How to coach SDRs on cold calling in 20 minutes a week

Coach SDRs on cold calling from scores, not from recordings. Pick the two skills the team scored lowest on this week, open one call per rep where that skill went wrong, write one note each, and send it before Monday's dials. The whole thing takes about twenty minutes for a team of five once the calls are scored on a single rubric. Coaching fails on most floors not because managers do not know what good sounds like, but because listening to whole calls does not fit in a week, so it stops happening by the second month.

The usual version of cold-call coaching is a manager with headphones on, a queue of recordings and good intentions. It works for three reps for three weeks. Then pipeline reviews, hiring and the quarter's number crowd it out, and the reps go back to learning from whoever sits nearest.

Why listening to whole calls does not scale

A connected cold call runs about a minute and a half on average and six or seven when it goes somewhere. Five reps, twenty connects a week each, is 100 calls. Listening to even a quarter of them is two hours before you have said a word to anyone.

Managers respond by sampling, and the sample is always the same: the calls the rep flagged, which are the good ones, and the calls that booked, which are also the good ones. The call where a rep collapsed on "we already have a solution" for the fourth time this week never gets picked, because nobody flags their own worst call.

Scores fix the selection problem. If every call is graded on the same six skills, the lowest score for each rep this week is a number you can sort by. You open that call, not the one the rep chose. Our post on building a cold call scorecard sets out the six skills and the weights. This post is about what to do with the scores once you have them.

The Monday format

The slot is twenty minutes, same time every week, before the first dials. Prepare from the scores on Friday afternoon or Monday morning. The output is one note per rep and one theme for the team. Nothing else.

The notes go out by message, not in a meeting. A rep who reads "at 1:42 she said the renewal was in six weeks and you kept pitching, next time stop and ask about the renewal" knows exactly what to do on Tuesday. The same thing said in a group session arrives as general advice about discovery.

The team theme does one more job. If four reps scored low on objection handling and the objection was "send me an email" in three of the four calls, the floor has a pattern, not five separate problems. One line on the board, such as "agree, then ask what the email has to answer", gets drilled by everyone that week. The post on handling "just send me an email" has the full exchange if you want to hand it out.

What to say, by skill

The note is the whole intervention, so it has to be specific. A note that says "work on discovery" changes nothing. These are the shapes that work for each of the six skills, with the sort of line that goes in the note.

Opener. The failure is almost always pace or a question that invites a no. "You said your name and the company in one breath at 0:03 and she asked you to repeat it. Slow the first line down. And drop 'is this a bad time', it got you a 'yes' twice this week."

Reason for call. The failure is a reason about you rather than about them. "Your reason at 0:14 was that you help logistics firms with compliance. Hers would have been that she took on twelve drivers in September. Lead with the thing that is true of her."

Discovery. The failure is a second pitch where a question belonged. "At 1:42 he told you the alerts were never set up. That was the call. You went back to the product for forty seconds. Next time, 'how long has that been the case?' and nothing else."

Objection handling. The failure is defending. "On 'we use Samsara' you said yours was better. Ask what they use it for. Never argue with the incumbent."

Value proposition. The failure is length and generality. "Your value line ran thirty-five seconds and covered four features. Cut it to the one that matches what she said at 1:42, in under fifteen seconds."

Close. The failure is a vague ask. "You ended on 'would it make sense to chat sometime'. Ask for fifteen minutes on Thursday at ten, then hold through the first soft no."

Each note has a timestamp, the prospect's words, and one line to try. That is the format. It takes two minutes per rep once you are reading from the right moment, which the score found for you.

The second call

A note changes behaviour when the rep tries the new line before the memory fades, and the worst place to try it is on the next real prospect, who may be the best account in the territory. The rep should run the same scenario again, with the same objection, within a day of reading the note.

That is the part of coaching most floors skip because there is nowhere to do it. Peer roleplay works but needs two people and a free half hour. An AI prospect built from the rep's own targets lets them run the exact call again, alone, in five minutes, and get a score on the one skill the note was about. On a floor doing this, the coaching loop is call, note, same call again, score, and the manager sees the second score without listening to anything. Our page for managers ramping and coaching SDRs shows what that looks like for a team, including the Monday digest that does the Friday sort for you.

What to track, and what not to

Track two things per rep per month: the lowest skill score and whether it moved. If a rep's discovery score goes from 4 to 6 over four weeks of notes about the second pitch, the coaching worked. If it stays at 4, the note is wrong or the rep is not running the second call, and both are worth a conversation.

Do not track the overall score week to week. It moves with the list, the prospect's mood and the time of day, and a rep who had a bad Tuesday looks like they are regressing. Do not track call count in the coaching slot either. Activity has its own meeting.

Two warnings from floors that have tried this. First, two skills a week, never six. A rep with six notes acts on none. Second, the note goes to the rep and only the rep. The theme goes to the board. Nobody's lowest score goes on the board, ever, or the scores stop being honest within a fortnight because reps start skipping the reviews on their bad calls.