Before we reskin Screen Time Overview, let's define what "better" means
A measurement plan for screen-time/v2/overview — page two on the dataviz reskin to-do list, built the same way as the School Comparison plan: from Vitaly Friedman's "How to Measure UX" and the open questions on design‑crit‑dataviz. The stakes are higher here — this page is a wellbeing/duty-of-care surface, not just a comparison tool.
Screen Time Overview is the project's "number with no context" problem in its purest form. The Learning Time bar reads 90%, but the page text separately says it "meets the 75% goal" — that goal is never drawn on the chart. Marathon Sessions (11%) and Late Night Learners (1%) show no benchmark at all. And "274 students need attention" has no denominator — 274 of 900 reads very differently than 274 of 8,300.
The stakes are higher than Comparison, too: these numbers describe student wellbeing, not engagement. Same method as page one — test today's page before anyone sketches a replacement.
Step 1 of the strategy. Run these on today's live page, before any sketch is chosen. This is the number every reskin direction has to beat.
| Task we give the administrator | What it actually tests | Why it matters here |
|---|---|---|
| "Is the district currently meeting its Learning Time goal, and by how much?" | Missing context | The 75% goal exists in the data ("meets the 75% goal") but is never drawn on the bar — the purest case of "a number with no context" in the whole product. |
| "Roughly what share of all students does the '274 students need attention' group represent?" | Missing context | No denominator is shown anywhere on this page. 274 could be alarming or trivial depending on total enrollment. |
| "Is 11% 'Marathon Sessions' something this school should be worried about?" | Missing context | No threshold or comparison point is given for either watch-list metric — admins have no way to judge severity from the number alone. |
| "Find the single grade — not grade band — with the lowest Learning Time percentage district-wide." | Content success | Requires expanding all four accordions. Tests whether switching from bars (bands) to plain numbers (grades) still lets people compare quickly. |
Step 2. Once "Heroes & KPI headers," "Gauges" (the goal-bar pattern) and "Bar & column" directions are narrowed to 2–3 each, don't pick with a vote — re-run the same loop.
Step 3. This is the bridge Friedman says is usually missing — connecting a design initiative to a business goal the org already tracks, through metrics we can actually measure.
Step 4. Applied to this page specifically, feeding the design-crit site's own "prove it beyond one page" step.
Step 5. This page already pairs green (good) with amber (watch) instead of red — good evidence for the colorblind-safe status pattern design-crit flagged as open.
Run a colorblind simulation on this exact screen, with a goal-line treatment added to the bar, before calling it final. Still no donuts — the Learning Time measure already proves a non-donut "percent of goal" pattern can work.
- 1Do we approve the 4-task baseline on the current Screen Time Overview before sketching it?Decision: pending
- 2Should every "watch" metric (Marathon Sessions, Late Night Learners, needs-attention count) require a visible benchmark or denominator in the reskin, not a nice-to-have?Decision: pending
- 3Which "Heroes," "Gauges" and "Bar & column" directions from round one do we adapt for this page's goal-bar and grade-bar charts?Decision: pending