Business Insights

Designing the Learning Layer for a Data-Heavy Product Feature

Quick Take

I designed a structured, in-product learning layer for a complex analytics feature that helps managers interpret sales performance, forecasts, and labor models — transforming a dense dashboard into a habit users return to as part of daily decision-making.

The goal wasn’t just comprehension. It was confidence, continuity, and repeat use.

The Product Feature

Business Insights is a data hub within a workforce management platform — surfacing sales trends, labor %, wage variance, budgets, forecasts, and labor models that inform smarter scheduling and operational decisions.

It’s the engine behind answering questions like:

  • Can we afford this schedule?

  • Where are we over or under budget?

  • What’s driving variance — and what should change next?

The Problem

Despite its power, the feature was under-adopted.

Users struggled to:

  • Understand what the data actually meant

  • Maintain accurate budgets and forecasts

  • Navigate a dense, multi-surface dashboard

  • Connect insights to real scheduling decisions

As a result, one of the platform’s most sophisticated features wasn’t becoming part of everyday workflows.

What I Discovered

Through user analysis and friction mapping, several patterns emerged:

  • Users didn’t know where to start

  • Core concepts didn’t reinforce one another

  • Budgets and forecasts became “set and forget”

  • Navigation increased cognitive load

  • Complex logic required scaffolding — not static documentation

The challenge wasn’t data availability.
It was sensemaking.

The Strategy

Design a habit-forming learning layer inside the feature itself — one that reduced cognitive load while reinforcing mental models over time.

The strategy centered on:

  • Foundations → Complexity: Build confidence before depth

  • Micro-steps, not macro-lessons: Teach one concept at a time

  • Concept repetition across workflows: Reinforce learning through use

  • Explaining why actions matter: Not just how

  • Cognitive load reduction: At every decision point

Learning needed to feel like guidance — not instruction.

What I Built

A system of 13 guided, in-product walk-thrus, including:

  • Business Insights Tour

  • Budget Setup and Management

  • Forecast Configuration

  • Sales Adjustments

  • Custom Metrics (Create / Edit / Delete)

  • Labor Models (Build / Edit / Iterate)

Each flow was designed to:

  • Teach a discrete skill

  • Reinforce a mental model

  • Encourage a repeat behavior

Together, they formed a cohesive learning layer — not isolated tutorials.

The Behavior Shift

The learning experience was intentionally designed to support these habits:

  • Checking insights before scheduling

  • Monitoring variance regularly

  • Keeping budgets and forecasts current

  • Revisiting and refining labor models

  • Tying decisions directly to data

  • Treating Business Insights as part of daily workflow

The goal was not mastery in one sitting — but ongoing fluency.

The Impact

Early signals showed:

  • Higher revisit rates

  • More accurate budget and forecast setups

  • Increased variance monitoring

  • Clearer mental models of how data connects

  • Fewer navigation-related issues

Even without sharing internal metrics, the pattern was clear:

When learning was embedded into the workflow, the feature became stickier.

My Role

I led the end-to-end learning and experience design, including:

  • User research and friction analysis

  • Content and learning architecture

  • Sequencing and mental model development

  • Microcopy and UX writing

  • In-product walk-thru build

  • Cross-functional collaboration with Product, Support, and Education

The result was a learning layer that didn’t explain the feature —
it made the feature usable.

Details have been generalized to respect confidentiality while preserving the structure and impact of the work.

Next
Next

UX Language Alignment for Enterprise Onboarding