[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"blog-en-what-products-to-use-for-listing-image-a-b-testing":3,"blog-translations-what-products-to-use-for-listing-image-a-b-testing":54,"blog-related-category-en-what-products-to-use-for-listing-image-a-b-testing":55,"blog-related-latest-en-what-products-to-use-for-listing-image-a-b-testing":68},{"id":4,"type":5,"locale":6,"slug":7,"title":8,"description":9,"body":10,"status":11,"section":12,"tags":13,"author":18,"cover_url":19,"published_at":20,"metadata":21,"template":5,"sort_order":49,"source_id":50,"search_vector":51,"created_at":52,"updated_at":53},"a6d6553f-2d7a-453b-9787-0f25b6fdfae1","blog","en","what-products-to-use-for-listing-image-a-b-testing","What products to use for listing image A\u002FB testing","Use Manage Your Experiments on high-traffic ASINs, panel pick tests before launch, and sequential tests in between. Test where a conversion gain compounds.","Use Amazon's Manage Your Experiments for any ASIN with enough traffic, panel-based pick tests for pre-launch images, and disciplined sequential tests for everything in between. The bigger question is which products to test on: high-traffic ASINs where a conversion gain compounds, not slow listings where no test ever reaches significance.\n\n## The short version\n\n- **Manage Your Experiments is the default.** Free, native, real purchase data. It needs Brand Registry and a traffic threshold.\n- **Panel pick tests answer a different question.** They measure preference before launch, not purchases after it.\n- **Sequential testing fills the gap.** Change one image, hold everything else, compare matched windows.\n- **Test your loudest ASINs first.** The value of a winning image is the conversion delta times the traffic behind it.\n- **The main image is the whole game.** It decides the click in search results before your gallery gets a vote.\n\n## The three options compared\n\nA common and expensive mistake is running image tests on the listings that feel neglected, the slow movers, where even a real improvement produces too few sessions to prove itself. Test where the traffic already is. A two-point conversion gain on your best seller pays for the entire testing program; the same gain on a trickle listing is a rounding error.\n\n| Option | What it measures | Needs | Best for |\n|---|---|---|---|\n| Manage Your Experiments | Actual purchase behavior, split in real time | Brand Registry, an eligible high-traffic ASIN | Established listings with steady sessions |\n| Panel pick testing | Stated preference between candidates | A panel, a handful of image candidates | Pre-launch, before you have traffic |\n| Sequential self-testing | Before-and-after conversion on matched windows | Discipline, stable price and stock | Mid-traffic ASINs below the experiment threshold |\n\nThe decision rule: if the ASIN qualifies for Manage Your Experiments, use it, because purchase data beats stated preference every time. If the product has no listing yet, panel-test the main image candidates before launch. Everything in between gets sequential testing, one variable at a time, with price, stock, and ad spend held steady so the comparison means something.\n\n## How we sequence image testing into a launch\n\nAt Flapen, image candidates are produced before a product ever ships in volume. Our validation phase runs 200 units at $5,000 to $10,000, with up to four products tested at once, and conversion rate is one of the gates a product must pass before we scale it. That means the main image gets pick-tested against alternatives while the validation stock is on the water, so the listing goes live with the strongest known candidate rather than a guess we fix later.