Skip to main content

What AI recommendations are based on

Recommendations depend on the quality of your notes and records.

S
Written by Sumit Kapoor
Updated over 3 months ago

Introduction

AI recommendations in Lynk are only as good as the information they’re built on. The system does not rely on assumptions or generic models alone—it learns from the data you actively record during your coaching workflow.

Clear, consistent inputs lead to clearer, more useful recommendations.

Core Inputs That Power Recommendations

Lynk AI combines multiple data sources to understand coaching context and learner progress.


Coach Notes

Coach notes provide qualitative insight into:

  • Performance observations

  • Challenges or constraints

  • Behavioural and focus-related signals

These notes help the AI understand why outcomes look the way they do.


Attendance Records

Attendance data reveals:

  • Participation consistency

  • Engagement patterns over time

  • Gaps that may affect progress

Irregular attendance often changes how recommendations are framed.


Progress Reports

Progress reports add structured evaluations through:

  • Holistic skill indicators

  • Technical skill assessments

  • Coach-written summaries

These act as formal checkpoints for AI interpretation.


Session and Batch History

Session history helps AI:

  • Identify long-term trends

  • Compare recent progress with earlier stages

  • Understand pacing and continuity

The more complete the history, the better the context.


Why Input Quality Matters

  • Clear notes lead to sharper suggestions

  • Consistent attendance marking improves pattern detection

  • Thoughtful reports strengthen insights

Incomplete or vague records limit recommendation accuracy.


Summary

AI recommendations in Lynk are built from coach notes, attendance, progress reports, and session history. The quality and consistency of these inputs directly determine how useful and relevant AI guidance will be—keeping the coach firmly at the center of the process.

Did this answer your question?