Every company wants to save money with AI. Salesforce already did that. But here’s the part nobody’s talking about.
In just over one year, Salesforce’s AI agents handled 3 million customer support conversations, cut case volume by 8%, launched live multilingual support in 7 languages for the first time in company history, and saved $100 million in annualized costs. That’s the headline.
The real story is what came next.
How They Did It: Phase One
In 2025, Salesforce deployed AI agents across help.salesforce.com – their customer-facing support portal. The agents handled common customer questions autonomously, maintained conversational context, and escalated to humans only when complexity demanded it.
The results were immediate. As AI handled routine volume, human teams were freed to do what they’d always struggled to find time for: proactive support, relationship management, and complex problem-solving. Customer satisfaction held. Costs dropped. Case volume fell even as the customer base grew.
This is Phase One. It’s where most companies stop.
Phase Two: AI Turns Toward Revenue
In 2026, Salesforce asked a different question: if AI can remove cost from service, can it add revenue from sales?
They looked at their CRM and found what they quietly called ‘sawdust.’ In any large B2B company, marketing generates massive inbound interest that never gets followed up. Sales teams focus on the highest-scoring prospects. The long tail sits in the database, untouched. Too many, too low-priority, too expensive for a human to work.
So Salesforce deployed an AI agent on those dormant leads. It sent personalized outreach, asked qualifying questions, responded in context, identified genuine buying signals, and routed promising prospects to humans. In a short period, it influenced more than 3,200 new sales opportunities – from pipeline that had been completely written off.
Why This Actually Worked
Three principles made this successful: First, they started with a constrained, measurable problem – customer service. Second, they understood that AI’s real unlock isn’t doing the same things cheaper – it’s doing things you never had capacity for. Third, they connected Phase One and Phase Two: the same infrastructure built for support became the foundation for revenue.
I’m Mike Partners, and I started VisionarySchool.com to bridge the gap between enterprise AI strategy and small business reality. Here’s your action plan.
How to Apply This to Your Business
Think about this in two phases, just like Salesforce did. First, find your support sawdust – pull every FAQ and repetitive ticket from the last 90 days and deploy an AI tool like Intercom, Tidio, or HubSpot AI to handle them automatically. Once that’s running and freeing up your team’s time, move to phase two: dig into your CRM and pull every lead that went cold in the last 12 months. Use an AI-powered outreach tool like Clay, Lindy, or Apollo AI to send personalized re-engagement at scale. The real question to ask yourself isn’t just “what did I save?” but “what can I now do that I couldn’t before?” That’s where the transformative ROI lives.
Frequently Asked Questions
How much did Salesforce save with AI?
Salesforce saved $100 million in annualized costs by deploying AI agents that handled 3 million customer support conversations and cut case volume by 8% in just over one year.
How did Salesforce use AI to generate revenue, not just cut costs?
Salesforce deployed AI agents to re-engage dormant leads in their CRM – prospects that sales teams had written off as too low-priority. The AI sent personalized outreach, qualified prospects, and routed promising leads to humans, influencing over 3,200 new sales opportunities.
What is “sawdust” in the context of Salesforce’s AI strategy?
Salesforce used the term “sawdust” to describe the massive volume of inbound marketing leads that never get followed up because sales teams focus on the highest-scoring prospects. AI allowed them to work this long tail of untouched leads at scale. VisionarySchool.com breaks down strategies like this for business owners looking to apply enterprise AI tactics.
Can small businesses use AI for both customer support and sales?
Yes. Tools like Intercom, Tidio, and HubSpot AI handle support automation, while Clay, Lindy, and Apollo AI power automated sales outreach. Mike Partners recommends starting with support automation first, then expanding to sales once you see the initial results.
What was Salesforce’s key insight about measuring AI success?
Salesforce measured capacity gained, not just costs saved. The critical question wasn’t “how much money did we save?” but “what can we now do that we couldn’t before?” This shift in thinking is what led them from $100M in savings to millions in new revenue.
