Quality and Call QA: Coaching
Objective. Open Coaching and read it as the step that closes the loop - distinguishing coaching assigned from coaching completed from behaviour actually improved.
Coaching is where findings become action. This screen answers 'did the quality finding turn into something that changes behaviour?' - the step that decides whether the whole loop was worth running.
See it in the demo
What this screen answers
Coaching tracks the actions created from quality findings - what was assigned, to whom, and whether it closed. It answers 'is the loop actually changing behaviour, or just generating scores?' A CSM uses it to confirm findings close into coaching and to tie coaching back to recurring-error patterns; Sales uses it to make the point that closing the loop is treated as part of quality, not an afterthought.
The task
Scope to the team and period and read coaching by status - assigned, in progress, completed - and by the finding or root-cause pattern each action traces to. Open one and follow it: the finding, the coaching action, its completion, and ideally the next-cycle score on the same error. The task is to verify the loop closes and to separate activity from outcome at each stage.
How to read it
The classic misread is treating coaching assigned, or even completed, as improvement. They are three different things: a coaching action created is not a coaching action done, and a coaching action done is not a behaviour changed. The value of the whole loop is realised only when the recurring error shrinks in the next cycle. Read completion as a necessary step, not a result, and always look for the follow-on evidence - did the same rubric item or root-cause pattern improve for that agent or team afterwards. A high coaching-completion rate with no movement in recurring errors means the loop is busy but not working.
Honest limits
This screen can show that coaching happened; proving it worked needs the recurring-error trend over the next cycle and a baseline. Attribution is never clean - other factors move quality too. Confirm with product how coaching workflows, assignment and completion tracking are configured and entitled; the demo shows the mechanism on seeded data.
Interpretation check
Coaching assigned, coaching completed, and behaviour improved are three different things - the loop only pays off when the recurring error shrinks in the next cycle, not when a coaching task is marked done.
In the field
OVO (contact centre): a coaching action on a recurring call-handling error only counts when the next sample shows that error declining for the coached agents - completion alone would be activity, not outcome.
Practical task
Open the Coaching screen, pick one completed action, trace it back to its finding and forward to any next-cycle signal, and write a three-line note separating assigned, completed and improved - then name the one metric you would baseline to prove coaching worked.
Self-check
Coaching completion is 95 percent. Has quality improved?
Unknown from completion alone. Completed coaching is a step, not a result; improvement is proven by the recurring error shrinking next cycle, not by a completion rate.
What makes coaching part of quality rather than an afterthought?
A finding that never closes into completed coaching changes no behaviour, so recurring errors persist. The loop's value is realised only when evaluation closes into coaching that moves the error.
What follow-on evidence should you look for after a coaching action?
The next-cycle score on the same rubric item or root-cause pattern for that agent or team - did the specific error actually decline.
Why is improvement attribution never clean?
Because other factors - staffing, volume, process change - also move quality. Coaching impact is a hypothesis supported by the trend and a baseline, not a clean causal claim.
Verified vs demo. Screen is live in the demo at /#/coaching; exact controls and figures are SME-validation-pending.