\n\nOnce real traffic arrives, the sequence is:\n\n1. Confirm the click side first. Watch main-image click-through in search before touching the gallery.\n2. Test the main image with the best available method for that ASIN's traffic level.\n3. Only then test gallery order, infographics, and lifestyle shots, which shape conversion after the click.\n4. Re-test whenever the competitive set changes, because a main image wins or loses relative to the grid around it.\n\nCreative for our clients comes out of our in-house Dubai studio, and the brief for every image set includes the test plan, not just the shot list. If you want that discipline applied to your own catalog, the audit at [Flapen](\u002Famazon-consulting) grades primary image click-through as one of its core checks.\n\n## What testing tool vendors will not tell you\n\nMost image tests are decided at thumbnail size. Your main image competes as a small rectangle in a crowded grid, and a design that looks premium at full resolution can be illegible at the size where the click actually happens. Evaluate every candidate at the size shoppers see first, or the test result will mislead you.\n\nThe second omission is statistical. An underpowered test produces a winner, every time, because something is always ahead when you stop counting. That winner is frequently noise. If the traffic is not there, no tool can conjure significance, which is exactly why the choice of which product to test on matters more than the choice of tool.\n\nThird, panels measure what people say, and shoppers do what they do. Pick tests are useful for eliminating weak candidates before launch, but a panel preference is a hypothesis, not a verdict. The verdict comes from purchase data.\n\n## Related answers\n\n- [How to fix low conversions on Amazon Spain](\u002Fblog\u002Fhow-to-fix-low-conversions-on-amazon-spain)\n- [Top tools for keyword-level Amazon KPIs](\u002Fblog\u002Ftop-tools-for-keyword-level-amazon-kpis)\n- [Alternatives to spreadsheet-based Amazon reporting](\u002Fblog\u002Falternatives-to-spreadsheet-based-amazon-reporting)\n- [How to measure organic lift from paid on Amazon](\u002Fblog\u002Fhow-to-measure-organic-lift-from-paid-on-amazon)\n- [Amazon account measurement and audits: the complete guide](\u002Fblog\u002Fmeasurement-and-audit)\n\nTo find out which of your images is costing you clicks, request the free 48-hour audit from [Flapen](\u002Famazon-consulting).\n\n## Keep learning\n\n- [Run Amazon advertising](\u002Fguides\u002Fadvertising)\n- [Run the review math](\u002Ftools\u002Freview-calculator)\n","published","advertising",[14,15,16,17],"product-images","listing-setup","brand-registry","amazon-fba","joel-turcotte-gaucher","\u002Fimages\u002Fblog\u002Fclusters\u002Fmeasurement-and-audit-06.jpg","2026-09-04T13:30:51.798+00:00",{"faq":22,"seo":43,"batch":44,"cluster":45,"cover_alt":46,"answers_prompt":47,"primary_benchmark":48},[23,27,31,35,39],{"id":24,"answer":25,"question":26},"what-products-to-use-for-listing-image-a-b-testing-faq-1","The ones where traffic times potential conversion lift is highest, usually your top sellers and any high-traffic listing with below-category conversion. Skip listings with too few sessions to resolve a difference within a reasonable window.","Which