Amazon Ads: first campaignAmazon Ads: первый запуск

Sponsored Products begins with a buyable Amazon offer. Before selecting keywords or increasing bids, confirm which product and seller offer you will promote, what the customer will see on Amazon, and how you will distinguish attributed advertising sales from the business’s orders, refunds and proceeds.

This guide follows a fictional kitchen-utensil brand run by adults, selling a stainless-steel kitchen-utensil set through a professional seller account on Amazon.com. The product, business decisions and all teaching numbers are invented. They are not SABSUS customer results, forecasts, market benchmarks or a commercial trial. The example covers an Amazon marketplace seller; it does not describe Sponsored Products across retailers or advertising that sends shoppers to an independent checkout.

The deliverable is a first-campaign record with a checked ASIN and SKU, a defensible targeting choice, understood bid adjustments and a dated reporting plan. Platform information was checked on 7 October 2026. Confirm current options in the intended marketplace and account before committing spend.

For this first scenario, choose the utensil set with stable stock, a complete detail page and a clear use case. Do not combine a premium set, a replacement utensil and an unrelated kitchen accessory merely because they belong to the same brand. Their prices, customer expectations and availability can require different decisions.

Record the marketplace, ASIN, seller SKU, selected variant and selling account. The ASIN identifies the catalog product; the seller SKU identifies the seller’s stock record. Keep the pair together when checking the offer. A familiar image or title is insufficient evidence that the intended pack size, color and seller are selected.

Standard Sponsored Products ads use CPC billing and promote individual listings, with placements on Amazon and selected outside properties. Their product destination is the Amazon detail page. A link to a SABSUS website is not the destination for the marketplace campaign covered here. Sponsored Products overview.

Write the practical buying question the page must answer. For the fictional set, it is whether the included utensils, dimensions and documented care instructions fit the shopper’s kitchen. Avoid claiming a heat resistance, material certification or dishwasher compatibility that the brand cannot substantiate. A broad “premium quality” message cannot replace those details.

Name an owner for inventory and offer problems as well as an advertising owner. If the stock situation changes, someone must connect that operational fact with campaign review. A reporting meeting a week later is a poor place to discover that the advertised variant was unavailable throughout the comparison.

Amazon’s current setup help requires an active, in-stock ASIN available to buy. Separately, Featured Offer eligibility and the offer currently holding that position affect delivery. Amazon warns that a campaign can show a delivering status while the advertised SKU lacks the Featured Offer and receives no impressions. Campaign prerequisites, Featured Offer eligibility.

Inspect the offer as a shopper in the intended delivery area. Confirm the variant, price, delivery promise and seller shown at purchase. Also check the product’s state inside the seller account. Record when the observation was made: availability and the selected offer can change after the launch checklist is signed off.

Do not assume that a brand owner automatically holds the Featured Offer at every moment. When several SKUs or sellers relate to one ASIN, verify the exact offer intended for advertising. A higher advertising bid does not solve an unavailable product or establish retail eligibility.

For the fictional brand, a failed eligibility check stops preparation at this stage. The next action belongs to the seller’s operating process: investigate stock, listing state or offer conditions. Record the unresolved issue in plain language. Do not hide it inside a recommendation to raise the budget.

A successful check proves only the observed state. It does not guarantee impressions, sales or continued ownership of the Featured Offer. Include the offer in routine campaign review, especially after a price change, stock receipt, fulfillment change or notice affecting the listing.

Review the detail page before buying attention for it. Show the exact set contents and scale. A shopper should understand whether the picture represents one utensil or a set, what is included in the package, and which option is selected. Keep the product title, images, description and variation selection consistent.

Use close views for meaningful details rather than decorative repetition. In our example, the useful images explain the size of the working head, handle shape and included pieces. If a feature is relevant to use or care, describe it in verifiable terms. An image of a different variant can create returns even when the advertisement attracts appropriate shoppers.

Read recent product questions and legitimate return reasons as evidence of missing information. A buyer who expected a larger utensil may be revealing an unclear scale photograph. Separate an inaccurate page from a manufacturing issue: the appropriate fix differs, even if both appear as a disappointing purchase experience.

Amazon’s U.S. Sponsored Products prompts can introduce a conversational step using product information, and eligible campaigns are automatically enrolled. Review available prompt text and controls as part of the page-and-ad check; prompts operate within CPC billing. Amazon’s prompts announcement.

This makes maintained product information part of advertising preparation. It does not justify inventing a conversation script that every shopper will see. Check what is available for the actual account and correct unsupported product claims at their source.

Prepare the configuration on paper or in a local draft before activating it. Use a name that identifies the marketplace, product family, targeting approach and start period. Record the intended product list, ad group structure, currency, dates, spending boundary and person authorized to change it.

Amazon documents Create campaign → Sponsored Products, then product, targeting, bidding and campaign settings before launch. Confirm the flow in the selected account. Sponsored Products creation overview.

