User Engagement Dashboard.

Build a user engagement dashboard for the analytics team at Trailmix, a fictional consumer habit tracking app with about 50,000 registered users on iOS, Android, and web. This is a read-only analytics view: no add or edit forms anywhere. Layout, top to bottom. A control bar with a metric toggle (DAU, WAU, MAU) and two filter chip groups: platform (All, iOS, Android, Web) and acquisition channel (All, Organic, Paid social, Referral). Below it, four stat cards: DAU (about 6,850), MAU (about 21,400), stickiness as DAU over MAU (about 32 percent), and week over week DAU change (about +2.1 percent). Then the hero: a large area chart of the selected metric per day across the last 26 weeks with a 7 day rolling average line. Next row, two charts side by side: a stacked area of daily actives by platform over the same period, and a donut of MAU share by acquisition channel. Finally a retention section: a multi-line chart of retention curves for the 8 most recent signup cohorts, percent retained from week 0 through week 8, plus a cohort grid beneath it where rows are cohorts, columns are weeks since signup, and each cell shows the percent with background intensity scaling to the value. Data to seed: one record per day for the last 26 weeks (about 182 records), each holding DAU, WAU, and MAU, with each of the three metrics split across the three platforms and separately split across the three acquisition channels (both splits sum to the same total), plus new signups by channel for that day; and 8 cohort records, each holding retention percentages for weeks 0 through 8. Keep values plausible for an app this size: DAU between 5,200 and 7,400 with gentle weekend dips. Plant this story: around 10 weeks ago daily actives fall roughly 15 percent for about two weeks, then recover. The stacked platform chart must show the dip comes almost entirely from Android (a bad release) while iOS and web hold steady. The one cohort that signed up during the dip retains about 10 points worse at every week than its neighbors, visible as the low line in the retention chart and the pale row in the grid. Behavior: the platform and channel filters recompute the hero chart and stat cards from the matching split in each daily record; stickiness and week over week change are computed, never stored. Default state shows DAU, all platforms, all channels. Visual identity: a warm amber accent on a soft warm-neutral base, used for active toggles, the hero area fill, and the retention grid intensity ramp.