Larridin measures AI use with a browser extension and a desktop app installed on employee computers (Larridin FAQ). That is one of five ways to measure how employees use AI at work. If you are looking for a Larridin alternative, the first question is which of the five fits what you need to know: shadow AI, adoption by team, cost, engineering delivery, or how AI is changing the way work gets done.
Worklytics wrote this guide, and Worklytics is one of the options. Each section says what the approach measures, how it is deployed, and where it falls short. For a side-by-side of the two products, see Worklytics vs Larridin.
| Approach | What it measures | How it is deployed | Best for |
|---|---|---|---|
| Workplace analytics from admin APIs (Worklytics) | AI adoption, depth, task type, fluency and usage-based cost across AI tools, next to meetings, focus time, collaboration and delivery | Read-only API connections; nothing installed on devices | Comparing AI tools by team and tying AI use to how work changes |
| AI vendors' own admin dashboards | Usage inside one tool | Built in; turn on in the admin console | A first look at a single tool |
| Engineering analytics platforms | Delivery metrics from Git, CI and issue trackers, sometimes with AI coding tool usage | Connect code and project tools | Engineering teams focused on delivery speed and quality |
| Endpoint and browser monitoring | AI sites visited, including tools IT has not approved | Browser extension or desktop agent on every employee computer | Finding shadow AI |
| Do it yourself: exports, BI and surveys | Whatever you export and model, plus self-reported value | Analyst time to export, join and maintain | Small rollouts or a one-off review |
Worklytics connects to the admin and usage APIs of your AI tools and of the systems where work happens: email, calendar, chat, documents, code, Jira and your HR system. Nothing is installed on laptops or phones. Personal identifiers are pseudonymized with SHA-256, results are shown only for groups above a minimum size, and prompt and response text is never stored.
Every major AI vendor now includes usage reporting for enterprise customers:
Where they fall short: each one covers its own tool. If you run more than one AI tool, you get several dashboards with different definitions of an active user, and none of them shows AI use by your own org structure or next to the rest of the workday.
Engineering analytics tools connect to Git, CI and issue trackers to measure delivery: cycle time, review time, deployment frequency and similar metrics. Several now add usage data from AI coding tools.
Where they fall short: they cover engineers only. Most AI use in a company happens outside engineering, in tools like ChatGPT, Claude, Copilot and Gemini.
This is Larridin's approach: a browser extension or desktop agent on each employee computer records which AI sites people visit. Its strength is finding unapproved AI tools by URL, which API-based tools cannot do.
Where it falls short: it needs software deployed and maintained on every computer, and Larridin's FAQ says iOS is not covered and Android needs a device management proxy (Larridin FAQ).
Most AI vendors let admins export usage data. You can join those exports to your HR data in a spreadsheet or BI tool, and add a survey to ask people what they use AI for and how much time it saves. Our guides cover the exports for ChatGPT Enterprise, Claude Enterprise, Claude Code, Microsoft 365 Copilot, GitHub Copilot, Cursor and Gemini.
Where it falls short: every export has its own format and history limit, the work has to be repeated each month, and surveys measure what people remember rather than what they did.
Some organizations combine approaches, for example an endpoint tool for shadow AI and Worklytics for adoption, cost and impact.
It depends on what you need. For shadow AI detection, other endpoint or browser tools. For adoption, cost and impact across several AI tools without installing anything, API-based workplace analytics such as Worklytics. For a single tool, the vendor's own admin dashboard.
Yes. Worklytics and the AI vendors' admin dashboards both work from admin APIs, so nothing is installed on laptops or phones.
Only for tools that have an admin or usage API. Finding AI sites that IT has not approved needs a browser extension or endpoint agent.
Worklytics reports usage-based cost for ChatGPT, Codex, Claude, Claude Code and Cursor by team and per active user. Vendor admin consoles show cost for their own tool only. See AI spend tracking across tools.
Want to see how the Worklytics approach works for your AI tools? Book a demo.