ASINs are worth testing images on?",{"id":28,"answer":29,"question":30},"what-products-to-use-for-listing-image-a-b-testing-faq-2","For Amazon's native experiments, yes, and the ASIN must also meet a traffic bar. Without Brand Registry you are limited to panel testing and sequential before-and-after testing, which still work if you control the variables.","Do I need Brand Registry to A\u002FB test images?",{"id":32,"answer":33,"question":34},"what-products-to-use-for-listing-image-a-b-testing-faq-3","Until it reaches significance or a sensible cap, not until one side looks ahead. Cutting a test short at the first exciting gap is the most common way sellers ship a worse image with confidence.","How long should an image test run?",{"id":36,"answer":37,"question":38},"what-products-to-use-for-listing-image-a-b-testing-faq-4","Main image first, always. It drives click-through from search, which multiplies through everything downstream. Gallery, infographic, and lifestyle testing come after the click side is won.","Should I test the main image or the gallery first?",{"id":40,"answer":41,"question":42},"what-products-to-use-for-listing-image-a-b-testing-faq-5","No. A deal badge or price move swamps the effect of the image and contaminates the result. Hold price, stock, and ad structure as steady as you can for the whole test window.","Can I test images during a promotion or price change?",{},"B16","C09","Flapen cover for What products to use for listing image A\u002FB testing: a re-shoot in the studio against a blank reference card","what products to use for listing image a\u002Fb testing","B7",0,null,"'\u002Famazon-consulting':604C,842C '\u002Fblog\u002Falternatives-to-spreadsheet-based-amazon-reporting':803C '\u002Fblog\u002Fhow-to-fix-low-conversions-on-amazon-spain':786C '\u002Fblog\u002Fhow-to-measure-organic-lift-from-paid-on-amazon':813C '\u002Fblog\u002Fmeasurement-and-audit':822C '\u002Fblog\u002Ftop-tools-for-keyword-level-amazon-kpis':795C '\u002Fguides\u002Fadvertising':848C '\u002Ftools\u002Freview-calculator':853C '000':417C,420C '1':488C '10':419C '2':507C '200':413C '3':523C '4':539C '48':837C '5':416C 'a':31B,79C,111C,118C,152C,182C,188C,208C,224C,244C,248C,288C,290C,373C,394C,403C,437C,476C,549C,638C,642C,646C,690C,761C,765C,768C 'a\u002Fb':8A 'account':815C 'actual':261C 'actually':663C 'ad':379C 'advertising':847C 'after':128C,309C,536C 'against':454C 'ahead':698C 'already':222C 'alternatives':455C,796C 'always':697C 'amazon':36C,784C,793C,801C,812C,814C,846C,864 'amazon-fba':863 'an':270C,686C 'and':24B,57C,110C,190C,308C,317C,378C,429C,530C,574C,645C,744C,817C 'answer':117C 'answers':777C 'any':42C 'applied':595C 'are':400C,628C,752C 'around':559C 'arrives':484C 'as':611C,637C 'asin':43C,275C,332C,519C 'asins':18B,77C,147C,322C 'at':372C,396C,415C,427C,602C,630C,651C,657C,668C 'audit':601C,839C 'audits':818C 'available':515C 'based':49C,800C 'be':655C 'beats':343C 'because':340C,548C,694C 'before':22B,124C,178C,298C,307C,362C,402C,441C,503C,758C 'before-and-after':306C 'behavior':263C 'behind':162C 'below':323C 'best':232C,256C,514C 'between':28B,64C,286C,366C 'bigger':66C 'brand':108C,268C,861 'brand-registry':860 'brief':576C 'but':760C 'can':654C,716C 'candidate':473C,667C 'candidates':287C,294C,361C,399C,757C 'catalog':599C 'change':135C 'changes':547C 'checks':616C 'choice':724C,735C 'click':174C,491C,499C,538C,609C,662C 'click-through':498C,608C 'clicks':833C 'clients':564C 'comes':565C,772C 'common':189C 'compare':141C 