The same search. Twenty years. A different answer every time.
A shopper types blue rain jacket, exactly as they did in 1999. Back then you ranked #1 and won the sale. Today the answer is written above the results and split across Google, ChatGPT, YouTube, TikTok, Pinterest and Meta AI, and your store shows up only where its content can be read.
Watch one query travel twenty years, and see where your store went.
Start in 1999A customer types any product-related query and page one gave you ten links. The shopper opened the top two or three stores, compared the products and the deals across those tabs, and bought from whichever won on price, product and quality. The better store made the sale. Your whole job was to rank in that handful and be the best option once they clicked.
In 2002 Google launched Froogle, a free product search that pulled listings from stores into one place and, crucially, added a rail of filters down the side, price, category, store. For the first time the shopper could narrow before they clicked. Your listing was free to appear, but it now sat inside a comparison grid instead of standing alone.
Price
- Under $50 61
- $50–$100 128
- $100–$200 44
Store
- YourStore 7
- RainCo 12
- eBay 90+
Category
- Shells
- Packable
- Kids
Google folded product search into the main results page, what it called universal search. Now a query returned web links and product thumbnails from stores, eBay and Amazon, stacked together with images and prices. The list stopped being ten links. Your product was suddenly competing for attention with pictures of the competition.
By now the shopping results were a proper grid you could shop and narrow without leaving the page. A rail of filters down the side, color, price, store, brand, sat beside rows of product cards with images, prices and star ratings. Google made these listings pay-to-play in 2012, then brought free listings back in April 2020, so the grid ended up showing paid ads and free results together in the same view.
Color
Price
- Under $50 120
- $50–$100 240
- $100+ 85
Store
- YourStore 9
- Amazon 300+
- RainCo 18
By now the results page was a stack of blocks: a swipeable product feed, an image pack, a video row, star ratings pulled from your markup, and "People also ask." The ten links were pushed below the fold. What ranked was no longer a page, it was a fragment, a review snippet, a pinned image, a how-to clip, each governed by its own rules.
In May 2024 Google put an AI Overview above everything. It reads the sources and writes the answer itself, then names three or four of them. The shopper's question is often settled before a single result is clicked. Ranking #1 now means ranking #1 underneath the answer. The prize stopped being the click and became the citation.
Now there is no single page. The same three words get asked of Google, of ChatGPT and the other assistants, of YouTube, of TikTok, of Pinterest, and of Meta AI inside Instagram. Each is a separate engine with its own index and its own slots. Your one query fans out into six answers, and your store has to earn its place in every one.
The same query, judged six different ways.
Chapter seven scattered one search across six engines. Each reads something different, and shows your store in its own set of slots. Here is what each one rewards, and exactly where you turn up.
| Surface | What it rewards, and where your store shows up |
|---|---|
| GGoogle results |
|
| AIChatGPT, Gemini & Perplexity |
|
| ▶YouTube |
|
| tTikTok |
|
| PPinterest |
|
| MMeta AI · IG, FB & WhatsApp |
|
The page didn't just change. The place people search moved too.
Under-30s didn't replace Google, they added five more front doors. A single purchase is now spread across platforms that each want the question answered in their own shape. The numbers are no longer a rounding error.
US searches that still send a click to the open web. The other 640 end on the results page itself.
SparkToro / Datosweekly ChatGPT users by early 2026, up from 400M a year earlier. A search surface with no blue links at all.
OpenAI, Feb 2026product listings in Google's Shopping Graph, refreshed roughly two billion an hour, feeding every AI answer.
Googlehigher conversion from AI-referred visitors than traditional organic, because the answer pre-qualified them.
Semrush, 2025Every era rewarded the same thing: the store the engine could read.
Froogle wanted a clean feed. AI Overviews want a page that answers, not just sells. TikTok wants spoken words and on-screen text it can index. The surface keeps changing; being readable and quotable is the constant.
Find where AI can cite your store.
One scan reads your catalogue and shows you the exact questions Google AI, Meta AI, YouTube, TikTok and Pinterest answer in your niche, and which of them name somebody else.