NOTE: Demo visuals use either blurred real data or synthetic placeholders to protect customer privacy.

In a high-growth sales organization, understanding why deals are lost is just as important as understanding why they’re won. But as Abnormal scales, manually tracking and analyzing loss patterns across hundreds of proof-of-value (POV) opportunities has become nearly impossible.

That’s why Mickey built an AI-powered POV Loss Analysis Agent: a tool that automates postmortems on lost opportunities, providing real-time insights into customer objections, competitive dynamics, and internal improvement areas.

Abnormal runs a massive number of POVs, pilots, and opportunities every quarter. Naturally, not every deal closes, but figuring out why can be slow, subjective, and inconsistent.

At the individual level, there are several challenges:

• Sales reps are busy: Time spent filling out detailed loss reasons in Salesforce is time not spent selling.

• Bias creeps in: Loss data often reflects a rep’s perception rather than the full picture: recency bias, incomplete context, and rushed input all contribute.

• No centralized visibility: GTM operations teams can’t easily review every loss report or correlate patterns across deals.

The result is a gap between what’s recorded in CRM and the reality reflected in calls, notes, and customer feedback.

To bridge that gap, Mickey built an AI workflow using Nora and Slack that automates the process of researching, summarizing, and analyzing lost POVs.

The solution includes two main components. First, a dedicated POV Loss Slack channel allows users to specify any opportunity ID. The AI agent fetches associated data from Salesforce, Gong, and Google Drive, compiles related transcripts and documents, and produces a full loss analysis report.

Each report includes:

• Executive summary of the loss reason.

• Supporting quotes from Gong transcripts or sales notes.

• Root cause analysis (the “reason behind the reason”).

• Strategic recommendations to prevent future losses.
  

Secondly, an additional Slack workflow aggregates all the individual loss reports. The agent runs cross-opportunity analysis to identify patterns, like top recurring loss reasons, competitive intelligence insights, and distribution of controllable (internal) vs. uncontrollable (external) factors. Results are summarized into an interactive report that can be filtered by timeframe, product, or region.

In Mickey’s demo, he triggered the Slack command “Nora, make me a loss report for this opportunity.” Within minutes, the agent returned a fully formed analysis citing a budget freeze as the cause, supported by Gong quotes and contextual notes.

He then ran an aggregate report across 47 previous POVs. The AI produced a summary showing:

• Loss reason distribution by category.

• Key competitive trends.

• Common customer objections.

• What Abnormal did well and what could be improved.

• Strategic recommendations to address root causes.

The AI Loss Analysis Agent transforms how Abnormal’s GTM organization learns from every deal.

• Faster insights: Reports are generated automatically. No manual auditing or note-gathering required.

• Greater consistency: Every analysis follows the same structured format, eliminating subjective variation.

• Data-driven adjustments: Leaders can identify which issues are within Abnormal’s control and act quickly.

• Continuous improvement: Aggregated reports highlight how loss reasons evolve over time, supporting better forecasting and enablement.
  

Mickey’s workflow also dramatically reduces time-to-insight. What once required hours of manual review can now be produced in minutes, and updated continuously as new opportunities close.

The real innovation lies in turning ad hoc loss reviews into a living feedback loop.

Instead of isolated Salesforce entries, Abnormal now has a dynamic, AI-driven record of deal outcomes that evolves with every new POV.

The system can:

• Automatically trigger reports whenever a Salesforce opportunity moves to “Closed Lost.”

• Store and analyze results over time for longitudinal insights.

• Enable leadership to track how process improvements impact win rates.
  

By integrating data across sources and running continuous analysis, the AI agent shifts loss reporting from reactive to strategic.

What makes Mickey’s POV Loss Analysis Agent remarkable is its speed, scale, and simplicity. Built in just two days using Abnormal’s GenAI architecture, it demonstrates how business logic can live entirely in plain English prompts, not code.

It’s another example of Abnormal’s culture in action: empowering employees across every function to use AI to solve complex problems fast. By combining curiosity, creativity, and cutting-edge tools, Mickey turned a manual, repetitive GTM challenge into an automated intelligence system that learns with every deal.