Does A Distinctiveness Score Prove Anything About A Produced Page

Does a distinctiveness score prove anything about a produced page


The question surfaces every time a spam update lands, and it arrives in the same shape: someone has a batch of generated pages, a utility that reports them as highly distinct from one another, and a suspicion that the number is not addressing the question they really have. They are right to be wary, and the explanation is easier to demonstrate than to argue about, because both halves of it can be checked on the same set of pages in around a second.


I generate location pages for one service, one page per town, from a skeleton with a lot of synonym slots. My checker says the pages hardly overlap, which everyone tells me is the goal. But the search engine's own policy lists synonymising as an example of the thing it acts against, so which one am I supposed to believe, my score or their page?


Both, since they are not measuring the same thing. The policy language is about what a batch of pages is for and how little it adds for the person reading it; the method is clearly not the offence, and the same sentence says so whether a individual or a machine did the writing. A similarity score, however, measures variation density over short runs of words. A pool can move that score as far as it goes and still say one thing in many costumes, which is the pattern the policy example depicts rather than the reverse of it. The first thing to establish is not how distinct the pages look but where their differences came from.



There is a test that reveals where the differences came from, and it takes one run. Replace every value that originated from data, the town name included, with a substitute, and measure the pool again. If what remains is many pages that are still unique, the template manufactured them and the reader receives one page every time. If what remains collapses to a single page repeated, the data was carrying the difference, and the pages are worth as much as the data behind them is. That second part the test cannot resolve for you, and no measurement can.



The write-up, with both pools assessed and the file that reproduces them: cold email spintax lint



Run the deletion pass before the publication decision rather than after, because it answers a question the dashboard never answers: what does a reader get on the second page that they did not see on the first. If the answer is a different arrangement of the same sentence, more variation will not fix it, and the number that claims otherwise is tracking the wrong surface.