Amazon Advertising Platform: What Senior Marketers Get Wrong

The Amazon advertising platform is one of the most commercially potent media environments available to performance and brand marketers today. It sits at the intersection of intent and transaction, where someone is not just searching but actively ready to buy. Used well, it compounds. Used carelessly, it becomes an expensive way to defend market share you already had.

Most brands using Amazon Ads are not using them badly. They are using them narrowly, optimising for the visible conversion while ignoring the structural work that drives long-term growth on the platform.

Why Amazon Advertising Is a Different Kind of Problem

Early in my career I was guilty of overvaluing lower-funnel performance. We had the numbers, the attribution looked clean, and the cost-per-acquisition felt like proof. What I came to understand over time is that a meaningful portion of what performance channels get credited for was already going to happen. The customer was already in motion. You were just the last door they walked through.

Amazon amplifies this problem. The platform is structurally biased toward capturing existing intent. Someone searching for a specific product category is already warm. Winning that click feels like marketing success. Often it is just efficient order-taking.

That is not a reason to reduce spend on Amazon. It is a reason to be honest about what that spend is doing and to build a more complete strategy around it. The brands that grow on Amazon are not the ones with the tightest ROAS. They are the ones that understand the platform as a media environment with multiple layers, not just a shopping cart with a bid management interface bolted on.

If you want to understand how Amazon advertising fits into a broader commercial growth strategy, the thinking I apply here connects directly to how I approach go-to-market and growth strategy more generally. The platform-specific tactics only make sense when the strategic frame is right.

The Three Layers of the Amazon Ads Stack

There are three distinct advertising layers on Amazon, and most brands treat them as variations of the same thing. They are not.

Sponsored Products are the workhorse. They appear in search results and on product detail pages. They are keyword-driven, auction-based, and directly tied to purchase intent. This is where most Amazon ad budgets live and where most of the optimisation energy goes. The mechanics are relatively straightforward: keyword selection, bid management, match types, negative keywords, and placement modifiers. Done well, Sponsored Products protect your own listings and take share from competitors at the moment of decision.

Sponsored Brands operate at a different level. They appear at the top of search results, include your brand logo and a custom headline, and can drive to a custom landing page or your Amazon storefront. The strategic purpose is different: you are building brand presence within the platform, not just fighting for individual clicks. Brands that run Sponsored Products without Sponsored Brands are leaving category ownership on the table. When a competitor’s brand appears at the top of a search and yours does not, you have already lost a portion of that impression regardless of what happens at the product level below.

Amazon DSP (Demand-Side Platform) is where the platform’s strategic scope changes entirely. DSP allows programmatic display and video advertising both on Amazon properties and across the broader web, using Amazon’s first-party purchase and browsing data for targeting. This is not a retail media channel anymore. This is a full-funnel media environment with some of the most commercially actionable audience data available anywhere in digital advertising.

The mistake I see repeatedly is brands that have optimised Sponsored Products to a fine edge but have never touched Sponsored Brands or DSP. They have built an efficient machine for capturing demand that already exists. They have done almost nothing to expand the pool of potential buyers or to build the brand signals that make future demand more likely.

Keyword Strategy: Where Most Budgets Leak

When I was at iProspect, managing paid search and performance media at scale across dozens of clients, one of the most consistent findings was that accounts over-indexed on exact match. It felt safe. The data was clean. The conversion rates looked strong. What it masked was the volume of relevant traffic being systematically excluded.

Amazon keyword strategy has the same failure mode. Advertisers lock into exact match on their highest-converting terms, celebrate the ROAS, and never interrogate what they are missing. Broad and phrase match, combined with disciplined negative keyword management, surfaces real search term data that exact match never would. Some of the most valuable keywords in any account are the ones you did not know to target at the outset.

The practical approach is a tiered structure. Run broad and phrase match in discovery campaigns with lower bids and a clear harvesting process. When a search term converts consistently, promote it to an exact match campaign with appropriate bid investment. Negatives flow both ways: exclude irrelevant terms from broad campaigns and exclude the harvested terms from discovery once they have their own campaign. This is not a novel approach. Most paid search practitioners know it. Far fewer apply it rigorously on Amazon.

