From keywords to audiences: the 4 eras of Amazon growth

Hacks, then retail readiness, then volume. Now Amazon targets the shopper as well as the search term. For brand owners and agency owners, with every era in one table.

Amazon growth has run through four eras. Launch hacks (2015-2018) rewarded bought reviews and forced rank. Retail readiness (2019-2022) rewarded complete listings. Volume (2023-2024) rewarded spend, price and speed. The fourth era, from 2025, rewards brands that reach the right shopper before the search. Those brands run ads, listing, catalog and stock as one system.

"Our ACoS looks fine. Sales are flat. The agency says everything is green."

You have said some version of this. Picture the weekly call behind it. Slide three is a keyword table. Forty rows, ACoS by search term, green cells and red cells. You ask who is buying from you and where they first saw the brand. Nobody on the call answers.

The table has no column for it. That gap is the shift from the third era to the fourth. It costs money every week it stays open.

What are the 4 eras of Amazon growth?

An Amazon era is a stretch of years when one kind of move wins more than any other. The era ends when Amazon blocks the move. Or it ends when every competitor copies it. There have been four: launch hacks, retail readiness, volume, and audiences and systems. The dates are my own reading of the market's history.

Era	Years	What won	What ended it	The question operators asked
1. Launch hacks	2015-2018	Review clubs, deep-discount giveaways, rebate launches to force rank	Amazon's ban on incentivized reviews in October 2016, then enforcement	How do I rank fast?
2. Retail readiness	2019-2022	Complete listings: title, full image set, bullets, A+, reviews, stock	Every serious seller caught up. Listing tools gave every seller the basics	Is my listing complete?
3. Volume	2023-2024	Ad spend, price, speed, deep stock, sales velocity on keywords	Everyone bought rank in the same auction. Rank cost more. Margin paid for it	How much do I spend to win this keyword?
4. Audiences and systems	2025 onward	Reaching the right shopper across search, AI answers, video and off-Amazon, with listing, catalog, price and stock in one plan	Still running	Where does my buyer first meet us?

Read the last column top to bottom. The question moved from the algorithm, to the page, to the auction, to the person. Each era kept the work of the one before it as the new minimum.

What won in the launch-hack era (2015-2018)?

In the launch-hack era, rank was something you forced. A seller shipped a container. The seller gave away a few hundred units at a deep discount. Then the seller asked those buyers for reviews. Sales velocity spiked. Rank followed. Organic sales carried the product for a few months.

This era came before my own account work. I read it from the public record. In October 2016, Amazon banned incentivized reviews: reviews in exchange for a free or discounted product. Its own Vine program stayed exempt.

Rank from bought reviews lasted only as long as the reviews counted.

Why did retail readiness stop being an advantage (2019-2022)?

Retail readiness stopped being an advantage once every serious seller reached it. Retail readiness is a listing and offer that pass a shopper's basic checks: a clear title, a full image set, benefit bullets, A+ content, enough reviews, a fair price and stock on hand. From 2019 to 2022, having all of it put you ahead of half-finished listings.

In July 2019, Amazon merged Enhanced Brand Content into A+ Content for sellers and made it free. Then everyone caught up. Listing agencies, image studios and AI writing tools gave every seller clean copy and professional photos. When every listing on page one looks complete, complete stops being a reason to buy. Retail readiness became the minimum to compete.

What did the volume era reward (2023-2024)?

The volume era rewarded whoever bought the most rank, the fastest. Launch budgets went into exact match on the head terms. Price cuts and coupons turned the click into an order. Deep stock kept the velocity going. Every brand bid on the same head terms. The brands with the deepest budgets and the lowest unit cost pulled ahead.

My own account work started in 2021, as a Seller Central specialist. By 2023 I was running PPC. Era 3 is the first one I watched from inside the ad console.

The volume era also hid the real size of a market. We screened six categories for a home and kitchen brand before its next launch. On the headline numbers, the market looked far bigger than it was:

Headline revenue in one category: $9.1M a month
Revenue open to an independent brand, after captive accessory lines and foodservice listings came out: $677,000 a month
Share held by one brand in a second category: 59%

These are market figures from a category screen. The headline figure sends stock and ad budget to the wrong market:

Headline figure against the revenue open to the brand: 13 times bigger

The volume era ended through arithmetic. When every competitor buys rank on the same terms, the price of rank goes up. Your margin pays for it first.

What changed in the fourth era: from keywords to audiences?

