Listening to everything isn't a plan, it's an excuse not to choose
A team of 8 reps making 25 calls a week each is already 200 conversations a week. No manager has 30 free hours to listen to all of them, so most managers listen to none — and when something goes wrong, they find out from the quarter's result, not from the call where the deal actually broke.
«I'm going to listen to more calls» is the New Year's resolution of sales coaching: it gets said at every kickoff, lasts two weeks, and dies the moment normal operations resume. The problem isn't discipline. It's that nobody designed a criterion for deciding which of the 200 to listen to.
Why random sampling doesn't work either
The intuitive fix is «I'll pick 5 at random each week». It feels fair — everyone has the same odds of being picked — but it's statistically useless: with 200 calls and 5 samples, the odds of landing on the one with a mishandled objection or an economic buyer who never showed up are close to zero.
Random optimizes for «feeling fair», not for «finding what matters». Those are different goals, and the second one is the one that actually moves the quarter's number.
The three filters for choosing what to listen to
Instead of random, three criteria — in this order, because each one filters on top of what the previous one left:
Large deals that moved stage without a clear reason, or that have gone more than two weeks without any activity from the buyer. This is where money gets lost without anyone noticing in time.
A new rep's first call with a new prospect, or the first time a recently changed playbook gets used. This is where the habit is still forming — correcting it here is worth ten times more than correcting it six months later.
The rep who's billing above average and the one who's billing below, in the same account segment. Comparing their calls is the fastest way to find what the winner is doing differently — without guessing.
Applying these three filters in order narrows the week's 200 calls down to 8 to 12 real candidates. That's the number a manager can actually listen to without it eating the rest of their week.
What to look for in each call you pick
Listening without a checklist is the same trap as random sampling, just after you've already spent the time. Four fixed questions per call, always the same:
- Did the rep confirm the pain in the buyer's own words, or assume it from the pitch?
- Did the call end with a next step that has a date, or was the close «I'll email you next week»?
- Was the economic buyer identified, or is the rep still only talking to their first contact?
- Was there a real objection the rep let slide without answering?
Feedback that's worth giving is feedback that can be applied on this week's next call — not a generic note about «good energy» that changes nothing.
The weekly cycle: pick, listen, give feedback, repeat
The framework only works if it becomes a calendar habit, not an exercise done «whenever there's time»:
- Monday: apply the three filters to last week and build the list of 8 to 12 calls.
- Tuesday to Thursday: listen to 3 to 4 a day with the four-question checklist, never all in one sitting.
- Friday: a 10-minute 1:1 per rep you listened to — one specific thing to keep, one to change.
A month of this cycle generates more real signal about the team than an entire quarter of forecast reports.
Why this shouldn't be done by hand either
Applying the three filters by hand means cross-referencing the CRM, the calendar, and the memory of who said what — every Monday, for a team that keeps growing. It scales badly, and it's exactly the kind of repetitive work that gets the weekly cycle abandoned the first time things get busy.
It's what Performy runs automatically on 100% of calls, not just the 8 or 12 a manager manages to pick by hand: it applies the same risk, novelty and divergence filters every week, and has the priority list ready — with the four questions already answered for each one — before the manager opens their calendar on Monday.