Category keywords also matter more than many brands acknowledge. Bidding only on branded and product-specific terms means you only appear when someone already knows what they want. Category terms, “best protein powder for muscle gain” rather than a specific brand name, reach buyers earlier in their decision process. The conversion rate is lower. The strategic value is higher because you are shaping consideration, not just closing it.

Share of Voice as a Strategic Metric

ROAS is the metric most Amazon advertisers optimise for. It is a useful operational number. It is a poor strategic one.

When I was judging the Effie Awards, the campaigns that consistently impressed were the ones with a clear theory of how advertising would change competitive position, not just the ones with the best cost-per-outcome. The same logic applies to Amazon. The question worth asking is not “what did we pay for each sale?” but “what percentage of visible impressions in our category do we own, and how is that changing?”

Share of voice on Amazon is measurable through impression share data in campaign reporting. It tells you something ROAS cannot: whether you are growing your competitive footprint or simply defending the ground you already hold. A brand with 40% impression share in its category and a moderate ROAS is in a stronger position than a brand with 15% impression share and a pristine ROAS. The first brand is building. The second is optimising itself into irrelevance.

BCG’s work on brand and go-to-market strategy has long made the point that sustainable growth requires coordination between brand-building and commercial execution. Amazon is a useful test case for that thesis. The brands winning on the platform long-term are not the most aggressive bidders. They are the ones treating it as a brand environment as much as a sales channel.

Product Detail Page Quality: The Variable Advertisers Ignore

Advertising on Amazon does not happen in isolation from the product listing. The detail page is the landing page. Its quality directly affects conversion rate, and conversion rate directly affects organic ranking, which in turn affects how much you need to spend on paid to maintain visibility. This feedback loop is one of the most important dynamics on the platform and one of the least discussed in media planning conversations.

I have seen brands run competent campaigns against listings with weak titles, thin bullet points, low-quality images, and no A+ content. The traffic arrives. The conversion rate is poor. The ROAS looks bad. The conclusion drawn is that the campaign is not working. The actual problem is that the destination is not working.

A strong product detail page on Amazon includes a keyword-rich title that still reads naturally, bullet points that address the actual purchase objections a buyer in that category has, high-resolution images that show the product in use rather than just on a white background, and A+ content that builds brand context for buyers who are still deciding. Reviews matter too, both volume and recency. None of this is advertising. All of it determines whether advertising converts.

The principle here is the same one that applies in any performance environment. You can drive all the traffic you want. If the thing you are driving traffic to does not do its job, the spend is wasted. Amazon just makes the feedback loop unusually tight and measurable.

Amazon DSP: The Part Most Brands Are Not Using

Amazon DSP deserves more attention than it typically gets in mid-market brand strategy. The minimum spend requirements and the managed service model have historically made it feel like an enterprise-only tool. That is changing, and the strategic case for using it is strong regardless of scale.

The core value of DSP is the audience data. Amazon knows what people buy, what they browse, what categories they shop in, and what price points they respond to. That data is not available anywhere else in digital advertising at the same level of commercial specificity. Running display or video against an audience of “people who have purchased in your category in the last 90 days but not from your brand” is a targeting capability that simply does not exist on most other platforms with the same fidelity.

DSP also enables retargeting of product detail page visitors, which closes a gap that Sponsored Products alone cannot address. Someone who viewed your listing but did not convert is a warm audience. Following them with a display unit on Amazon-owned properties or across the web is straightforward with DSP and measurably improves conversion rates on that audience segment.

The broader strategic shift DSP enables is moving Amazon from a last-click channel to a full-funnel one. Paired with Sponsored Brands at the top of search and Sponsored Products at the point of decision, DSP lets you run awareness and consideration activity against precisely defined commercial audiences. That is a fundamentally different media strategy from bidding on keywords and hoping for the best.

Forrester’s research on go-to-market execution challenges consistently surfaces the tension between short-term conversion focus and the investment needed to build durable market position. Amazon DSP is one of the cleaner ways to address that tension within a single platform environment.