For the teaching scenario, one coherent ad group contains only the selected set. That is an editorial choice for interpretability, not an Amazon requirement or a universal account structure. If variants have materially different stock, pricing or performance, explain how they will be reviewed before grouping them.

Choose a start condition that leaves time for an operating check. Identify how the selected campaign can be paused, who monitors notices, and how a product-level issue reaches the advertising owner. Do not assume that a campaign’s end date automatically cancels every other active campaign or advertising product in the account.

The first-campaign document should be readable by someone who did not create it. They should know what will be advertised and where a shopper will arrive. A campaign name alone will not communicate whether automatic targeting, placement boosts or other rules can change its delivery.

Automatic targeting uses four groups: close match, loose match, substitutes and complements. Manual targeting lets the advertiser select keywords or product targets. These are different controls: an automatic group is not a keyword match type. Amazon targeting guidance.

For our fictional first campaign, automatic targeting asks where Amazon finds relevant demand for the prepared set. Before activation, read the available group settings and bids. Keep a note of why each group is included. “Automatic” does not remove the need to review relevance and spending.

Later, a manual campaign can examine a narrower, evidence-based question, such as demand for the intended utensil type. A product target can address shoppers considering an appropriate alternative. Do not copy a whole search-term export into a new campaign without checking whether terms actually describe the product.

Current Amazon help allows close variations for manual keyword matching, including exact match. Treat “exact” as the name of a targeting control rather than a promise of literal character-for-character searches. Verify actual matched terms in reporting. Keyword match types.

Use negative targeting to exclude a demonstrated mismatch. For example, a recurring search for a utensil the set does not include may justify an exclusion after checking its meaning. A relevant term with a few clicks and no mature sales needs a different analysis. Avoid removing useful demand just because its purchase window has not finished.

Amazon’s current bidding help distinguishes down-only, up-and-down, fixed and eligible rule-based strategies. With dynamic up-and-down, bids can currently rise or fall by up to 100% across placements. Down-only can reduce bids. These choices affect auction bids, not a guaranteed click price or sales outcome. Current Sponsored Products bidding strategies.

For a straightforward first review, the fictional team chooses down-only and avoids optional boosts until it has a reason to use them. This is a deliberately simple teaching configuration, not a claim that it produces the best result. Check every active modifier before assuming the entered base bid describes maximum exposure.

Placement adjustments can increase bids for top of search, rest of search and product pages; current help also describes Amazon Business placement adjustments. They operate in addition to the bidding strategy, and applicable audience or video boosts may also combine. Sponsored Products bid adjustments.

Read these controls as a complete set. Ask which adjustment can apply, under what condition, and who approved it. A fixed bidding strategy does not erase an explicitly configured placement adjustment. Record optional automated rules and additional delivery settings even if the base bid is unchanged.

Use suggested bids as platform guidance, not proof of affordability or a requirement to spend. If a proposed increase exceeds the business’s approved limit, revise the plan before activation. The budget and acquisition-cost guide explains the broader economics; this guide concentrates on understanding the controls that implement the decision.

Sponsored Products uses an average daily budget. Amazon’s budgeting policy allows higher spending on an individual day using available underspend, with account settings for the permitted increase, including up to 100%. Do not treat the displayed daily amount as a strict same-day ceiling. Sponsored ads budgeting policy.

Read the account’s current budget setting and any campaign budget rules. Check the campaign duration, currency, payment arrangement and other active campaigns drawing from the same business resources. A correct per-campaign amount can still coexist with an unintended total commitment across the account.

Before launch, the owner should understand what is being authorized and how it will be observed. Set a review process that can detect an unavailable offer, an incorrect target or unexpected spend promptly. Do not invent a total-spend enforcement feature if the selected account does not provide one.

Review the final product, destination, targeting, bids, boosts and dates together. When preparation is complete, active spend requires the authorized owner’s decision. Keep the final record accessible to the person who will compare the campaign against the approved plan.

Start each review by recording marketplace, account type, currency, date basis and extraction time. Then identify the report’s attribution window and whether a sales column covers the promoted item, other items or both. Keep that information beside exported figures so it survives a spreadsheet copy.

For seller Sponsored Products reporting, Amazon defines promoted sales as the advertised SKU’s attributed sales within seven days after the click, and halo sales as other-SKU sales in that window. Vendor definitions can differ. Use the actual column definitions for the selected account. Advertised product report.

Different reports answer different questions. Product reporting helps identify the advertised item; targeting reporting helps review delivery choices; placement reporting shows where performance occurred. A strong total can conceal a problem in one part of the campaign. Keep comparisons within the same reporting scope before drawing a conclusion.

Amazon’s search-term report includes terms with at least one click. It can also show inferred terms for non-search contexts, rather than a literal phrase typed by a shopper. Its impression total therefore need not equal the campaign’s total impressions. Search-term report definition.