In the fourth era, the keyword still captures demand. The brand meets the shopper earlier. Three Amazon changes from the last six months show it. The AI assistant reads the person as well as the query. Sponsored ads and DSP became one platform. One campaign, in open beta, now runs across search, display, video and streaming TV.

The assistant reads the person. Amazon's May 2026 launch of Alexa for Shopping brought Rufus and Alexa+ together. Per Amazon, the assistant draws on a customer's Amazon activity and Alexa interactions: shopping history, browsing, purchases and conversations. The same words from two shoppers no longer promise the same answer. Your listing now has to win the shopper and the shopper's history.

The ad platform became one platform. Amazon says it brought sponsored ads and Amazon DSP together into one ad platform, Amazon Ads Agent. AdExchanger reports the name arrived at Amazon's unBoxed event in September 2026. Amazon says its targeting and bidding run on shopping, browsing and streaming signals from:

Authenticated customers: over 300 million

The campaign became one campaign. In September 2026, Amazon opened Full-Funnel Campaigns as an open beta for US advertisers who sell on Amazon. One campaign covers sponsored ads, display, video and streaming TV. The advertiser sets products, budget and an optional ROAS threshold. Amazon's AI picks the ad formats and moves budget between funnel stages.

Together, they change the unit of targeting:

Keyword-era habit	Audience-era question
Which keyword do I bid on?	Which shopper do I want at each stage?
ACoS by search term	First-time buyers, repeat buyers, and the path to purchase
Listing copy for the index	Listing copy for the index, the AI assistant and the shopper who saw a video first
PPC and listing as separate jobs	One system with one owner

A shopper still types "air fryer 6 quart". A Sponsored Products ad still wins or loses that click.

What are Amazon DSP, AMC and Amazon Attribution?

Amazon DSP, AMC and Amazon Attribution are the three Amazon products behind the audience shift. DSP buys ads by audience on and off Amazon. AMC is a clean room for building those audiences and measuring journeys. Attribution measures what off-Amazon traffic does to Amazon sales.

Amazon DSP is Amazon's programmatic ad buying for display, video and audio, on Amazon and across third-party sites and apps. It now lives inside Amazon Ads Agent. AdExchanger reports Amazon replaced the DSP name with DVA. Amazon says programmatic features, such as audience targeting and deal IDs, "all live in DVA+ with advanced settings." DVA stands for display, video and audio. AdExchanger describes DVA+ as the AI-run campaign mode, on by default unless the advertiser opts out. Amazon says advertisers do not need to sell on Amazon to run these campaigns.

Amazon Marketing Cloud (AMC) is, in Amazon's words, "a secure, privacy-safe, and cloud-based clean room solution." Amazon built it on AWS Clean Rooms. Advertisers run analytics and build custom audiences from engagement records, conversion events and segment information, across Amazon Ads signals and their own data. Those audiences go into sponsored, video and display campaigns. Amazon lists AMC as "available at no cost to eligible advertisers."

Per Amazon, Amazon Attribution shows the on-Amazon impact of marketing on non-Amazon channels. Those channels include search, social, display, video, email and affiliate or influencer campaigns. Amazon calls it "a free measurement solution" for eligible vendors, sellers and registered brand owners.

What am I learning about audiences on Amazon?

I run search, listings, catalog and profit in live accounts every day. DSP and AMC are the parts of this era I am studying now. I write about them as a learner. I have not run them as a service. I will say so until I have.

What I am learning, so far:

A keyword records what a shopper typed. An audience records what a shopper did. The first is one search. The second is a history.
Audience traffic lands on the same product page as search traffic. If your page does not convert a shopper who searched for the product, it will not convert one who saw a streaming ad last week. Retail readiness comes before audience spend.
As I understand it so far, AMC is where you ask what the campaign console never shows. Two examples I am studying: shoppers who saw a video before they searched the brand name, and first orders from shoppers new to the brand. I am learning to write the questions before I learn the queries.
Amazon calls Attribution free for eligible sellers and vendors. If your brand already pays for Meta, TikTok or Google traffic to its listings, Attribution is the cheapest first step into off-Amazon measurement I have found.
When Amazon's AI picks the audience, the operator's job moves up one level. The machine decides who sees the ad. Someone still has to decide what the spend is for. Someone has to decide when a number is good enough.

Why does the whole brand decide ad results now?

