About
About SEOCheck
We run an SEO backlink API and a free checker. Link data over HTTP, nothing else.
SEOCheck research team ·

The short version
SEOCheck is a developer API for backlinks, referring domains, anchors, top pages and a 0-100 Domain Rating. A free web checker sits on top of it, and both read from the same index.
We crawl the web ourselves and keep our own link graph. It holds 181,132,097 domains, 2,774,625,913 domain-to-domain links and about 798.5 billion backlinks. It refreshes roughly once a month. That is a real limit, and we put it on the front page for that reason.
The account, keys and credit balance are shared with sister sites that sell the same product under other names. One sign-in, one balance.
Two lines of history are enough. The index you query today was built from crawl runs that finish roughly every month, and each run replaces the last. There is no archive of past states to query, so store anything you want to compare over time.
Where the data comes from
Every number you get back comes from a link graph we crawled ourselves. We do not resell another vendor's index and we do not wrap anyone else's API.
That has a cost. Our crawl is smaller than the biggest commercial indexes and it updates about once a month. Compare a fresh Ahrefs crawl to ours and you will see links they have and we lack. You may also see the reverse.
The 0-100 Domain Rating comes from a public formula, round(100 log(ref_domains + 1) / log(2 10^7)), so you can check our arithmetic by hand. Details sit on the DR checker page.
What we believe
Output first
If it cannot be parsed in one line of jq, it is not done.
Price the call, not the seat
One credit is $0.0001. You pay for what you run.
Say what is missing
No keyword database. No rank tracker. For those, Ahrefs and Semrush are fuller products.
How this site works
Google account only
No passwords to store, no forms to fill. One click and a key is waiting.
$100 on sign-up
That is 1,000,000 credits, no card needed. Enough for a lot of curl.
USDT (TRC-20)
Add credit from the account page when the free amount is spent.

What we will not build
We get asked for keyword volumes, SERP snapshots and rank tracking. The answer is no. Those need different data and different crawlers, and Ahrefs and Semrush already do them well.
We also will not promise that a link tool improves your position in Google. Rankings depend on far more than a link count, and every core update, like the ones Google shipped in March 2024 and November 2024, reshuffles the board in ways no vendor predicts.
Smaller scope, fewer lies. That is the deal we offer. If it is a fit, you will know within a single curl call, and the first four free actions let you test without spending anything.
Who the site is for, and who should look elsewhere
A link API suits some people very well and others not at all. Honesty saves time. Being honest about that saves everyone a week. The sections below describe the readers we have in mind, and the ones we would rather point to a different tool.
If you are not sure, start with the free actions. Start small. They cost nothing and show within minutes whether the data answers your question. A seo backlink api earns trust by answering the question you brought, not by promising more.
Developers who script link checks
If you already write small scripts for reports, the API will feel familiar. One URL, JSON in and out, and a credit meter in every response. You can drop a call into a cron job or a spreadsheet function without learning a new interface.
That is the reader we build for. The docs favour curl examples, because a command you can paste and read in one line is the fastest way to understand an endpoint.
Marketers who need keyword data
Keyword volumes, rank tracking and daily crawls are outside what we do. Ahrefs and Semrush cover them well, and paying for a suite can be the right call. We would rather say so than sell you a feature we do not have.
Small teams with a monthly list
A freelancer or a small agency often needs one answer a month. Which domains are strong, who links to a rival and which prospects are worth a second look. A per-call price suits that rhythm, because you pay for the month you actually work.
Testing before you build on it
Many developers ask whether anyone has used an API before they build on top of it. The fair test is your own. Run the free actions, then one small paid call on your own domain, and judge the response against what you need.
Keep a copy of the first responses. If the API changes later, that copy is the baseline you can compare against.

How to reach us
Questions about billing, bugs and data all start on the account page. Sign in there. It is your point of contact. Please note the action, the domain and the time of any problem.
Start here. Read the endpoint reference or the free DR checker. The home page has the shortest working example. Everything here is written by the SEOCheck editorial team. We publish no named staff and no customer quotes.
Plain text is fine. A curl command pasted into your request is the clearest way to show what happened.
How SEO and AI answers fit into our work
Search has changed shape over the last two years, and the change is still moving. Still moving. Generated answers now sit above many results. We do not claim to know how those answers are chosen, and we say so on every page where it matters.
What we can offer is narrower and steadier. Link data, measured the same way each month, that you can combine with your own observations.
Is SEO dying, or changing shape?
Our view is that the work is changing shape more than it is disappearing. Clear pages, real expertise and crawlable text still matter. The fine detail of how a model picks a source is not something we can confirm.
Where a link index helps with AI answers
A link index cannot tell you what a chatbot says. It can show you who links to you and to the brands that are named. That is one input among several, and it belongs next to your own prompt tests rather than in place of them.
What we do not claim
We do not claim that a link count predicts a mention, and we do not promise a place in any answer. If a vendor sells you a guaranteed route into AI results, the guarantee is the part to question first.
Why we keep the scope small
A small product is easier to describe honestly. We would rather do one thing reliably than promise a dashboard that covers everything and does none of it well. A narrow scope also means fewer claims to retract.
Decide in one afternoon whether we fit
Five small steps. None of them needs a card, and none of them commits you to a plan.
Sign in and take the free credit
Sign-in is with a Google account. The $100 credit appears on the account page, with no card needed.
Run the free actions
Call overview and stats on a domain you know. Both are free and need no key.
Test one 1-credit call
Send compare with your domain and two rivals. Check the _credits field to confirm the price.
Pull one 2-credit listing
Ask for refdomains on a rival, then compare the result with your own list. Decide whether the rows help.
Write down the decision
Note what the data answered and what it did not. If it does not fit, the same notes will help you pick a suite instead.
How we keep the product honest
Trust in a data product comes from plain statements about what it does and does not do. The rules below are the ones we apply to our own pages, so you can hold us to them.
The price list is the cost table
Every action has a credit cost, and the cost table in the endpoint reference is the price list. A call is billed once, not once per row. If a number on a page does not match the table, the table wins, and you should tell us through the account page.
Limits stated up front
The refresh cycle, the deep pagination cap and the lack of keyword data are all stated on the pages where they matter. We would rather you learn about a limit before you build on it than after a month of work.
Who writes the pages
Everything here is written by the SEOCheck editorial team, and we do not name individual writers. Each page carries a date that shows when it was last revised, so you can judge how current a claim is. Each page is written for one question, so you can read only the pages that match your work.
If a page disagrees with the responses you actually receive, trust the responses. Then tell us through the account page, and include the call that failed.