Website traffic for a list of domains: hedwigai against DataForSEO, SpyFu, Semrush and Ahrefs

Which of these 200 importers is a real business online? Is this competitor growing? An analyst sizing a market or a prospect list asks how much traffic a website gets, usually for many sites at once.

hedwigai now answers that from your agent. Ask one question with ask, or hand a whole list to the traffic program in the shell. Both return DataForSEO’s estimate of monthly visits from Google organic search. We priced it against the places you would otherwise get the same number, each at its own list price, on a real list of 124 domains taken from our own prospect workbooks, on 14 October 2026.

ask answers a question whose answer is a value, without starting a research run. It picks the program that can answer, runs it, and cites what it ran. Asked about PitchBook, it chose the traffic program (confidence 0.84 against every other program the account has) and answered in under four seconds:

ask itself is free. The question cost what the program it ran costs: $0.004 for the request and $0.0001 for the domain, $0.0041 in all.

It compares, too. Ask whether one site gets more traffic than another and each side’s figure is the one its own lookup returned, compared in code, so the answer is a yes or a no with both numbers beside it:

A comparison looks up both sites, $0.0082 in all. To rank a longer list, hand it to the program, which answers every domain in one request.

In the shell your agent gets from hedwigai, traffic estimate takes up to 1,000 domains, or full addresses it trims to the domain. It answers in JSON, so it pipes straight into jq, a spreadsheet or a workbook:

The first call bought the three estimates from DataForSEO (upstream_calls: 1). The second bought nothing: estimates are recomputed monthly, so hedwigai keeps each domain’s figure for the month and answers again from its own store, for every account that asks.

Providers charge in two ways: a fee per request, or a price per domain. These are per-call prices only; the monthly plans some providers require come on top. A per-request fee is expensive for one site and cheap for a long list; a per-domain price is the reverse. hedwigai charges a small fee per request and a fraction of a cent per domain, which is why the gap widens with the list.

Where we are not the cheapest. Per call, SpyFu is cheaper for lists of up to ten domains ($0.0005 a domain against our $0.0041 for one), and Serpstat for a single domain. Both need a monthly plan first: $119 for SpyFu’s Pro + AI, which includes $40 of API use, and $100 for Serpstat’s Team. From eleven domains up we are cheaper than SpyFu per call as well, and from two up cheaper than Serpstat. Against DataForSEO, SE Ranking and Adyntel we are cheaper at every size.

Against DataForSEO directly. DataForSEO is the source of our figures, and it charges $0.012 a request plus $0.00012 a domain. Going to it yourself also means an account, a verification step and a prepaid balance. We charge less because most answers come from our store, so we do not buy the same domain twice in a month, whoever asks for it.

Against Semrush and Ahrefs. Both sell far more than one number, and both price API access as a subscription: Semrush’s API comes with its Advanced plan at $549 a month, before API units, and Ahrefs’ uncapped API with Enterprise at $1,499 a month on a yearly contract. For the question “how much search traffic do these sites get”, that is a large fixed cost for a small, occasional answer.

Measured from the same machine on the same day. DataForSEO itself takes under a tenth of a second to produce an estimate; the rest is getting the request there and back.

It is an estimate of monthly visits from Google organic search in one market, the United States unless you name another. It ranks sites against each other well. It says nothing about direct, paid or social visits, and it is not a site’s analytics. Every answer names the month and the source, so a figure in a workbook can be cited as exactly what it is.

Of the 124 domains in our list, 116 had a non-zero estimate. The rest were very small sites, which is itself an answer: a prospect with no measurable search traffic is a different conversation from one with a hundred thousand visits a month.