Artificial intelligence has quietly revolutionized customer support over the past two years, with companies routinely achieving deflection rates above 60%—meaning AI handles more than half of all customer inquiries without human intervention. Now, this same transformation is beginning to reshape sales teams, and the implications could be just as dramatic.
The concept of “deflection” in customer service refers to AI systems resolving customer issues automatically before they reach human agents. Rather than simply routing calls or providing basic information, these AI agents can access customer data, process transactions, and handle complex workflows independently. The results have been remarkable: companies are saving hundreds of thousands of dollars monthly while improving customer satisfaction.
This success story is now migrating from post-sale support to pre-sale interactions, with the first AI-powered sales agents entering the market. The question isn’t whether AI will impact sales—it’s how quickly sales teams will adopt the deflection playbook that has already transformed customer support.
The numbers from leading AI support platforms reveal the scale of this transformation. Decagon, an AI customer service platform that recently raised $131 million at a $1.5 billion valuation, reports that businesses using their technology achieve average deflection rates approaching 70%. Some companies, including language-learning platform Duolingo, are pushing well above 80% deflection rates.
Bilt, a credit card and rewards company, now handles 70% of their 60,000 monthly support tickets through AI agents, generating savings in the hundreds of thousands of dollars each month. Meanwhile, Intercom’s AI assistant called Fin achieves 86% resolution rates, managing everything from account changes to complex technical troubleshooting.
These aren’t the simple FAQ chatbots of years past. Modern AI support agents can navigate customer databases, apply business rules, process refunds and upgrades, and intelligently escalate issues that require human expertise. Gorgias, which provides AI customer service tools for e-commerce businesses, routinely achieves 60% deflection rates even among smaller companies with limited time to train their AI systems.
Enterprise software giant Salesforce reports that companies using their Agentforce platform see double-digit percentage increases in both customer satisfaction and deflection rates, with a 50% improvement in case resolution speed. Their Service Cloud offering helps businesses deflect 30% of cases immediately upon implementation.
The pattern is consistent across the industry. Zendesk partners report 70% ticket resolution rates through AI, while Swedish financial services company Klarna’s AI assistant now handles two-thirds of their customer service volume—equivalent to the work of 700 full-time human agents.
While support deflection has matured rapidly, sales deflection is just beginning to take shape. The earliest AI Account Executives—automated sales agents that can handle basic sales interactions—are starting to manage specific types of transactions that don’t require relationship building or complex negotiation.
Current AI sales agents excel at handling existing customer expansions, such as when a client needs to add five more software licenses to their existing subscription. These transactions involve minimal complexity and follow predictable patterns that AI can manage effectively.
Routine renewals represent another early deflection opportunity. Annual software subscriptions with standard terms and minimal customization requirements can often be processed automatically, freeing human sales representatives to focus on more strategic accounts.
Inbound product inquiries from qualified prospects also suit AI handling, particularly when customers seek specific information about features, pricing, or compatibility. These interactions often follow similar patterns and can be resolved with access to product databases and pricing information.
Demo scheduling has become another area where AI adds value, using intelligent routing to match prospects with the most appropriate sales representatives based on company size, industry, or specific product interests.
The parallels between support and sales deflection are striking. Five years ago, customer support teams worried that AI would eliminate their jobs entirely. Today, those same teams report that AI has actually elevated their roles by handling routine inquiries while humans focus on complex problem-solving and relationship management.
Jesse Zhang, CEO of Decagon, frames this transformation positively: “AI is often seen as destroying jobs, but at Decagon, we believe the opposite. Our AI agents are enhancing jobs, not replacing them.”
Support representatives have evolved into AI managers, spending their time configuring and overseeing automated agents rather than personally responding to every basic customer inquiry. Sales teams appear headed down the same path, though they’re roughly 18 to 24 months behind support in terms of AI adoption and sophistication.
This reframing helps clarify what organizations should expect from AI Account Executives. Rather than completely replacing human sales representatives, the technology focuses on deflection—handling routine sales interactions instantly and automatically, while escalating more complex opportunities to human representatives.
Forward-thinking sales leaders are beginning to identify their equivalent of support tickets—routine sales interactions that could be handled effectively by AI systems.
Low-value deals at standard list prices represent prime deflection candidates, particularly when customers have already researched the product independently and may prefer a quick, automated purchasing process over lengthy sales conversations.
Simple upsells to existing customers, such as adding standard features or increasing capacity limits, often follow predictable patterns that AI can manage without human intervention.
Straightforward renewal processing for annual subscriptions with minimal negotiation requirements can be automated, allowing sales representatives to focus on accounts that require strategic attention or custom arrangements.
Standard pricing requests for qualified inbound leads can be handled automatically, with AI systems providing accurate quotes based on prospect requirements and company pricing rules.
