Code search across every repository an organisation has, with precise cross-repository navigation and batch changes that open pull requests everywhere at once.
Do not build this at scale. Search across thousands of repositories is the product, and it is an indexing problem rather than a grep. Fast, structural, cross-repository search with references and definitions resolved means language analysis plus an index that stays current across a very large amount of code. At that size it is genuinely hard. FIRST, CHECK YOUR SIZE - A handful of repositories? `ripgrep` locally is faster than any hosted search, and it costs nothing - Your forge's own code search covers most cross-repository questions for free - Zoekt — the open indexer this category is built on — is available and runs happily over a few hundred repositories on one machine WHEN TO KEEP PAYING - Thousands of repositories, or a large organisation where finding every caller of a function is a daily question - Automated refactors across many repositories at once, which is the feature that genuinely saves weeks WHAT TO BUILD IF YOU WANT SOMETHING - Zoekt in a container, indexing your organisation's repositories on a schedule, with a small search page in front. That is an afternoon and it covers regular-expression and symbol search across everything - Add per-repository permissions if the code is not all open internally, checked in the query rather than filtered afterwards - A saved-search alert: notify me when a pattern appears anywhere. That is how you catch a deprecated call being reintroduced, and it is a cron job THE THING WORTH MORE THAN THE SEARCH - **A codemod runner.** A script that clones every repository, applies a transformation, runs the tests, and opens a pull request per repository. That is the actual value of large-scale code intelligence, and it is a few hundred lines with an abstract syntax tree library - Rate-limited, resumable, and with a dry run that shows the diff before anything is opened. Forty pull requests opened by accident is a bad afternoon for everybody THE ONE-LINE VERSION Count your repositories first. Under a few hundred, Zoekt and ripgrep cover it — and build the codemod runner, which is where the real saving is.
What you lose
- Indexed search across thousands of repositories that stays fast, which is a serious infrastructure problem
- Precise navigation across repository boundaries, built from compiler output rather than from text matching
- Batch changes that open and track a pull request in every affected repository
If you would rather not build
- GitHub code search, free and often enough
What it costs
as published on their pricing page
| Plan | Billed monthly | Billed yearly | Last read |
|---|---|---|---|
| — | $59/mo | — | — |
Their pricing page is where these came from. Seeing a different price? Tell us.
The escape hatch
open source · no votes, no paid placement
Zoekt
$0The fast trigram index behind Sourcegraph, usable on its own.
sourcegraph/zoektfree · open source
ripgrep
$0Fast enough across local clones that a service is unnecessary at small scale.
BurntSushi/ripgrepfree · open source
Why this verdict
our own opinion · changed only by a person
34/100
Verdict no at 34 above a certain size. Under fifty repositories ripgrep is genuinely better; the value appears exactly where building stops being feasible.
History
tracked since 10 Aug 2026 · nothing is ever overwritten
Questions about Sourcegraph
answered from the record above
Is Sourcegraph free?
No — the plan we track is $59 a month. Enterprise Starter from around $59 per user per month billed monthly; pricing above that is quoted.
Can you replace Sourcegraph by building your own?
KEEP IT. The value is the network, the data or the infrastructure. Keep paying. Replacement score 34 out of 100, build time longer than it saves. Read what you lose before you decide.
How much does Sourcegraph cost?
$59 a month on Enterprise Starter — $708 a year. Recorded 10 Aug 2026.
What do you lose by replacing Sourcegraph?
Indexed search across thousands of repositories that stays fast, which is a serious infrastructure problem; Precise navigation across repository boundaries, built from compiler output rather than from text matching; Batch changes that open and track a pull request in every affected repository. If any of those carry weight for you, keep paying.
Is there an open-source alternative to Sourcegraph?
Yes: Zoekt, ripgrep. The prompt on this page is for when you want it your way instead.
Related entries
same category first, most replaced first
Every week, something stops being worth paying for.
New verdicts, prices that moved, entries added. One email a week. Unsubscribe in one click. Nothing is being sent yet — your address is kept here, and the first issue is the first thing it is used for.
free forever · no tracking pixel · stored here, never passed to anyone