PPC is one lever. The brand is the system. Two years ago, the keyword auction was the main input. Good campaign work from a strong agency was enough to lift sales. Today the ad platform finds the shopper on its own signals. The result depends on everything that shopper meets after the click.

The system, layer by layer:

PPC and listing conversion: the ad buys the visit. The page earns the order.
Retail readiness: stock, price, reviews and the Buy Box. A stockout wastes every audience dollar you send to the page.
Creative, A+ and Store: video and display ads need assets a shopper trusts.
Catalog and variations: a wrong parent-child setup splits reviews and sales data across separate listings.
Audiences and measurement: who you reach, plus proof of what they bought.
External traffic: what off-Amazon spend does to Amazon orders.
Ownership: one person who connects all of it and answers the question from slide three.

My job, in one line: I help brand owners and agencies run ads, listing, catalog and stock as one Amazon system, with one person accountable for the result.

For brand owners, this is a hiring question. Somebody inside the brand has to own the channel, even when an agency runs the campaigns. For agency owners, it is a reporting question. A weekly report with only campaign metrics describes the last era. Your client already lives in this one.

What to do this week

Pull one month of search term data and split ad sales into branded and non-branded terms.
Read 30 recent reviews of your top product. Write one sentence about who buys it and why. Check whether your main image shows that person.
If you pay for any off-Amazon traffic, check your account is eligible for Amazon Attribution. Then set up tags for one channel.
Before you launch a new campaign in Amazon Ads Agent, check which campaign types default to AI-managed settings.
Brand owners: ask your agency for one page that shows ad spend, stock, price and listing changes on the same timeline.
Agency owners: add one line to the weekly client report. What changed outside the campaigns this week.

Next reads: "How do Amazon keywords work when AI reads your listing?" for the search side. "Amazon agency, in-house or VA: who owns the channel?" for the ownership side.

FAQ

What are the four eras of Amazon growth?

Launch hacks (2015-2018), retail readiness (2019-2022), volume (2023-2024), and audiences and systems (2025 onward). Each era rewarded one kind of move. Then Amazon blocked it. Or every competitor copied it.

Are keywords dead on Amazon now that ads target audiences?

No. Search terms still capture demand. The search term report is still the most honest file in the account. Audiences and AI assistants now reach the shopper around the search.

What is the difference between Amazon DSP and Sponsored Products?

Sponsored Products bid on search terms and product pages inside Amazon. Amazon DSP, now DVA inside Amazon Ads Agent, buys display, video and audio placements by audience, on Amazon and across third-party sites and apps.

Is Amazon Attribution free?

Amazon describes Amazon Attribution as a free measurement solution for eligible vendors, sellers and registered brand owners.

The keyword table on slide three is still worth reading. Add one page next to it: who buys and where they first met the brand. PPC is one lever. The brand is the system.

Sources

TechCrunch, "Amazon bans incentivized reviews tied to free or discounted products", 3 Oct 2016. https://techcrunch.com/2016/10/03/amazon-bans-incentivized-reviews-tied-to-free-or-discounted-products/
Amazon Ads, Amazon Ads Agent product page, fetched 7 Nov 2026. https://advertising.amazon.com/adtech-solutions/amazon-ads-agent
Amazon Ads, Amazon Marketing Cloud product page, last published 4 Nov 2026, fetched 7 Nov 2026. https://advertising.amazon.com/solutions/products/amazon-marketing-cloud
Amazon Ads, Amazon Attribution product page, last published 4 Nov 2026, fetched 7 Nov 2026. https://advertising.amazon.com/en-us/solutions/products/amazon-attribution
AdExchanger, "Amazon Ads and the Amazon DSP get an agentic rebrand", 29 Sep 2026. https://www.adexchanger.com/platforms/amazon-ads-and-the-amazon-dsp-get-an-agentic-rebrand/
Amazon Ads, "Full-Funnel Campaigns: from discovery to sales", 29 Sep 2026. https://advertising.amazon.com/resources/whats-new/full-funnel-campaigns-from-discovery-to-sales
Amazon, "Meet Alexa for Shopping, your personalized, agentic AI assistant on Amazon", 13 May 2026. https://www.aboutamazon.com/news/retail/alexa-for-shopping-ai-assistant
Web Retailer, "Amazon A+ Content: all your questions answered" (A+ and Enhanced Brand Content merged for sellers, July 2019), updated 26 Feb 2024. https://www.webretailer.com/b/amazon-a-plus-content