Basic product questions about feature comparisons and compatibility can be resolved through AI systems with access to comprehensive product databases, eliminating the need for sales representative involvement in preliminary research conversations.
Customer support’s journey toward high deflection rates offers a clear roadmap for sales transformation, suggesting a three-phase evolution.
Phase 1, where sales currently operates, involves AI handling basic inquiries and scheduling meetings. Most sales teams with AI capabilities are operating at this level, using automation for lead qualification and calendar management.
Phase 2, which appears to be emerging now, will see AI processing simple transactions and managing standard renewals. Early adopters are beginning to experiment with these capabilities, though widespread implementation remains limited.
Phase 3 represents the future state where AI manages entire deal cycles for routine purchases, deflecting the need for sales representative involvement until complications arise or strategic decisions become necessary.
Support’s experience suggests that deflection rates can increase rapidly once AI systems prove reliable. Companies like Duolingo didn’t gradually progress from 10% to 80% deflection—they experienced dramatic improvements once their AI agents demonstrated consistent performance across various scenarios.
The financial benefits of deflection create powerful incentives for rapid adoption. ClassPass achieved a 95% cost reduction in support conversation handling through AI implementation. For sales organizations, every routine interaction managed by AI rather than human representatives represents immediate cost savings and frees up valuable time for complex deal management.
Research from HubSpot, a marketing and sales software company, indicates that sales teams using AI tools are 1.3 times more likely to see revenue increases, with 40% of sales professionals already incorporating AI into their workflows. However, sales remains in early adoption phases compared to support’s mature AI implementation.
The economics become particularly compelling when considering the cost differential between AI interactions and human sales representative time. While a senior sales representative might cost $200 per hour when factoring in salary, benefits, and overhead, AI systems can handle routine interactions for pennies per transaction.
Based on customer support’s evolution pattern, sales deflection will likely follow a predictable timeline over the next several years.
During 2024 and 2025, AI will primarily handle inbound qualification and simple customer expansions, achieving deflection rates between 5% and 15% as organizations begin testing and implementing these capabilities.
The 2026 to 2027 period should see AI processing routine renewals and standard upsells more broadly, with deflection rates climbing to 20% to 35% as the technology matures and organizational comfort levels increase.
By 2028 to 2030, AI systems will likely manage complete transaction cycles for commodity-like sales, potentially achieving deflection rates between 40% and 60% for routine business.
This timeline assumes continued technological advancement and organizational adaptation similar to what occurred in customer support. However, sales cycles often involve more complexity and relationship management than support interactions, which may slow adoption in certain industries or deal types.
Sales leaders should begin preparing for this deflection-driven future by taking several concrete steps now, rather than waiting for the technology to fully mature.
Identifying deflection candidates requires analyzing current sales activities to determine which deals require minimal human judgment or relationship building. These transactions often involve existing customers, standard products, and predictable pricing structures.
Measuring baseline metrics helps establish current time allocation between routine and complex sales activities. Understanding how representatives currently spend their time provides a foundation for measuring AI impact and optimizing deflection strategies.
Piloting AI interactions with simple customer requests allows organizations to test capabilities and identify potential issues before broader implementation. Starting with low-risk scenarios builds confidence and reveals optimization opportunities.
Training teams for the transition ensures sales representatives understand how to work alongside AI agents rather than competing with them. This preparation helps smooth the adoption process and maximizes the benefits of human-AI collaboration.
Customer support achieved deflection rates between 60% and 80% by allowing AI to handle routine inquiries while human agents focused on complex problem-solving and relationship building. The first AI Account Executives are now applying this same approach to simple sales transactions, though the technology remains in early stages compared to support applications.
The companies that begin measuring and optimizing for deflection rates now will likely gain significant advantages as AI sales agents become mainstream tools. The question isn’t whether AI will eventually deflect a substantial portion of sales volume—customer support has already demonstrated this possibility. The critical question is how quickly individual organizations will adapt to this emerging reality.
This transformation won’t eliminate sales jobs any more than it eliminated support roles. Instead, it will likely elevate sales representatives to focus on strategic relationships, complex negotiations, and high-value opportunities while AI handles the routine transactions that currently consume significant time and resources.
The deflection model represents a fundamental shift in how organizations think about sales efficiency and resource allocation. Rather than hiring more representatives to handle growing transaction volumes, companies will increasingly rely on AI to manage routine interactions while human talent focuses on activities that truly require emotional intelligence, strategic thinking, and relationship management.
Join hosts Anthony, Shane, and Francesca for essential insights on AI's impact on jobs, careers, and business. Stay ahead of the curve – listen now!
Join hosts Anthony, Shane, and Francesca for essential insights on AI's impact on jobs, careers, and business. Stay ahead of the curve – listen now!
No hype. No doom. Just actionable resources and strategies to accelerate your success in the age of AI.
