A rep takes a call on Monday afternoon. It doesn't go well — they fumbled the discovery, let the prospect control the conversation, and closed without getting a clear next step.
The manager reviews the recording on Wednesday. They pull out the key moments, add some timestamp comments, and write up a coaching note.
The rep reads it on Thursday morning.
By then, the rep has taken eight more calls. The moment is gone.
The Timing Problem in Sales Coaching
Feedback is only as useful as the behaviour it's connected to. That connection — the gap between action and correction — determines whether feedback produces learning or just information.
This is not a new idea. It's one of the most replicated findings in behavioural science. In study after study, across domains from surgery to mathematics to language acquisition, the speed of feedback is one of the primary determinants of skill development. Not the quality of the feedback. Not the expertise of the coach. The timing.
The mechanism is straightforward: when feedback arrives close to the behaviour, the learner can connect the correction to the specific decision they made, the specific words they chose, the specific moment where things went differently than intended. When it arrives two days later, that connection is weak. The rep knows intellectually that they should have handled the discovery differently. They don't have access to what it felt like to make the wrong choice in that moment.
And because the feeling isn't there, the correction doesn't update the behaviour. It updates the rep's beliefs about what to do — and then the next time the situation arises, the old pattern fires anyway.
Why the Standard Coaching Cycle Is Broken by Design
The typical sales coaching workflow — record call, manager reviews, feedback in 1:1 — is broken not because managers aren't trying, but because the structure makes timely feedback nearly impossible.
A manager with ten direct reports can't review every call. They review a sample — the ones flagged by the recording platform, the ones from deals the manager is already paying attention to, the ones that happened to come up in a 1:1 conversation. The rep who needs feedback on Monday's call may not get it until Friday. Or the following week. Or the week after a deal closes and a retrospective happens.
Meanwhile, the rep is accumulating repetitions. Every call they take in the gap between the mistake and the correction is a repetition of the uncorrected pattern. The behaviour is being reinforced while the feedback is in transit.
The Compounding Effect of Delayed Correction
In learning science, this is called the distributed practice effect — and it cuts both ways.
When correct behaviours are repeated consistently, they become more automatic over time. The rep doesn't have to think about it. The right move fires before the conscious mind gets involved.
The same is true of incorrect behaviours.
A rep who has been handling discovery calls with a specific bad habit — leading with features, not asking about consequences, failing to get explicit buy-in before moving on — has been practising that habit on every call they've taken. By the time the manager flags it, the behaviour may be well into the process of becoming automatic. The correction isn't just teaching the right approach; it's also trying to dislodge an already-forming pattern.
The later the feedback arrives, the more repetitions of the wrong behaviour have accumulated. The more entrenched the pattern. The more practice is required to overwrite it.
The Difference Between Knowing and Not Repeating
There's a moment every sales manager recognises: the rep who receives feedback, understands it completely, agrees with it entirely — and then does the same thing on the next call.
This isn't defiance. It isn't laziness. It's the natural result of a feedback loop that arrived too late to do what feedback is supposed to do.
Feedback that arrives after the behaviour has been reinforced can change what a rep believes about the right approach. It can't, by itself, change the automatic response that fires when the situation actually occurs. That automatic response was built through repetition. It can only be changed through repetition — specifically, repetitions of the correct behaviour, ideally with immediate feedback, until the new pattern becomes more fluent than the old one.
This is why reps can "know" the right answer and still not do it. Knowledge and automatic behaviour are different things, stored differently, updated by different mechanisms. Feedback that arrives on Thursday about a call that happened on Monday is highly effective at updating knowledge. It is much less effective at updating behaviour.
What Immediate Feedback Actually Requires
The ideal feedback loop — in Ericsson's model of deliberate practice, in learning science more broadly — arrives close enough to the behaviour that the learner can feel the connection between the correction and the specific decision being corrected.
In music, this means a teacher who stops you mid-phrase and shows you exactly where the timing slipped. In sport, it means a coach who corrects the movement pattern before you've done it wrong twenty more times. In surgery simulation, it means a system that flags the technique error before the trainee moves to the next step.
In sales training, it means feedback that arrives before the conversation is forgotten — ideally, before the rep has taken another call.
This is not possible in the standard coaching workflow. A manager cannot review every call in real time. They cannot attend every customer conversation. The structure makes immediate feedback impossible, so delayed feedback becomes the default — and delayed feedback, for all its value, is fighting against the timeline of how skill actually develops.
The Practical Implication
The solution isn't to make managers faster at reviewing calls. The bottleneck isn't the manager's speed — it's the structure of a system where feedback depends entirely on a human reviewing a recording hours or days after the fact.
What changes the feedback loop is moving practice out of the live environment, where feedback is structurally delayed, and into a setting where correction can happen in close to real time — while the behaviour is still happening, or immediately after.
The rep who practises an objection scenario and receives a score and specific feedback before they've moved on to the next thing is getting something qualitatively different from the rep who gets a coaching note two days later. The feedback is attached to the moment. The moment is still fresh. The correction has somewhere to land.
Speed of feedback isn't a nice-to-have. In how skill actually develops, it's the mechanism.
Decibl scores every practice session immediately — so feedback arrives while the conversation is still in the room, not two days after it's been forgotten.