Campaign Structure: The Architecture That Determines Everything

Bad campaign structure is the silent killer of Amazon ad performance. You can have the right keywords, the right bids, and the right creative and still produce mediocre results if the account architecture does not give you clean data and clean control.

The principles that apply here are similar to what I would apply to any large-scale paid search account. Separation of match types into distinct campaigns gives you bidding control without match types competing against each other. Separation by product category or SKU group gives you budget control and prevents high-volume products from consuming budget that should be allocated to strategic launches or underperforming lines that need support. Automatic campaigns run in parallel with manual ones to surface new search terms, not as a replacement for manual control.

Dayparting and placement bid adjustments are levers that most accounts underuse. If your conversion data shows that performance drops significantly in certain hours or that top-of-search placements convert at two times the rate of rest-of-search, those signals should be reflected in your bid strategy. Amazon’s placement modifier allows you to increase bids specifically for top-of-search positions. Using it intelligently based on actual placement performance data is a straightforward efficiency gain that many accounts leave unclaimed.

The growth loop principle, the idea that good data produces better decisions which produce better outcomes which produce better data, applies directly to campaign architecture. A well-structured account generates clean, actionable signals. A poorly structured one generates noise that looks like data but does not tell you anything useful. Tools like Hotjar’s growth loop framework make this point in a web analytics context, but the logic transfers cleanly to any performance media environment.

Measurement: What Amazon Attribution Actually Tells You

Amazon’s attribution model is last-click within a defined window. It is clean and consistent, which makes it useful for operational decisions. It is also limited in ways that matter strategically.

The first limitation is that it does not capture the halo effect of Amazon advertising on sales outside the platform. Brands with strong Amazon presence often see organic search and direct sales lift that is not attributed to any Amazon campaign. This is real value that the platform’s own reporting systematically misses. Amazon Attribution (the external measurement tool, separate from the campaign reporting) goes some way toward addressing this by tracking off-Amazon traffic sources, but the inverse problem, measuring Amazon’s influence on off-platform behaviour, remains genuinely difficult.

The second limitation is the new-to-brand metric, which Amazon does surface in Sponsored Brands and DSP reporting. This is one of the most strategically important numbers in the platform and one of the least discussed. New-to-brand tells you what percentage of attributed sales came from customers who had not purchased from your brand on Amazon in the previous 12 months. A campaign with a lower ROAS but a high new-to-brand rate is doing something more valuable than a high-ROAS campaign that is predominantly selling to existing customers. It is growing the customer base, not just monetising it.

I have sat in enough post-campaign reviews to know that new-to-brand rarely gets the attention it deserves. The room gravitates toward ROAS because it is familiar and it looks like efficiency. New-to-brand is harder to defend in a short-term budget conversation. It is also the metric that tells you whether you are building a business or just running a very expensive loyalty programme.

Growth strategy frameworks from sources like Semrush’s growth examples consistently show that sustainable growth requires expanding the customer base, not just deepening engagement with the existing one. Amazon’s own data infrastructure gives you a way to measure that expansion directly. Using it changes the conversation.

Where Amazon Advertising Fits in a Full Go-To-Market Plan

Amazon is not a standalone channel. For most consumer brands, it is one part of a broader go-to-market system that includes owned channels, retail partnerships, social commerce, and brand media. The question worth asking is not “how do we maximise Amazon?” but “how does Amazon fit into the overall system and what role should it play?”

For some brands, Amazon is the primary sales channel and the advertising strategy should reflect that weight. For others, Amazon is a visibility and consideration environment where people research before buying elsewhere, and the advertising investment should be calibrated to that role rather than optimised purely for on-platform conversion.

Creator-led campaigns and social commerce are increasingly driving Amazon discovery, particularly for consumer goods categories where influencer content generates direct product searches. Later’s research on creator-led go-to-market campaigns points to the growing importance of coordinating creator activity with retail media investment, so that the awareness generated off-platform converts efficiently when it arrives on Amazon. That coordination requires intentional planning, not just running both channels independently and hoping they reinforce each other.

The strategic framing I apply to Amazon advertising sits within a broader view of how growth actually works. If that thinking is useful, more of it lives in my writing on go-to-market and growth strategy, where the channel-specific tactics connect to the commercial logic behind them.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions