Rankings
SERP Volatility Checker
Tell a ranking drop from a shuffling page
Enter a keyword and get how much its top ten has changed across recent checks, which URLs entered and left, the largest single move, and whether the churn is spread across the page or concentrated in a few positions.
Get early accessWhat it returns
- A churn figure for this keyword across the checks we hold, built from how far each URL moved and how many entered or left the top ten.
- The URLs that appeared and the URLs that dropped out between checks, named, because a new entrant explains a drop that a position number does not.
- The largest single position change in the set, and the positions it happened at, since movement at four and movement at twenty-four mean different things.
- Whether the movement is spread evenly or concentrated, with the window and the number of checks behind the figure stated plainly.
When this matters
A page falls from four to nine and the client wants to know what broke. Before anyone touches the page, the question worth answering is whether this keyword moves like that all the time. If its top ten reshuffles every few days and two of the URLs above you are different ones each week, the drop is weather and the right action is to wait for another check. If the page has been still for a month and only you moved, something specific happened and it is worth finding.
How to answer this today
Rank tracking is where volatility actually becomes measurable, because the measurement is a history and not a request: it checks your keywords on a schedule, keeps the full position series, and shows a move against the baseline that keyword normally runs at. This tool answers the narrower question for a keyword you do not track, from the checks we happen to hold, which is a thinner dataset and says so.
Rank trackingWhy this is harder than it looks
Volatility needs history, not a check
A single fetch cannot show movement. Any figure here comes from comparing checks taken at different times, so the answer is only as good as how many of those exist for this keyword and how far apart they are. Both get reported with the number.
Normal differs per keyword
Some queries reshuffle daily by design, especially news, product and anything seasonal. Others barely move for months. A churn figure means nothing without the baseline for that keyword, which is why a raw score read on its own misleads more than it helps.
Noise looks like movement
Location, device, personalisation and plain test traffic all shift a results page without anything having changed. Comparing two checks that ran under different conditions manufactures volatility, so the conditions have to be held constant before any difference is attributed to Google.
Repeated fetches cost money
Measuring change means paying for more than one results page, which is why checks are metered and rate limited per IP. The limit protects the shared budget from scripts pulling thousands of keywords, which is a paid product and not a free one.