Sales ops · Production

How Granicus Doubled Win Rates with AI-Powered Account Scoring

The problem

Granicus's account scoring model, in place since 2017, had grown stale — it was static, used only 5-10 variables, and did not leverage modern AI or machine learning, resulting in decreased sales team engagement and a narrowing gap between high- and low-target accounts.

First attempt

The existing static model with only 5-10 variables was no longer effective, causing decreased sales team engagement and eroding scoring differentiation between high- and low-target accounts.

Workflow diagram · grounded in source

Multi-source data ingestion

trigger

“Incorporates over 50 different variables, including CRM data, website information, federal agency data, and more”

ML propensity scoring

ai_action

“Utilizes advanced machine learning techniques to identify patterns and predict propensity to buy”

Salesforce real-time sync

integration

“Integrates seamlessly with Salesforce for real-time updates and easy access for sales teams”

Dynamic model refresh

feedback_loop

“Adapts dynamically as new data becomes available, ensuring the model stays current”

High-target account identification

output

“We're now able to identify high-target accounts for us that have sort of double the win rate from what our average is”

Reported outcome

The new AI-powered scoring model produced a 200% increase in win rates for high-target accounts, revealed that 25-30% of pipeline efforts had been directed at low win-rate accounts enabling better resource allocation, and drove adoption across sales, marketing, and executive teams.

Reported metrics

Win rate increase for high-target accounts200%

Pipeline spend on low win-rate accounts25-30%

Territory planning improvementImproved territory planning and go-to-market strategies

Model adoption across teamsIncreased adoption and trust in the model

Reported stack

PursuitSalesforce

Source

https://pursuit.us/case-studies/granicus

Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

The new AI-powered scoring model produced a 200% increase in win rates for high-target accounts, revealed that 25-30% of pipeline efforts had been directed at low win-rate accounts enabling better resource allocation,…

What tools did this team use?

Pursuit, Salesforce.

What results were reported?

Win rate increase for high-target accounts: 200%; Pipeline spend on low win-rate accounts: 25-30%; Territory planning improvement: Improved territory planning and go-to-market strategies; Model adoption across teams: Increased adoption and trust in the model (source-reported, not independently verified).

What failed first in this deployment?