Case Study
Strengthening Usage Intelligence for EdTech Product Renewal Decisions
Key Result Highlights
- Delivered an evidence-backed inventory of 1K+ source files and ~700 interactions, giving the client a clear view of what was being measured.
- Identified account-resolution logic affecting the attribution of 10M+ usage events across large institutions, consortia, library systems, and multi-campus accounts.
- Created a phased implementation roadmap for governed reporting within the Databricks and Power BI stack the client already used, without requiring a new analytics platform or additional licensing.
- Provided the client with a clearer basis for strengthening renewal reporting, account-level analysis, and future usage-intelligence initiatives.
The Client
The client is an educational reference and digital media publisher serving universities, K–12 schools, public libraries, and multi-institution consortia in the US.
The Challenge
As the client continued consolidating products into a unified platform, usage information needed to be interpreted across LibLynx, Google Analytics 4, Databricks, Power BI, product migration states, and multi-level account relationships. They wanted to define a governed reporting foundation that could support consistent renewal decisions as the platform evolved.
Critical Success Parameters
- Validate the collection and attribution of records for referral or discovery activity.
- Trace each measure from the original user action from collection to reporting.
- Distinguish genuine changes in usage from differences caused by instrumentation, platform migration, or account hierarchy.
- Prioritize tracking changes that improve renewal and product analysis.
- Establish clear definitions, confidence ratings, ownership, lineage, quality controls, and monitoring requirements.
- Build the recommended future state within the client’s existing Databricks and Power BI environment.
Our Approach
- Developed testable hypotheses for duplication, attribution, and reconciliation between analytics sources.
- Scanned platform source files and interactions for a full activity measurement inventory.
- Reproduced interactions in live sessions and inspected LibLynx and GA4 network traffic to cross-check against real platform behavior.
- Traced key measures through instrumentation, collection, Databricks processing, account attribution, and downstream reporting.
- Reviewed production pipelines, notebooks, source schemas, migration signals, and master-recipient account relationships.
- Designed a phased Bronze, Silver, and Gold data architecture with ownership, lineage, scheduled quality checks, and failure alerts, using the client’s existing technology stack.
The audit combined full-code analysis with live verification. Recommendations covered immediate priorities and architectural improvements.
Need Similar Results?
Talk to our team to see how we can help.