Let's start with the thing most keyword tools won't say out loud: Amazon does not publish search volume. Not in Seller Central, not in an API, not anywhere.
Every exact number you've ever seen in a keyword tool, including ours, is an estimate. Amazon gives you rankings and query-level performance data, and everyone else models volume on top of that.
So the skill isn't finding a number. It's knowing which source to trust for which job.
Here are the four that matter, roughly in order of how much access they require and ends with the one that requires none at all.
1. Brand Analytics Top Search Terms (from Amazon)
If you're brand registered, Amazon hands you the Top Search Terms report inside Brand Analytics. This is real Amazon data, which is the actual most-searched terms on the platform that is updated weekly.
The catch: it's rank, not volume. You'll see that "creatine gummies" is search frequency rank #1,200 and another term is #4,800. That tells you relative demand. It does not tell you whether #1,200 means 250,000 searches or 25,000.
Use it for: Validating that a keyword has meaningful demand, spotting rising terms week over week, and checking the top-3 clicked ASINs for any search term (a criminally underused competitive signal).
Don't use it for: Forecasting units, sizing a market, or comparing keywords that are far apart in rank. The gap between rank #100 and #200 is enormous; the gap between #100,000 and #100,100 is basically nothing.
2. Search Query Performance (from Amazon or through a tool)
This is the most accurate data in this entire post, and it's the dashboard almost nobody opens.
Search Query Performance (SQP), also in Seller Central for brand-registered sellers, shows actual query counts: impressions, clicks, cart adds, and purchases for search terms where your brand appeared. Not estimates. Not rank. Real counts.
Two catches:
- It only covers queries where your brand already shows up. Perfect for keywords you own, useless for scouting new ones.
- The native dashboard is painful. One query at a time, clunky exports, no trending. This is why tools like SmartScout pull SQP data in via API. The same Amazon-sourced numbers, but queryable across your whole catalog with history intact.
Use it for: finding keywords where you get impressions but no clicks (listing problem), or clicks but no purchases (price/review problem). It's a diagnostic tool disguised as a keyword tool.
3. SmartScout Search Term Explorer
This is where estimated volume becomes an actual number you can work with.
SmartScout tracks over 10 million top-performing Amazon keywords, each with estimated monthly search volume and multi-year history.
Why history is the whole game: a single volume number can't tell you if you're looking at a trend or a spike.
Real example. "Creatine gummies" pulled roughly 101,000 monthly searches in January 2024.
By January 2025, search volume more than doubled to roughly 224,000, and it currently holds steady at around 250,000 monthly searches.
While the initial surge settled, the shift in consumer preference remained intact. At the same time, "creatine monohydrate" still comes about 417,000 monthly searches.
Looking at a single data point only tells you that the legacy term is larger, but analyzing the historical trend reveals the market's true trajectory. This allows you to distinguish between catching a lasting wave of momentum and launching into a passing fad.
Use it for: Sizing keyword opportunities before a launch, seasonality checks, and separating durable demand shifts from TikTok-fueled sugar highs.
4. Amazon Search Trends (Free)
If you just need a fast answer, SmartScout's Amazon Search Trends tool is free. No login, no credit card. Type a keyword, see the trend.
It's the triage step.
Check a keyword here first, and only bring in the heavier tools once something looks worth the effort. Most keyword research dies at this stage, which is exactly the point.
Kill bad ideas for free.
No single source gives you the full picture, because Amazon never published one. Stack them, and you get something better than a number: you get context.