'compared':187C 'comparison':385C 'competes':636C 'competitive':545C 'complete':820C 'compounds':34B,82C 'confirm':489C 'conjure':717C 'conversion':32B,80C,157C,228C,310C,430C,535C 'conversions':782C 'core':615C 'costing':831C 'counting':702C 'creative':561C 'crowded':643C 'data':105C,342C,775C 'decided':629C 'decides':172C 'decision':328C 'default':100C 'delta':158C 'design':647C 'different':119C 'discipline':314C,594C 'disciplined':58C 'do':746C,749C 'dubai':572C 'eligible':271C 'eliminating':755C 'else':140C 'enough':45C 'entire':237C 'error':250C 'established':276C 'evaluate':665C 'even':207C 'ever':89C,405C 'every':346C,578C,666C,692C 'everything':62C,139C,364C 'exactly':721C 'expensive':191C 'experiment':325C 'experiments':13B,40C,97C,260C,337C 'fba':865 'feel':201C 'few':213C 'fills':132C 'find':824C 'first':148C,493C,673C 'fix':479C,780C 'flapen':397C,603C,841C 'for':5A,41C,52C,61C,235C,257C,334C,517C,562C,577C,754C,789C 'four':424C 'free':101C,836C 'frequently':706C 'from':773C,809C,840C 'full':652C 'gain':33B,81C,229C,242C 'gallery':180C,506C,527C 'game':170C 'gap':134C 'gates':436C 'gets':181C,367C,450C 'goes':467C 'grades':605C 'grid':558C,644C 'guess':477C 'guide':821C 'handful':291C 'happens':664C 'has':351C 'have':300C 'held':381C 'high':16B,75C,273C 'high-traffic':15B,74C,272C 'hold':138C 'hour':838C 'house':571C 'how':388C,778C,804C 'hypothesis':766C 'if':330C,348C,590C,708C 'illegible':656C 'image':7A,137C,154C,166C,195C,293C,360C,391C,398C,449C,497C,511C,551C,579C,607C,626C,635C 'images':56C,829C,856 'improvement':210C 'in':27B,63C,175C,265C,365C,407C,501C,570C,641C 'in-house':569C 'includes':581C 'infographics':529C 'into':393C 'is':68C,98C,155C,167C,193C,223C,247C,432C,460C,487C,684C,696C,705C,711C,720C,764C,830C 'it':106C,129C,163C,171C,253C,339C,444C,560C 'its':614C 'itself':217C 'just':586C 'keep':843C 'keyword':791C 'keyword-level':790C 'known':472C 'kpis':794C 'later':480C 'launch':23B,55C,125C,297C,363C,395C,759C 'learning':844C 'level':522C,792C 'lifestyle':531C 'lift':808C 'list':589C 'listing':6A,246C,353C,466C,858 'listing-setup':857 'listings':85C,199C,277C 'live':468C 'looks':649C 'loses':554C 'loudest':146C 'low':781C 'main':165C,359C,448C,496C,510C,550C,634C 'main-image':495C 'manage':11B,38C,95C,258C,335C 'matched':142C,312C 'math':852C 'matters':731C 'means':386C,446C 'measure':122C,740C,806C 'measurement':816C 'measures':254C 'method':516C 'mid':320C 'mid-traffic':319C 'mislead':679C 'mistake':192C 'more':732C 'most':625C 'movers':205C 'must':439C 'native':102C 'needs':107C,255C 'neglected':202C 'no':87C,352C,714C 'noise':707C 'not':83C,126C,585C,622C,712C,767C 'of':151C,292C,434C,567C,613C,725C,736C,827C 'omission':683C 'on':14B,73C,197C,230C,243C,311C,461C,730C,783C,811C 'once':428C,481C 'one':136C,370C,433C,612C 'only':524C 'option':251C 'options':186C 'or':553C,674C 'order':528C 'organic':807C 'our':409C,563C,568C 'out':566C,825C 'own':598C 'paid':810C 'panel':19B,48C,114C,281C,289C,356C,762C 'panel-based':47C 'panel-test':355C 'panels':739C 'pass':440C 'pays':234C 'people':742C 'phase':411C 'pick':20B,50C,115C,282C,452C,750C 'pick-tested':451C 'plan':584C 'point':227C 'pre':54C,296C 'pre-launch':53C,295C 'preference':123C,285C,345C,763C 'premium':650C 'price':316C,376C 'primary':606C 'produced':401C 'produces':211C,689C 'product':350C,404C,438C,727C,855 'product-images':854 'products':2A,70C,425C 'program':239C 'prove':216C 'purchase':104C,262C,341C,774C 'purchases':127C 'qualifies':333C 