← Back to overview
CLASSLINK · ANALYTICS+ · DATAVIZ RESKINMOCKUP FOR TODAY'S CRIT

School Comparison, reskinned two ways

Built from Faina's four chosen sketches — Pulse tiles / Bento board for the hero row, Strips / Bubble grid for the comparison table — using the real Frost tokens (cl-protoapp-components) and the sketches' own computed styles, not approximations. Today's real numbers from schools/comparison.

Why this looks like this — two PMs described a data-trust problem, not a visual one, and a third handed us Stephen Few's preattentive-attribute and color-function framework to design against instead of a taste call. Full feedback & framework →  ·  This page's measurement plan →
01Hero KPI row — two finalists, your call

Live page today: 197m time spent, 59 digital resources, 8.3K students & staff, 20.5K logins — four flat-colored cards, one number each, no trend. One tile is shown selected (the sketches' own "tabs" convention — clicking a tile would move this state, not add a new color per card).

Dashboards D

Pulse tiles

Small multiples, one visual language — every card carries a living wave baseline and a pinned peak ring, so a trend is visible before anyone reads the number.

Time Spent
197m
Digital Resources
59
Students & Staff
8.3K
Logins
20.5K
Dashboards H

Bento board

Each measure keeps the same mini bar-chart form — the sketch's own restraint, not four different icon metaphors — with plum reserved for the one card that already represents two roles combined.

Time Spent
197m
This Month
Digital Resources
59
This Month
Students & Staff
8.3K
This Month
Logins
20.5K
This Month
02Students · Staff · Total — two finalists, your call

The live table is already a blue heatmap, but every cell is colored — nothing marks which school actually leads. These keep the ramp only for the one standout cell per column (Strips) or encode it redundantly in size and shade (Bubble grid), straight from the sketches' own technique. Adapted from a 7-day grid to this page's 3-metric grid; rows ordered by Total, busiest school first.

Heatmaps B

Strips

Position beats color — one common 0–100% scale per column; only the column's single busiest school keeps the dark ramp color, the rest stay light.

School ComparisonThis Month
SchoolStudentsStaffTotal
Polk High School
2h 36m
2h 22m
4h 44m
Roosevelt Elementary
3h 3m
1h 17m
4h 5m
Lincoln Elementary
2h 18m
54m
4h
Madison Middle School
4h 42m
1h 13m
3h 50m
Adams High School
2h 33m
1h 55m
3h 24m
Washington Middle School
3h 52m
1h 8m
3h
Kennedy Middle School
3h 49m
2h 21m
2h 42m
Jefferson High School
3h 4m
1h 2m
2h 34m
Truman Elementary
2h 44m
1h 6m
2h 21m
Jackson Elementary
3h 15m
1h 17m
2h 7m
Only Madison (Students), Polk (Staff) and Polk (Total) keep the dark ramp color — one standout per column, same technique the sketch uses for a row's busiest day.
Heatmaps D

Bubble grid

Redundant encoding — radius and shade both say the same thing (area ∝ value, not diameter), so it still reads for colorblind viewers.

School ComparisonBubble size and shade = share of column max
SchoolStudentsStaffTotal
Polk High School
2h 36m
2h 22m
4h 44m
Roosevelt Elementary
3h 3m
1h 17m
4h 5m
Lincoln Elementary
2h 18m
54m
4h
Madison Middle School
4h 42m
1h 13m
3h 50m
Adams High School
2h 33m
1h 55m
3h 24m
Washington Middle School
3h 52m
1h 8m
3h
Kennedy Middle School
3h 49m
2h 21m
2h 42m
Jefferson High School
3h 4m
1h 2m
2h 34m
Truman Elementary
2h 44m
1h 6m
2h 21m
Jackson Elementary
3h 15m
1h 17m
2h 7m
Radius ∝ √(share of column max), same formula read off the sketch's rendered bubbles — not a free-hand size.
OPEN QUESTION FOR THE ROOM
The Analytics PM's ask was "simplify so people trust the numbers." Strips reads faster at a glance; Bubble grid is the safer pick if we also care about colorblind access. Does one matter more for District Admins than the other?