Review the most recent period cautiously. The attribution window may still be open, and reporting can update. Preserve the extraction time and return to the same click period after sufficient maturity. Comparing incomplete recent sales with a mature earlier period can make a stable campaign look worse than it is.

The following is a wholly hypothetical future exercise, not observed October performance. Suppose a seller report reviewed on 21 October covers advertising clicks from 1–7 October and uses a documented seven-day attribution window. Spend is $240. Promoted-item attributed sales are $1,000 and halo sales are $200, giving $1,200 total attributed sales.

Reported ROAS is $1,200 divided by $240, or 5. Reported ACOS is $240 divided by $1,200, or 20%. These ratios describe that advertising report and its sales definition. They do not subtract product costs, fees, fulfillment or the financial consequences of returns.

Attributed sales amounts and purchase counts do not identify unique or new customers. Repeat buyers, multi-item orders and unknown customer status remain distinct questions. Platform attribution also does not establish which sales would have been lost without advertising; that requires a separate incremental-effect study.

A separate fictional seller ledger covers orders placed on 1–7 October, also reviewed on 21 October. Item proceeds before refunds are $1,800, and recorded refunds are $180. The remaining amount is $1,620 before fees and costs. For simplicity, these teaching figures exclude tax and separate shipping charges. They are not a settlement or payout.

The two views need not contain the same orders or products. A purchase after 7 October may belong to the advertising click period. Halo sales can refer to another SKU. Conversely, an order in the seller ledger may arise from an earlier advertising interaction or from a route outside this campaign.

Do not subtract $1,200 from $1,800 and call the remainder organic sales. Do not divide all retained seller proceeds by this campaign’s spend and label the result its ROAS. A valid reconciliation requires compatible scopes and evidence for the relationship. This example assumes no available order-level advertising join.

Amazon’s attribution help says payment failures and purchases canceled within 72 hours are deducted from attribution metrics; processing those deductions can take up to seven days. It also states that attributed sales do not reflect cart discounts and exclude tax. Those rules do not establish a complete profit or refund ledger. Ad campaign attribution.

Use seller order, refund and financial records to understand the actual business outcome. Record the item, order date, refunded amount and review date at the level available to the business. Keep partial refunds distinct from full reversals. A return request, a received return and a processed refund can represent different stages.

Then investigate reasons. In the utensil example, repeated size-related returns call for a page review; damaged goods call for an operating investigation. Do not diagnose the cause from a refund total alone. The task is to identify what can be fixed, not to assign every disappointing order to advertising.

If later corrections change a report, preserve the earlier extract and explain the updated basis. Do not quietly overwrite a launch review and leave the decision attached to different figures. Equally, do not freeze immature sales as a permanent account of campaign performance.

Profit requires the relevant costs and timing. Marketplace fees, fulfillment, product cost and other expenses are outside this guide’s short report example. Use the actual accounting basis rather than treating “amount after refunds” as cash received or contribution margin.

For no impressions, begin with active stock, the exact SKU’s Featured Offer, listing eligibility, campaign status, dates and targeting availability. Check bids only in that context. Increasing a bid first can conceal the fact that the intended offer cannot serve.

For impressions with little engagement, inspect the product’s fit, visible image, price and delivery proposition. For clicks with few mature purchases, examine the detail page and offer at the time of traffic. A product that was unavailable during part of the period should not be evaluated as if shoppers had a stable chance to buy it.

For sales with disappointing economics, separate attribution scope, actual refunds and operating costs. Look at relevant search terms and placements without assuming the highest-sales row is the most valuable row. Halo sales can be commercially useful, but require an understanding of the purchased products rather than automatic credit to the advertised set’s margin.

Change one interpretable part of the plan, document the reason and choose the observation period. Pause an unsuitable product, repair an inaccurate page, exclude a confirmed mismatch or adjust a justified bid. A small early sample supports a bounded next step, not a guarantee that scaling will preserve the same ratios.

SABSUS ADS lists Amazon Ads in its published provider registry and offers checks of connection capabilities. The 14-platform registry does not prove 14 universally live integrations. Confirm the selected account and operation using the connection acceptance checklist.

This article claims no native SABSUS seller-order integration, automatic Amazon refund import or verified live Sponsored Products launch. The Amazon marketplace order remains distinct from a purchase through a SABSUS website. A campaign card or OAuth connection alone does not establish a shared order identifier or a working conversion transport.

If the business maintains its own website as well, report those online orders separately unless a verified data model deliberately relates them. Preserve source, currency, event meaning and timing. A consolidated dashboard is useful only when its labels retain these distinctions.

Finish with a short decision record containing the checked offer, final settings, report definitions, observed issues and next action. Keep the person responsible for listing accuracy and the person responsible for spend visible. They need a shared explanation of what the campaign is testing.

The initial launch is ready when the product can be bought, the page supports an informed decision and the owner understands delivery and spending controls. The first review is useful when it can explain attributed sales alongside actual orders and refunds without inventing a relationship between them. The next change should respond to a specific finding.