A GitHub contribution graph is a prompt for questions about someone’s work, not a measure of how well they performed. It records only activity that meets GitHub’s counting rules, hides private work from most viewers by default, and says nothing about quality, design judgment, mentoring or impact. This guide explains what the squares do and don’t show, why some work goes missing, and how developers and managers can use the graph as one input alongside stronger evidence.
What the contribution graph actually shows
GitHub describes the profile graph as a record of contributions to repositories on GitHub. The profile shows a year of graph activity, plus a separate contribution activity timeline listing commits (including co-authored work), pull requests and issues. That timeline is the better place to look, because it points to specific work you can open and discuss rather than a color intensity. (GitHub Docs: Contributions on your profile)
As an Amazon Associate I earn from qualifying purchases.
Why contributions go missing
An empty-looking day, or a lower total than you expected, is not proof that no work happened. The causes fall into a few groups.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Commit eligibility rules
According to GitHub’s profile contributions reference, a commit counts only when all of the following hold:
#1 Best Overall
- The author email on the commit is associated with the GitHub account.
- The commit is in a standalone repository, not a fork.
- The commit is on the default branch (or on
gh-pagesfor a project site). - At least one relationship condition is met: you are a collaborator or organization member, you forked the repository, or you opened an issue or pull request in it.
The most common practical failure is a work laptop configured with an email address that is not linked to the account. Commits made on feature branches also won’t show until they land on the default branch.
Issues, pull requests and discussions
These have their own criteria. Items opened in a fork do not qualify under the stated rules, and GitHub notes limits on how many such items may appear. See the same reference for the details.
Rank #2
Private work
The graph defaults to public repository activity. A user can enable anonymized private contribution counts, but anyone without access to those repositories sees only daily counts, never what the work was. Much professional work is in private repositories, so a sparse public graph can sit alongside a very busy working year. (GitHub Docs)
Time and history rewriting
Profile contributions use UTC, so a late-evening commit in the Americas or an early-morning one in Asia may land on a different calendar day than the developer experienced. For commits, the profile uses the Git author date, while repository views use the commit date. Rebases, amends and force pushes can make those differ, so sequences may look out of order. (GitHub Docs)
Work that never reaches a commit
Design documents, incident response, architecture discussions, code review feedback, mentoring, planning, customer calls and work in other tools are not a green square. These are editorial examples of valuable work, not things GitHub claims to track.
Don’t confuse the profile graph with the repository contributors chart
The repository-level contributors graph is a different feature with different rules. It shows at most the top 100 contributors, excludes merge commits and empty commits, and may omit someone whose commits are not merged to the default branch or whose author email is not connected to their account. Numbers from one view should not be compared against the other. (GitHub Docs: Viewing a project’s contributors)
Why activity volume is a poor performance score
Productivity research is consistent on this. The authors of the SPACE framework (Forsgren, Storey, Maddila, Zimmermann, Houck and Butler, ACM Queue, 2021) write that developer productivity “is about more than an individual’s activity levels or the efficiency of the engineering systems relied on to ship software, and it cannot be measured by a single metric or dimension.” (The SPACE of Developer Productivity)
Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallLinkedIn’s Developer Productivity Framework, in its “Metrics and Performance Reviews” section, is blunter: “It is dangerous to use numbers representing the volume of output of a software engineer to determine their job performance—numbers like ‘lines of code produced,’ ‘number of changes submitted to the repository,’ ‘numbers of bugs fixed,’ etc.” (LinkedIn dph-framework)
The practical risk is incentive distortion: if squares matter, people split changes into tiny commits, avoid hard work that produces few commits, and spend less time on reviewing and mentoring, none of which helps the product.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.How to use graph evidence in a review
For managers and reviewers
- Establish scope first. Check which repositories and work types were visible, whether private contributions were enabled, and whether the work met GitHub’s counting rules.
- Anchor on role and assignment. Ask what the person was expected to do over the period. A platform engineer on incident rotation, a staff engineer writing designs and a developer on a greenfield feature will have very different graphs.
- Treat density as a question. Quiet stretches, bursts, review-heavy periods and co-authored work are all things to ask about, not to score.
- Ask for evidence through approved channels. Do not guess at the content of anonymized private squares. Request internal artifacts such as pull requests, design documents and tickets.
- Do not compare raw counts across people with different roles, codebases, or private/public work mixes.
- Give feedback with specific examples and outcomes, never a request to “commit more” to fill the graph.
For developers preparing their own review
- Check that the email on your commits is linked to your account in GitHub settings, and decide whether to enable private contribution counts.
- Pick a handful of pieces of work and describe the outcome: what shipped, what got more reliable, what a customer or teammate could do afterwards.
- Include non-commit work: reviews you gave, designs you wrote, people you helped, incidents you handled, documentation you improved.
- Explain unusual patterns yourself, such as time on leave, an on-call week, a long investigation that produced one small fix, or work in systems outside GitHub.
A multidimensional frame
SPACE names five dimensions: satisfaction and well-being, performance, activity, communication and collaboration, and efficiency and flow. It is a way to think about evidence, not a ready-made scorecard or universal rubric, so adapt it to the role and team. One workable layout for a review period:
| Question | Evidence that can inform it | What the graph contributes |
|---|---|---|
| Role and context: what was this person asked to do? | Assignments, team goals, on-call and support load | Nothing directly |
| Outcomes and quality: what changed because of the work? | Shipped or maintained systems, reliability, customer impact, review outcomes | Pointers to pull requests worth opening |
| Communication and collaboration | Code review, design discussion, mentoring, documentation, stakeholder feedback | Partial: issues, pull requests and co-authored commits |
| Efficiency and flow | Blockers, handoffs, tooling friction, interruptions | Little; patterns may prompt questions |
| Satisfaction and well-being | Conversations, workload, sustainability | Nothing reliable; do not infer from weekend or late activity |
| Activity | Commits, pull requests, issues, reviews | A partial, rule-bound view |
When comparing graph data itself, at minimum align the time window, repository scope, visibility setting, branch and account eligibility, and role before reading anything into a difference.
Free tools Windows power users keep installed
One-click scans. No signup required.
The Bottom Line
Use the green squares to start a conversation, then let outcomes, quality, collaboration and role context decide the review. A full graph does not prove strong performance, and a sparse one does not prove weak performance.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
