Case study · Toddle · Insights

School data, asked as questions.

Toddle runs teaching and learning for K-12 schools: planning, assessment, attendance, family communication. I designed Insights, the analytics layer that tells school leaders and coordinators what is happening across all of it.

RoleProduct designer PlatformWeb, admin and coordinator views ProgrammesIB DP, MYP, UbD Year2023–24

Pick a question. Each one is a real chart title from the product.

toddleapp.com/analytics/performance/year-groupDP

Year group analytics: a bar chart comparing the average IB total of Batch of 2022 (36.5) and Batch of 2023 (32.4), with a table of year groups below

85%

adoption within three months

92%

satisfaction

70%

use AI insights every week

Where it started

Analytics was one tile among many.

Toddle's home is a grid of modules: announcements, attendance, classroom review, calendar, family communication, progress reports. Every one of them produces data. Who read the announcement. Who turned in the essay. Who has missed nine days of school.

The data existed. What a head of school or a DP coordinator lacked was one place to read it, at the level they think at: the whole school, a year group, a subject, a class, one student.

The first explorations organised that place the way the database does: report types in a sidebar, then a list of students to click into. It was complete, and it made the leader do the analysis.

The DP Home screen: pastel tiles for Announcements, Attendance Dashboard, Classroom Review, Calendar, Family Communication, Progress Reports and Analytics, then DP core projects CAS, EE and TOK, then My classes — The DP Home a coordinator lands on. Analytics sits bottom left, next to everything it measures.

Scope

Four families of insight, one sidebar.

I split Insights by the question being asked, not by the data source. Each family has its own landing page and goes deeper from there.

User analytics

Admins

How often do users sign in to the platform?

StudentsStaffFamiliesVisitorsDaily / weekly

Usage analytics

Admins

Are people using the modules we pay for?

Unit planningTasks & LEsClass streamPortfolioAnnouncementsMessagingAttendance

Performance

Coordinators, leaders

How is the overall academic performance of DP like?

Year groupSubjectClassDP · MYP · UbD

Student insights

Coordinators, teachers

How has the student's overall performance been like?

OverviewGradebookAttendanceClass streamAI overview

The principle

Every chart is titled with the question it answers.

A label like "Read rate by recipient" asks the reader to work out why it matters. A question tells them before they look at a single bar. These are titles from the shipped designs, kept in the school's own words.

How is the overall academic performance of DP like? How often do users sign in to the platform? Are users reading the announcements sent? How has the student's overall performance been like? How are individual subjects performing in DP? How has year group performance over grading periods been like? How are the students performing in the year group? How are individual classes performing in DP? How is announcements being used across the school? How is the overall performance of MYP like?

Early exploration: a sidebar of report types and a plain list of students with their classes — EarlyReports named by type. The list shows who exists, not who needs attention.

Final year group analytics: a question as the chart title, batch averages, and the number of students at risk per year group — FinalThe question leads, the answer follows, and students at risk are counted in red on every row.

The decision

Three ways to move between modules. One won.

Usage analytics covers seven modules, and Toddle keeps adding more. I explored the same Announcements page three ways and judged each one by how it would hold up as the module count grew.

Analytics · Usage analytics · AnnouncementsType 2

Type 1: the module is chosen from a Module filter dropdown at the top right

Type 3: modules shown as a row of tabs above the page

Type 2: each module is an item nested under Usage analytics in the sidebar

Drill-down

From the whole school to one student.

Each level answers its own question and points at the next one. A coordinator can start at "how is DP doing?" and end on the one student who needs a conversation this week.

1Programme

Is DP doing well this semester?

How is the overall academic performance of DP like?

Two batches side by side, by IB grade or by score. The table underneath keeps the number that matters most on every row: students at risk.

Batch of 2022 36.5Batch of 2023 32.4At risk 24 and 12

2Year group

Which students in this batch need help?

How has year group performance over grading periods been like?

The batch page opens with the count, then the trend, then the names. Top performers and students at risk of failing sit next to each other, so nobody has to export a spreadsheet to find either list.

Students 120Avg. IB total 32.4At risk 24Top performers 18

3Student

What is going on with Albus?

How has the student's overall performance been like?

Quarter grades, presence, assessment and eCoursework completion, then every subject with weak quarters marked in red. The fourth quarter drop to 25 out of 45 is the first thing you see.

Q1–Q3 32 · 32 · 38Q4 25/45Presence 80%Assessments 75%

4Student · Attendance

Is it attendance?

The same student, one tab over. Attendance, gradebook and class stream share the student's header, so switching context never loses the person you are looking at.

Analytics · Performance · Year group analyticsLevel 1

Batch of 2023: performance overview, trend across grading periods, top performers, students at risk, subject averages and a distribution of students by IB total

Student insights for Albus Dumbledore: quarterly IB grades, presence, completion rates, subject performance table and completion charts

Student insights, attendance view for the same student

Outcome

Schools used it, and kept using it.

85%

Adoption in three months

Measured from launch across schools on Toddle.

92%

Satisfaction

From the leaders and coordinators using Insights.

70%

Weekly AI insights use

Seven in ten users come back to AI insights every week.