NOTE: Demo visuals use either blurred real data or synthetic placeholders to protect customer privacy.
Reclaiming Time for CSMs
Before this project, CSMs faced constant friction just finding reliable information. Their work depended on a patchwork of old docs, Slack threads, and spreadsheets, which made prep time unpredictable and error-prone. Each QBR could take hours of manual digging.
Top frictions included:
- Fragmented knowledge across multiple sources with no apparent authority.
- Time-consuming QBR prep that could stretch from 2 hours to a full afternoon.
- Repeated customer questions about integrations or product features that required manual research.

One Source of Truth for Success
The Customer Success Companion brings everything into a single Glean-powered experience. CSMs can now query natural-language prompts like “Show me risks for Yale University’s next QBR” or “How does our Okta integration work?” and receive verified, contextual answers from active sources such as Confluence, Highspot, and Salesforce.

Key capabilities include:
- QBR Ready Summary pulling data from Gong, JIRA, and Salesforce to generate executive summaries and account health snapshots.
- Product Support Lookup that filters for verified knowledge only from maintained sources.
- Product Release Tracker summarizing weekly roadmap changes from Google Docs and Slack.
- Data Logic Helper connecting to Databricks dashboards to explain metric discrepancies.
As Noam described, “It goes from a couple hours of prep to five to ten minutes.”
Together, these features turn the Companion into a daily launchpad for CSMs. It’s already become the default homepage inside Glean.

Measurable Impact for Teams and Customers
Since its October launch, usage has surged: 300-plus runs and adoption by roughly 80% of CSMs each month. Feedback calls it “essential” and “priceless” for understanding account history and surfacing upsell opportunities.
Key impact highlights:
- 80% active user rate across Customer Success.
- Hours reclaimed weekly from reduced QBR prep.
- Improved data trust with all sources verified and current.
- Better customer outcomes through quicker, more confident responses.
Next step: integrating Gainsight and Slack product feeds to extend the Companion’s reach across the CS tech stack.
From Feedback to Flywheel
Early users describe the tool as “instrumental” and “the first place I go in the morning.” The feedback loop is tight: CSMs share real-time insights via a dedicated Slack channel, and Ellie’s team rapidly iterates on new data integrations. The culture around it reflects Abnormal’s builder-first mindset: start with user pain, ship fast, learn faster.
The Companion is now a model for how AI can serve internal teams as effectively as external customers. Next up: tighter Gainsight integration and expansion into post-QBR analytics, continuing the cycle of trusted, efficient customer engagement.
