Open source SEO tools, and what they actually replace
An open tool costs nothing in licence fees and moves the spending elsewhere: to a server, to a data source, to your time. This page gathers the ones that replace part of a paid tool presented here, and says which part, precisely.
The rule for getting in
One criterion only: the tool replaces, in whole or in part, a paid tool already presented on this site, and one can write what it replaces and what it does not. A useful open project with no paid equivalent here does not appear — not because it is worth less, but because this page exists to decide between two ways of spending, not to take a census.
A second check, added after nearly going without it: the vitality of the project. A public repository exists indefinitely, its licence reads perfectly, and none of that says whether anyone is still there to repair it. One open crawler was left out of this list for that reason alone, when it would have been its best entry.
What “open” does not mean
Open is not free, and free is not open. A licence says what you are allowed to do with the code; it says nothing about the price of the service. Several open projects sell a hosted version, and the most useful free tool in search — the engine’s console — is entirely closed.
Open does not mean reliable, it means legible. On a closed tool, a figure is either believed or rejected. On an open tool, the way it is computed can be read. That is a real advantage, and it is all of it: the method remains debatable, and the data may be bad.
What gets installed, and what gets queried
These tools fall naturally into two families. Some get installed: you run the software, and the data is the data you produce. Others get queried: they are public datasets you interrogate, with nothing to run.
The second family is absent from this page, and that is a research finding, not an oversight. The most-cited datasets in the field — public crawl archives, aggregated load measurements — declare no identifiable data licence, or declare one only for the documentation describing them, which is not the same thing. None could be presented here without inventing a fact.
The tools kept
advertools — MIT licence
It is an open library, not software you click: you write a few lines to crawl a site, read a robots.txt file or a sitemap, and you get a table back. It replaces part of a paid crawler, at the price of the script.
Lighthouse — Apache-2.0 licence
It is an open audit tool, published by the browser's vendor, which loads a page under conditions it sets and hands back scores per panel. It assesses one page at a time, in the lab, and replaces neither a crawler nor a measurement of real audience.
SerpBear — MIT licence
It is an open rank tracker, installed on your own server instead of rented. It records the positions of your phrases and keeps them. The software costs nothing; hosting and collecting the positions remain your responsibility.
Unlighthouse — MIT licence
It is an open tool that runs the browser's audit across all of a site's pages, discovered automatically, and gathers the results into a single dashboard. It fills the gap between a page-by-page audit and a crawler, without replacing either.
Where no open tool exists
None of those entries concerns link analysis, and the reason is not a passing one. What you pay for in a link tool is not the software: it is an index, built by a robot crawling the web continuously, over years. An open project can write the same software; it cannot equip itself with the same inventory, whose cost is the cost of crawling the web without stopping.
The same reasoning holds for the suites selling phrase databases and histories of results. Where the work consists of crawling or auditing, open catches up with paid without difficulty. Where it consists of owning an index, it does not — and saying so is more honest than offering an approximate substitute.
What open costs anyway
Three items replace the subscription, and none of them appears on an invoice: hosting when the tool has to run continuously, data access when the tool consumes data, and time — installation, updates, and the half-day of the day something stops. That last item is the only one nobody budgets in advance, and regularly the heaviest.
What you buy with a subscription is therefore not software, it is continuity: the name of someone to write to when the measurement stops. If nobody at your end wants to be that person, the paid tool remains the right choice, whatever the quality of the open alternative.
Frequently asked questions
- Is an open tool necessarily free of charge?
- No, and the confusion is constant. An open licence says what you are allowed to do with the code, not what the service costs: several open projects sell a hosted version. Conversely, the engine’s console is free of charge and entirely closed.
- Why so few tools in this list?
- Because a single criterion gets one in: it replaces, in whole or in part, a paid tool already presented here, with the substitution written down plainly. A directory of open projects without that criterion would fill up fast and help nobody decide anything.
- How do I check that an open project is still alive?
- By looking at three things before installing: the date of the last published release, the date of the last work on the repository, and the state of the project’s own site. An abandoned project stays perfectly installable and simply stops being repaired.
- Does open code guarantee reliability?
- It guarantees legibility, which is not the same thing. You can read how a figure is computed instead of believing it, and that is a real advantage over closed tools. It says nothing about whether the method is sound or the data any good.
Sources
- Licences and histories recorded from each tool’s public repository — Respective vendors and project maintainersaccessed on
- About Search Console — Google Search Consoleaccessed on
Block revalidated on