In a recent LinkedIn post, sales enablement expert John Barrows discusses a critical flaw he observes in how many professionals are currently utilizing artificial intelligence tools. Barrows argues that the effectiveness of AI is being hampered not by the tools themselves, but by the poor quality of context provided to them.
Barrows highlights a common practice among sales representatives and leaders: exporting call transcripts, copying them into AI models like ChatGPT, and then asking for summaries or follow-up actions. He categorizes this approach as a “workaround” rather than a robust system, leading to significant drawbacks.
“That’s not a system. That’s a workaround. And it creates two big problems: 1) You lose context across conversations and 2) You now have sensitive data floating around in random places.”
According to Barrows, this method results in a loss of crucial context across multiple interactions and introduces security risks by scattering sensitive data. He points out that this approach prevents AI from building a comprehensive understanding of a client’s history or a deal’s progression.
The Shift Towards Connected Systems
Barrows’ perspective shifted after seeing a demonstration of Otter.ai’s new integration tools. He explains that these tools offer a different paradigm by connecting meeting data directly into AI platforms like ChatGPT and Claude, thereby bypassing the need for manual copy-pasting.
This integration allows for more sophisticated and context-aware queries. Instead of simple requests like summarizing a single call, users can leverage AI to answer complex questions based on their entire history of interactions.
“So instead of saying: ‘Here’s one call, summarize it’ You can ask: • What have all my clients said about pricing this quarter? • Where are my deals getting stuck based on recent calls? • What did this client care about across every conversation we’ve had?”
As John Barrows notes, this allows AI to operate with a deeper understanding of the business’s reality, rather than making assumptions based on isolated data points. He emphasizes that AI’s improvement is contingent on the quality of its inputs and the richness of its context, not just advancements in the AI models themselves.
AI’s Evolving Landscape
Barrows articulates a vision for the future of AI in business, one that moves away from extensive manual prompting and towards more integrated, intelligent systems.
“AI doesn’t get better because the model improves. It gets better when the context improves. That’s where this is all going. Less prompting. More connected systems. Better inputs.”
He concludes by posing a question to his audience, inquiring about their current AI practices: whether they are still relying on copy-pasting methods or beginning to connect their systems for more effective AI utilization. This encourages a broader discussion on the practical adoption of AI in professional settings, underscoring his belief in the power of connected data for unlocking AI’s true potential.
📝 About This Content
This article is based on insights shared by John Barrows on LinkedIn.
📅 Originally posted on March 19, 2026 | View original post on LinkedIn →