Salesforce, Momentum, Mangomint Share AI Sales Lessons

The Gist

• Salesforce admits AI rollouts are bumpy even with messy CRM data
• AI excels at following up on neglected leads that reps cherry-pick
• Start with AI where data is cleanest: support and ops, not sales

• AI is easier to deploy on the service and ops side than on the sales side due to structured data and defined processes.
• Salesforce's Agentforce rollout faced challenges due to unclean data and unprepared processes, highlighting common AI deployment issues.
• AI agents can consistently follow up on leads, enrich data, and surface opportunities that humans might ignore.
• Clean CRM data is essential for AI effectiveness, as AI cannot reconstruct missing or unstructured data.
• Companies winning with AI in sales invest heavily in ops and rev ops to manage data and architect systems.
• AI can significantly improve retention by flagging dissatisfaction signals in real time and automating support tasks.

• Start AI deployment in areas with the cleanest data, such as support or operations, rather than sales.
• Ensure all sales and customer data flows into a single source of truth (e.g., CRM) before deploying AI agents.
• Invest in operational infrastructure and rev ops to support AI systems and maximize efficiency gains.
• Use AI to improve retention by automating support tasks and flagging dissatisfaction signals in real time.

• 35 logos a month (Marchelle's top rep's monthly performance in closing deals.)
• 7x ARR-to-OTE ratio (Marchelle's top performers' performance against quota in SMB SaaS.)
• 40% (Percentage of customers in financial services who want more personalized outreach from vendors.)