AI Web & Product Analytics Dashboard from CSV

The state of product analytics for most companies is: an analytics tool licence (PostHog, Amplitude, Mixpanel) that costs five or six figures, a handful of dashboards built by someone who has since left, and a steady stream of Slack DMs asking 'can you pull X for me?'. The reason is unchanged across tools — building a new dashboard requires a query language, the query language is owned by data, and data is busy. Sreniq breaks that loop by reading raw event exports directly. The PM does not learn SQL; the PM uploads an event CSV and asks 'what is the D7 retention for users who completed onboarding step 2?'.

DAU / WAU / MAU and the stickiness ratio (DAU / MAU) are computed from the same event table; Sreniq calculates both as standard outputs. Funnels are user-definable in chat — describe the steps in plain English, get a conversion-per-step bar chart and the absolute drop-off counts. Retention curves are produced by signup cohort, with the typical D1 / D7 / D30 columns and a heatmap if you want a longer view.

Feature-adoption analysis is the other workflow that pays for itself. After a launch, the question 'what fraction of weekly active users tried the new feature in the first 14 days?' is the leading indicator of whether the feature will stick. With a single event export filtered to the launch window, Sreniq can answer it plus the obvious follow-ups — adoption by plan tier, by cohort, by user persona if the column is in the file. None of these require a new dashboard build.

Event tables get large fast. Sreniq supports files up to 1 GB, but for most product analyses pre-aggregating to daily user-event counts before upload makes the chat snappier and the dashboard quality is identical. For long-running event tables, exporting one month at a time and consolidating in chat is the typical pattern.