How Lily AI went from 30 hours of pipeline reviews to trusted data

Overview

Lily AI is an AI-driven company that helps large retailers connect shoppers with the right products. When Julian Dimery joined as Head of Sales at the start of 2024, his mandate was clear: scale the sales machine, hit ambitious yearly targets, and build the kind of repeatable process that grows with the company.

What he found was a sales org running almost entirely on word-of-mouth. No formalized methodology. No stage-gating. Deals qualified by opinion, not data. And a team spending close to 30 hours a week in pipeline review meetings just trying to understand what was actually happening in the field.

[testimonial]

[list#base]

Before Backstory

• No formalized sales process or methodology - deals qualified by rep opinion
• 30 hours per week consumed by pipeline review meetings across reps and leadership
• 10 sales stages with no enforcement criteria to move deals forward
• Deals going stale or perpetually pushing out with no mechanism to flag or cut them
• Forecasting unreliable - leadership making revenue calls without confidence in the data

[/list]

[tag#Use case 01]

MEDDPICC and standardized deal qualification

Giving sellers a consistent framework - and making deal health visible to everyone

Dimery's first priority was replacing opinion-based deal reviews with a process everyone followed. He worked with his Backstory CSM to build Opportunity Scorecards in Salesforce based on MEDDPICC, customized to Lily AI's specific sales motion. Just as importantly, he introduced stage-gating. Ten stages became four. The scorecard became the agenda for every 1:1, every team meeting, every pipeline review.

[testimonial]

[list#brand]

AFTER BACKSTORY

• MEDDPICC Opportunity Scorecards embedded in Salesforce and used in every deal review
• Stage-gating introduced - deals must meet defined criteria before advancing
• 10 sales stages consolidated to 4, removing ambiguity from the pipeline
• Deal health visible at a glance - gaps surfaced in the scorecard, not discovered at close

[/list]

[tag#Use case 02]

Forecasting accuracy and pipeline hygiene

From 30 hours of meetings to a forecast leadership can actually trust

The forecasting problem at Lily AI was downstream of the qualification problem. When deals weren't consistently scored, the pipeline couldn't be trusted - and if the pipeline couldn't be trusted, every revenue call was a guess.

With Opportunity Scorecards as the foundation, Lily AI built a report that uses scorecard completion percentage to weight the ACV of each deal. Leadership can now see through deals that look good on paper but lack the qualification depth to close.

[list#img]

AFTER BACKSTORY

• Forecast weighted by scorecard completion - deal ACV adjusted by qualification depth
• Pipeline reviews reduced from 30 hours per week to a few hours - with better output
• Stale deals surfaced and cut earlier - time-in-stage tracked and flagged automatically
• Revenue calls made with confidence - leadership trusts the data behind the number

[/list]

[tag#Use case 03]

Relationship Maps and enterprise buying groups

Mapping complex stakeholder networks at large global retailers

Lily AI sells to large retailers where buying decisions involve multiple departments, geographies, and levels of seniority. Managing those relationships without a clear picture of who's in the deal meant deals could stall at any stage for reasons that weren't visible until it was too late.

Backstory's Relationship Maps gave Lily AI's sellers a structured way to visualize buying groups, track stakeholder relationships, and identify gaps - including who at the executive level needed to be looped in.

[list#accent]

AFTER BACKSTORY

• Buying groups mapped visually - relationships between stakeholders tracked across every deal
• Gaps in stakeholder coverage identified proactively, not discovered when deals stall
• Leadership and founder network deployed strategically based on map insights
• Better multi-threading across large global retail accounts

[/list]

What Lily AI achieved

In under six months, Lily AI moved from a sales org dependent on anecdote to one that runs on data. Thirty hours of weekly pipeline meetings collapsed into a few focused hours. Win rates improved in the first quarter of use. Sales cycles shortened. Forecasts are built on scorecard-weighted deal data rather than rep optimism.

[testimonial]