Instead of replacing third-party tools for initial emails, a custom-built agent excels at the follow-up stage. By integrating all first-party data (past event attendance, call transcripts, proposals), it crafts hyper-contextual follow-up emails that off-the-shelf tools cannot replicate.

While AI can speed up outbound campaigns, its greatest opportunity is fixing the 'last mile' where customer experience often fails. By automating the next interaction after a lead responds, AI ensures no touchpoint falls through the cracks, making communication more complete rather than just faster.

If you use AI to fully automate outreach without personal effort, you haven't earned the right to ask for a prospect's valuable time. Instead of using AI as a silver bullet for mass messaging, leverage it for deeper research to understand what a specific persona at a specific account truly cares about.

The best use of AI in sales is not to automate generic emails, which can make reps lazy. Instead, it should be used as a research and efficiency tool to triple the number of hyper-personalized, human-centric calls a salesperson can make daily.

AI models like Claude produce ineffective outbound because they can follow infinite plausible-but-wrong reasoning paths. Without strict, hierarchical rules (e.g., "the cause for meeting is most important"), the AI defaults to templates that sound good but lack a strategic reason for the prospect to respond, resulting in a 0% reply rate.

Standard AI sales agents are limited to CRM data. A custom-built agent for renewals can outperform them by pulling from a company's entire data universe—website, social media, podcasts, even founder emails—to create hyper-personalized, context-rich pitches that third-party tools cannot match.

A powerful two-stage workflow combines different AI tools. Use scalable, off-the-shelf AI SDRs for initial outreach to get a response. Once the prospect engages, deploy a custom, context-rich agent to deliver a hyper-personalized pitch or deck as the more impactful follow-up.

Many companies fail with AI prospecting because their outputs are generic. The key to success isn't the AI tool but the quality of the data fed into it and relentless prompt iteration. It took the speakers six months—not six weeks—to outperform traditional methods, highlighting the need for patience and deep customization with sales team feedback.

Off-the-shelf AI go-to-market tools fail because they are purely transactional. ElevenLabs' CRO built custom AI agents for SDRs, proposals, and customer success that assist humans by drafting personalized messages, which are then reviewed, sent, and used to fine-tune the models, leading to actual revenue generation.