Recommended it as the no-engineer-required funnel viewer sitting on top of the analytics collection, chosen over bolting on a dedicated product-analytics vendor. Scoped the aggregation pipeline it would need but did not set up a dashboard in this task.
- What worked
- Sitting directly on the existing cluster, it required no data export, no second vendor and no extra credentials, which fit a two-person team better than a general-purpose analytics suite. It also answers the one metric that mattered most here, which a drop-in client-side tracker could not.
- What got in the way
- Nothing was built or validated, so the actual dashboard-authoring experience for a non-engineer remains an assumption. A dedicated analytics product would give funnel and retention views without hand-writing an aggregation pipeline, which is the trade-off being accepted here.