM). vCPM sounds attractive — pay for eyeballs — but it bills against view-based attribution: a shopper scrolls past your ad half-visible for a second, later buys through an organic click, and the sale is credited to the vCPM campaign. You didn't earn that sale; you photographed it. The classic pattern: vCPM campaigns reporting spectacular ROAS while organic sales fall by almost exactly the same amount — pure attribution shift, zero new revenue.

Amazon is a bottom-of-funnel platform; shoppers arrive with intent, and CPC bills you only when someone engages. Default to CPC. The exception: vertical video currently requires vCPM — if you run it, watch click-attributed sales separately, and compute your effective CPC (spend ÷ clicks) against your CPC campaigns to see what those impressions really cost.

The other default to switch off

⚠️ Watch out. New Sponsored Brands campaigns enable “cost controls” by default — which disables individual keyword bids and hands Amazon blanket control of bid allocation on your most expensive ad type. Turn it off at creation, every time (unlike Sponsored Display, SB lets you reverse it later, but even one day of automated bidding skews your early data).

The complete deployment

1. Identify keyword themes.Not single keywords — clusters. “Fasting electrolytes,” “hydration packets,” “hangover hydration” are three campaigns, each with its own creative.
2. Build single-theme campaigns.Keyword-dense headline with CTA, theme-matching lifestyle image, relevant product selection, modified broad + phrase + exact coverage, and negatives (your brand, competitor brands, other themes' terms) to keep themes clean.
3. Pair with Sponsored Products.Run SP campaigns on the same keywords. Video at top of search + product in an SP slot + another video mid-page = three brand touches on one search, and the familiarity measurably lifts the SP campaigns' conversion.
4. Test in rounds.Weeks 1–2: three headlines, same image and products. Weeks 3–4: winning headline, three images. Weeks 5–6: winner + winner, different product selections. Pause losers, scale winners, repeat.
5. Optimize placements weekly.RPC-based modifiers, base bid at the weakest placement, bulk file for per-placement control.
💡 Tip. Launch tactic: feature a new, low-review product alongside your established bestsellers in a Product Collection. The veterans lend their review counts and credibility while the newcomer collects exposure it couldn't buy alone — most clicks still go to the store, but the awareness jumpstarts review velocity.

The flywheel to close on: better headlines → higher CTR → better quality score → lower CPCs → more clicks per rupee → more conversions → better organic rank → less dependence on ads. Every piece of this guide feeds the next one, and the whole thing starts with taking the format seriously. If you want that system built on your account — themes, creatives, tests, placements — the first diagnosis is free.

Want this run on your account instead? The first diagnosis is free.

Book Free Consultation

New guides, straight to your inbox.

Practical D2C playbooks as we publish them. No fluff, no spam — unsubscribe anytime.