Amazon search result pages were already personalized before everyone started talking about it. Now they are even more so. And that creates a real problem if you are trying to build accurate keyword rank tracking.
So let us be honest about what personalization actually means.
What changes a shopper's SERP
Based on their shopping history — what they searched, clicked, bought, returned — two shoppers searching the same keyword right now will see a different page. That is the personalization nobody can replicate. You are not getting their Amazon account data.
But that is only one variable. The rest are fully controllable:
- Zip code. A shopper in New York and a shopper in Texas see different prices, different delivery windows, sometimes different Buy Box winners. The SERP reflects all of that.
- Device. Mobile layout returns 21 results per page. Desktop returns 60. Sponsored placements stack differently. Rank position 4 on desktop can be position 9 on mobile after the ad block.
- Time of day. Lightning Deals go live and expire. Inventory levels change. Sponsored bids fluctuate. The SERP at 9am is not the same SERP at 9pm.
Three of the four personalization variables are things you can set, control, and track systematically. Most rank trackers ignore all three.
What most rank trackers actually give you
One location. Desktop only. Scraped once a day at a fixed time nobody tells you about. You get a number and you trust it represents your real rank. It represents one person, in one place, on one device, at one moment. That is a very narrow slice of what your shoppers are actually seeing.
The shopping history piece — you cannot fix that. You are never getting a representative sample of personalized browsing sessions. Accept that and move on.
Everything else? You should have been controlling it already.
What accurate rank tracking actually looks like
You need an API that lets you set the variables that matter. Not just a keyword and a marketplace. Zip code. Device type — mobile or desktop. Time of day. And the ability to run the same keyword multiple times across all of those combinations so you build a picture of where you actually rank for the shoppers who are most likely to buy your product.
Tracking a keyword in 5 states tells you something. Tracking it on mobile and desktop tells you something different again. Running it at 8am and 8pm shows you how rank moves through the day. Stack all of that together and you have rank intelligence. Run it in one place on desktop once a day and you have a number that may or may not be relevant to any real shopper.
Our search endpoint gives you control over all of it. Zip code, device type, scheduled frequency — per request. The data compounds into something useful the longer you run it.
We want to grow your brands, not get you in trouble with Amazon legally. There are other scrapers out there who will tell you they can do it. Until they disappear.