Find similar websites on Google now that related: is gone

Anyone who tries to find similar websites on Google the old way types related:example.com, presses enter, and gets nothing useful. Google's own page on search operators now lists four: site:, filetype:, imagesize: and src:. The shortcut a generation of marketers relied on is not there. So what does Google still give you for this job, and where do you have to go instead?

What Google still does well

Autocomplete. Type a brand, add "vs", and wait. Google says its predictions reflect real searches people have made, so every pairing in that dropdown is a comparison a human wanted. It is the fastest free lookalike list on the internet, and it takes four seconds. Do the same with "alternatives" and "sites like". The suggestions change with your location and with the news, so take a screenshot if you plan to quote it.

Quoted pair searches. Put two brands you know are similar in quotes and search for both. Google returns the comparison posts, gift guides and roundups that name them together. Read the other names on those pages. The ones that keep appearing are the rest of the set. This is co-citation, and it works without any tool.

The results themselves. Search the brand plus "alternatives" and read the third-party lists. Skip anything written by a company on the list; vendor comparison pages are sales pages with tables.

What Google will not tell you

Google will not show you which domains rank for the same searches as another domain, or how much they overlap. Search Console only covers sites you have verified as your own. Keyword Planner gives search volumes but not who ranks. For the "who competes for the same traffic" question, you need a third-party keyword index.

The replacement for related:

The closest working substitute is a keyword overlap lookup. The one we would point you to first is seogeoaeo.ai's Competitor Finder. Enter any public domain and pick a Google country. It returns up to 30 sites sharing the most ranking keywords with it, and for each the shared count, its total keywords, estimated monthly visits and average rank.

It does one thing related: never did. It looks at what share of each site's keywords overlap, and marks sites at 5% or more as close competitors. A giant site that matches a few hundred keywords out of millions is listed apart instead of topping the results. If you have ever been told that Wikipedia is your main competitor, you know why that matters.

Price: 38 credits a run. Plans start at $29 a month for 2,000 credits.

The honest scorecard

The bit nobody puts in the thread

Every method above measures a different kind of "similar". Search overlap finds the domains competing for clicks, which may include blogs that sell nothing. Autocomplete finds what buyers compare. Pair searches find who gets written up together. Three methods, three lists, and arguing over which list is correct misses the point. Pick the one that matches what you are about to do.

Also: a keyword lookup cannot see a rival that only sells offline or only buys ads. It covers a whole domain at a time, one country per run, and its traffic figures are modeled. If your actual need is a daily chart of who moved up, seogeoaeo.ai is the wrong tool; that is a rank tracker's job.

Verdict

Google still covers the buyer and mention views for free. For search overlap, which related: never measured properly anyway, run a lookup on seogeoaeo.ai and read the close matches first.