Skip to content

Stockertim Dictionary · Investing

Version control

Also called versioning, revision control, version history

In short

Keeping each saved version of a file or model, with a note of what changed and why, so that any published version can be found and compared later.

  • Advanced
  • Concept
  • Checked

In detail

Research models and reports change constantly: new results arrive, assumptions are revised, mistakes are fixed. Without a record, it soon becomes impossible to say which version produced the figure quoted to a client last month, or why the valuation moved. Version control solves this by keeping every saved version, labelled with the date, the author and a short description of the change.

In software, dedicated systems track every edit to code line by line, allow several people to work in parallel and merge their changes, and can restore any earlier state. Analysts increasingly use the same tools for models written in code and for data pipelines. For spreadsheets and documents, simpler methods are common: a numbering scheme, a change log tab, a shared drive with history switched on, and a rule that published versions are frozen and never overwritten.

The benefits are reproducibility, accountability and safety. Any published number can be traced to the exact model and data that produced it, reviewers can see what changed between versions, and an error introduced today can be undone without losing other work. In regulated firms, keeping such records also helps answer questions from compliance teams, auditors and supervisors.

Version control works best when combined with dated inputs and recorded sources, so that both the model and the data behind it can be recovered.

Worked example

A change log for a valuation model of Kestrel Foods.

  1. v1.0, 1 Aug 2026: first published, value ₦52 a share
  2. v1.1, 15 Aug 2026: fixed a tax-rate error, value ₦50
  3. v2.0, 20 Oct 2026: updated for third-quarter results, value ₦55

A reader asking why the value changed from ₦52 to ₦55 can see both reasons and open each version.

In practice

Software teams and quantitative researchers use distributed version-control systems as standard, and government digital services require code to be kept under version control. Banks' model risk policies expect models used for pricing or capital to be versioned and change-controlled.

Watch out

  • Files named final, final2 and final-really are not version control; record what changed and why.
  • Versioning the model without the input data still leaves results hard to reproduce.

Learn it properly

Official sources

Facts last checked on 4 November 2026.

Education, not advice: the Stockertim Dictionary explains what terms mean. It never recommends buying or selling anything.

English