GitClear's Data Deep Dive agent provides a chat-based interface to find the metrics that are most relevant to any question you have about your team's data.
linkWhat can it do?
It built to answer any question you might have about the 100+ metrics that GitClear has collected in the API and described in the Developer Analytics Encyclopedia.
Some common questions:
Is AI-assisted code increasing our volume but decreasing code quality or long-term durability?
How does our engineering output compare before and after adopting AI coding tools?
How much engineering effort goes to new features versus maintenance and bug fixes?
What is the ROI of our investment in developer tooling and infrastructure?
How fast is my team, or an individual developer, producing durable code that advances the project?
How much code is being written and then rewritten or discarded?
How long does it take to go from first commit on a feature to it being in production?
Is engineering output increasing, flat, or declining over time?
Are we delivering more customer value with the team size we have?
The agent has been trained on about 50 types of questions. As shown in the screenshot, it will return CSV data that you can download to prepare an analysis, or you can visit the pages that implement AI usage, velocity, and various cohort reports.
linkLimits per subscription level
Questions are counted per month.
Trial level | Queries in trial |
Basic | 1 |
TrialPLUS | 10 |
Subscription level | Queries per month |
Starter | 3 |
Pro | 10 |
Elite | 50 |
Enterprise | 200 |
linkSharing a Data Deep Dive
You can always revisit a past Data Deep Dive by visiting its home under the "Work in Review" tab. When you ask a new question, if you click the "Share" button near the top of the question window:
Then you'll get a link inserted into your clipboard that can be pasted into a group chat or email to allow others to see the data you're seeing.
linkRecommended use
GitClear's Starred Reports allow a convenient one-stop tab to follow the metrics that matter most on a per-team basis.
When you ask a question to the agent, it will highlight several metrics most relevant to your query. When you're seeking to build out your personal collection of Starred Reports, the fastest way to find relevant charts is often to visit the charts that are referenced by the Data Deep Dive Agent.
Once you've collected your preferred reports together, then you can leverage Outlier View to understand which teams are leading the way in code quality, PR/issue solve rate, sprint velocity, etc.
linkExample use video
See how it works in a one-minute micro demo: