05 / IBM Training · 2023

Learning analytics dashboard

Program-manager view of learner outcomes across cohorts — from progress rings down to per-module drop-off.

Role

Senior Product Designer

Team

1 designer · 3 engineers · 1 PM

Duration

4 months

Platforms

Web dashboard

IBM Training — Learning analytics dashboard

Problem

Program managers had reports, not answers.

IBM Training's program managers received weekly PDFs. The PDFs answered 'how are cohorts doing?' but not 'where is a learner stuck, and what should I do about it?'

The team wanted a live dashboard, but the risk was drowning managers in charts.

Process

Design around three questions, not thirty charts.

  1. 01 · Question framing

    Interviewed program managers to distill the dashboard into three questions: who's at risk, what's blocking them, and which module needs a rework?

  2. 02 · Risk-first layout

    The 'at-risk learners' list is the top-of-page primary. Everything else is context around that decision.

  3. 03 · Module drill-through

    One click from a module tile takes you to the per-question drop-off inside that module — turning 'what's broken' into 'here's the exact quiz item.'

Results

Weekly PDFs quietly disappeared.

2.6×

at-risk interventions per week

−31%

cohort drop-off after 60 days

12

modules reworked in first quarter

Program managers ran more targeted interventions in the first month than the previous quarter combined. The module drill-through surfaced twelve high-drop-off quiz items that were rewritten within the first quarter.

Next case study

Employee learning home

IBM Training

Read case