The modern sales landscape is often characterized by a “broken” coaching model where managers, overwhelmed by data and deals, fall back on reactive, subjective feedback based on a tiny sample of calls.
In a recent webinar How to Build Personalized Coaching Paths for Every Rep featuring leaders from Docebo, Glyphic, and Enso Connect, several innovative frameworks were introduced to solve the scalability crisis in sales enablement.
In this article, I overview four concepts for building a high-performing sales team in the age of AI.
1 The Nine-Box Framework for Managing “Brain Fry”
One of the most common mistakes sales leaders make is “shoving the entire elephant” down their team’s throats by introducing too many changes at once. To combat this, Mark Kosoglow (CRO at Docebo) introduced a nine-box framework designed to measure a team’s “available capacity of change”.
The framework plots initiatives on two axes:
Competency: Moving from Aware (passed a quiz) to Competent (internal role-play) to Mastery (proven in a real-world call).
Cognitive Load: The amount of effort a rep must exert to change their existing behavior, categorized as Low, Medium, or High.
| Cognitive Load / Competency |
Awareness (1-3 pts) |
Competence (2-6 pts) |
Mastery (3-9 pts) |
|---|---|---|---|
| Low | 1 Point | 2 Points | 3 Points |
| Medium | 2 Points | 4 Points | 6 Points |
| High | 3 Points | 6 Points | 9 Points |
Under this system, a team is given a 17-point quarterly budget. A “High Cognitive Load Mastery” project is worth 9 points, meaning a team can only handle one such major behavioral shift per quarter. This forces leadership to make trade-offs rather than overwhelming reps with “brain fry”, where employee satisfaction plummets.
2 Moving from Call Samples to Aggregate Data Sets
Traditional coaching relies on managers manually reviewing a handful of “cherry-picked” calls — typically four or five a week — which provides a distorted view of performance. AI shifts this paradigm by providing an aggregate data set.
When every single call is automatically scored against a rubric, several things change:
Accountability: Reps behave differently when they know every interaction is being reviewed.
Action over Collection: Managers stop spending hours collecting data and instead spend those hours actioning.
Personalization: Leaders can see team’s strengths and weaknesses, allowing them to tailor coaching to specific individual needs rather than generalities.
3 Forecasting via Binary Risk Assessment
Personalized coaching is described as the most critical driver of forecast accuracy. Many organizations struggle with “haircuts”, where each layer of management arbitrarily reduces a forecast because they don’t trust the underlying details.
To fix this, the speakers suggested creating a binary forecasting system. Instead of using vague terms like “best case” or “most likely”, AI-driven scorecards can determine risk based on objective, binary yes/no criteria.
By decoupling deal maturity (the sales stage) from risk assessment, one organization reported increasing their forecasting accuracy from 54% to 97%.
Sales teams often assume that deals in later stages are likely to close. In reality, stage ≠ certainty. Evaluating actual risk signals separately improves forecast accuracy.
A deal can be in the late stage and still be high risk because a) the decision-maker hasn’t been involved, or b) a competitor is actively being considered, or c) budget is not fully confirmed.
By separating deal stage (where it is in the process) from risk signals (how healthy it actually is), teams can make more realistic forecasts and avoid last-minute surprises.
4 Rewarding the “Exposure of Risk”
AI tools now allow for proactive risk identification by tracking “competitive mentions” on calls and instantly alerting managers via Slack. This data is only useful if the organization fosters a culture where exposing risk is rewarded rather than punished.
The speakers emphasized that “losing alone is a fireable offense”. The goal of AI-supported deal reviews is to ensure that the minute a rep feels a deal is in danger, they flag it so the rest of the team can step in and work to save it.
AI provides the “receipts” to show reps that specific behaviors (like quantifying a pain point) can increase win rates significantly, turning the scorecard from a “Big Brother” surveillance tool into a “self-service” empowerment platform.
Conclusion
A recurring theme across these concepts is the need to move beyond individual call reviews toward continuous, contextual understanding of rep behavior.
Check out the Projects section to learn more about the technology behind the Sales Signals app:
This is the core idea behind Sales Signals — a system turning conversation data into actionable signals that connect behavior, coaching, and deal outcomes over time.
See Also
Here are some of my other posts related to sales strategy, coaching, and professional development:
| Title | Reading Time | |
|---|---|---|
|
Using Transit Time to Rethink Hotel Search | 14 min |
|
The 7-Step KPI Blueprint from Business Intelligence Analytics Perspective | 14 min |