Two competing views on AI

One view: AI will narrow performance gaps by giving lower performers templates and coaching. Other view: AI will widen gaps by automating routine work, freeing top performers for higher-value activities.

The research question

The real question is which effect dominates in real jobs where workers perform bundles of activities, not just single tasks. The answer matters for hiring, training, and team organization.

How the study was done

Researchers interviewed eight C-level revenue executives from the Bain Capital Ventures CRO Advisory Board, held more than a dozen advisor meetings with sales leaders, and surveyed two hundred thirty-six sales professionals at Stripe and Checkr.

The main finding

While AI broadly helps workers, the largest benefits are accruing to top performers, at least in leading-edge AI adopting firms. The gap is widening, not closing.

Time savings alone don't explain it

Higher-performing reps realized greater time savings than lower-performers. Managers assigned higher-value tasks to forty-four percent of top-quartile reps, compared with twelve percent of bottom-quartile reps.

Top performers use AI differently

Top performers are more fluent with AI tools and apply them across the deal lifecycle, not just to tactical questions.

Quality improves for top performers

Managers were sixteen percentage points more likely to report quality improvements for top-quartile reps than for bottom-quartile ones, suggesting gains did not come at the expense of customer experience.

AI literacy is the key

Each one-point step up in AI literacy was associated with a zero point seventy-one-point average increase in reported benefits.

Top performers are autodidacts

Top performers do not wait for training.They build literacy through trial, rapid iteration, comparisons with others, feedback, and peer knowledge sharing.Success factors include executive role modeling, encouragement to experiment, and concrete examples from power users.

Checkr's negotiation copilot

One rep at Checkr built a negotiation copilot before these tools became widely available. According to COO Lindsey Scrase, this kind of creative, account-specific use distinguished the best from the rest.

“this kind of creative, account-specific use was what distinguished the best from the rest”

Hire for learning ability

Companies should select for the ability to adapt, not just AI usage. Interview questions should probe whether candidates have built workflows, tested competing outputs, spotted errors, or decided not to use AI for sensitive situations.

Redesign entry-level roles

Brex CRO Garrett Marker automated twenty to forty percent of SDR manual work so ambitious reps could take on full-cycle deals independently. SDRs meeting targets while closing full-cycle deals were automatically promoted to account executive.

Build versus buy technology

Buy when the task is generic, the market solution is mature, and the tool does not require proprietary context. Build when the task requires sensitive data, product complexity, or domain-specific logic.

Centralize context, decentralize discovery

Companies should centralize data, permissions, knowledge bases, and guardrails. But they should decentralize discovery, letting top performers and power users find customer-specific uses that central teams would not identify quickly.

Productize the best discoveries

Leading companies centralize context, decentralize discovery, and diffuse or productize the best tools and answers. This means integrating, securing, measuring, documenting, training on, and retiring duplicates.

The managerial task ahead

AI is not simply replacing work or equalizing performance.Productivity gains depend on whether reps convert saved time into better customer understanding and deal strategy.Leaders must decide where AI should automate, augment, or remain human.