'question':67C,120C 'rate':431C 'rather':474C 're':541C 're-test':540C 'reaches':90C 'real':103C,209C,266C,482C 'rectangle':640C 'registry':109C,269C,862 'related':776C 'relative':555C 'reporting':802C 'request':834C 'resolution':653C 'result':677C 'results':177C 'review':851C 'rounding':249C 'rule':329C 'run':845C,849C 'running':194C 'runs':412C 's':37C,520C 'same':241C 'say':743C 'scale':443C 'search':176C,502C 'second':682C 'see':672C 'self':304C 'self-testing':303C 'seller':233C 'sequence':390C,486C 'sequential':25B,59C,130C,302C,368C 'sessions':214C,280C 'set':546C,580C 'setup':859 'shape':534C 'ships':406C 'shoppers':671C,745C 'short':93C 'shot':588C 'shots':532C 'side':492C 'significance':91C,718C 'size':632C,659C,670C 'slow':84C,204C 'small':639C 'so':383C,464C 'something':387C,695C 'spain':785C 'spend':380C 'split':264C 'spreadsheet':799C 'spreadsheet-based':798C 'stable':315C 'stated':284C,344C 'statistical':685C 'steady':279C,382C 'stock':318C,377C,459C 'stop':701C 'strongest':471C 'studio':573C 'tell':623C 'test':29B,72C,88C,144C,218C,357C,508C,526C,542C,583C,676C,688C,729C 'tested':426C,453C 'testing':9A,131C,238C,283C,305C,369C,392C,618C 'tests':21B,26B,51C,60C,116C,196C,627C,751C 'than':475C,733C 'that':200C,445C,518C,593C,648C,703C 'the':65C,92C,99C,133C,149C,156C,160C,164C,168C,173C,184C,198C,203C,220C,236C,240C,324C,327C,331C,349C,358C,384C,435C,447C,457C,462C,465C,470C,485C,490C,505C,509C,513C,537C,544C,557C,575C,582C,587C,600C,658C,661C,669C,675C,681C,709C,723C,734C,770C,819C,835C,850C 'then':525C 'there':713C 'they':121C,748C 'third':738C 'three':185C 'threshold':113C,326C 'through':500C,610C 'thumbnail':631C 'time':267C,347C,374C,693C 'times':159C 'to':3A,71C,215C,418C,423C,556C,596C,728C,779C,797C,805C,823C 'too':212C 'tool':619C,715C,737C 'tools':788C 'top':787C 'touching':504C 'traffic':17B,46C,76C,112C,161C,221C,274C,301C,321C,483C,521C,710C 'trickle':245C 'two':226C 'two-point':225C 'underpowered':687C 'units':414C 'up':422C 'use':4A,10B,35C,338C 'useful':753C 'validation':410C,458C 'value':150C 'variable':371C 'vendors':620C 'verdict':769C,771C 'version':94C 'volume':408C 'vote':183C 'want':592C 'watch':494C 'water':463C 'we':389C,442C,478C 'weak':756C 'what':1A,252C,617C,741C,747C 'when':699C 'whenever':543C 'where':30B,78C,86C,206C,219C,660C 'which':69C,533C,719C,726C,826C 'while':456C 'whole':169C 'why':722C 'will':621C,678C 'windows':143C,313C 'winner':691C,704C 'winning':153C 'wins':552C 'with':44C,278C,375C,421C,469C,512C 'yet':354C 'you':299C,591C,624C,680C,700C,832C 'your':12B,39C,96C,145C,179C,231C,259C,336C,597C,633C,828C","2026-09-04T16:27:40.916874+00:00","2026-09-04T16:38:14.114968+00:00",[],[56,60,64],{"slug":57,"title":58,"published_at":59},"measurement-and-audit","Amazon account measurement and audits: the complete guide","2026-09-04T16:31:51.798+00:00",{"slug":61,"title":62,"published_at":63},"who-manages-amazon-vendor-central-in-the-uk","Who manages Amazon Vendor Central in the UK","2026-09-04T13:34:51.798+00:00",{"slug":65,"title":66,"published_at":67},"who-handles-amazon-dsp-in-uae","Who handles Amazon DSP in UAE","2026-09-04T13:33:51.798+00:00",[69,73,77],{"slug":70,"title":71,"published_at":72},"brand-tiers","Amazon brand management tiers: the complete guide","2026-09-04T16:33:51.798+00:00",{"slug":74,"title":75,"published_at":76},"geography-and-marketplaces","Amazon marketplaces by geography: the complete guide","2026-09-04T16:32:51.798+00:00",{"slug":57,"title":58,"published_at":59}]