-
1
Define the four rubric dimensions in behavioral language, not adjectives
Before any review happens, the rubric has to score behavior you can see on a recording, not vibes. The four dimensions that matter are discovery depth, question quality, objection handling, and close ask. For each dimension, write 1 to 5 anchors in behavioral language so a 3 means the same thing to every manager and every rep. Discovery depth at a 4 might read as 'rep asked at least one quantified business-impact question and tied the answer back to a stakeholder named earlier in the call'. Question quality at a 2 might read as 'rep stacked three questions in a row without waiting for the buyer to finish'. Behavioral anchors kill the arguments about what a score means and make scoring portable across managers.
- Write 1 to 5 behavioral anchors for each of the four dimensions in a shared doc the whole sales org can see.
- Pressure-test the anchors by scoring the same recording with another manager; disagreements over one point mean the anchor language is still too loose.
- Pin example clips next to each anchor so new managers can calibrate against real moments, not written theory.
- Version the rubric with a date so iterations are visible and reps know which version their current scores reference.
Tip: Resist the urge to add a fifth or sixth dimension in month one. Four dimensions scored well beat eight dimensions scored sloppily, and the rubric gets heavier every time you add a column.
-
2
Separate the review cadence from ride-alongs and shadow sessions
A call recording review is not a ride-along and not a shadow, and conflating the three is the fastest way to make the cadence feel pointless to reps. A ride-along is live, where the manager is on the call and may jump in. A shadow is silent, where the manager observes without scoring to learn the rep's style. The recording review is async-sourced and sync-discussed: both parties watched the two chosen calls before the meeting, both bring scores, and the 30 minutes is spent on the gaps between the two scorecards and on the two or three timestamped moments that matter most. Reps who understand which surface they are in behave differently in each one, which is the point.
- Document the three surfaces in a one-pager so reps can tell at a glance which type of session is on their calendar this week.
- Color-code the three surfaces on the manager's calendar so reps see the pattern across the quarter, not just this week.
- Keep ride-along feedback in a separate note from recording review scores so the rubric trend is not polluted by live-call context the recording never captured.
- Reserve shadow sessions for the first two weeks of ramp and after any major territory move; do not run them every week by default.
Tip: If reps are asking at the start of each meeting what kind of session this is, the three surfaces are not distinct enough yet. Rename them and re-publish the one-pager before you blame the reps.
-
3
Pick two calls per rep per week, not a full call library
A 30-minute review of two calls beats a 30-minute scan across ten. Each week the rep picks one call they want feedback on and the manager picks one call the rep did not flag, usually from a stage transition that matters (discovery to demo, demo to proposal, or a stalled late-stage deal). Two calls is enough to see pattern, few enough to review deeply, and balanced enough that the rep never feels the manager is cherry-picking only the worst moments. Use the conversation intelligence tool's filters to find the manager-pick quickly: filter by stage, duration, and deal size, then skim transcripts rather than listening end to end.
- Ask the rep to drop their one chosen call link in the shared review log by end of day Thursday for a Friday review.
- Pick the manager's choice using filters that force variety across weeks (different stages, different buyer titles, different deal sizes).
- Pre-queue the two or three timestamped moments on each call that the review will anchor on so the live 30 minutes stays on discussion.
- Avoid using the same call for two consecutive weeks; even a great teaching call gets stale and the rep learns to tune out.
Tip: If the rep's pick is always their best call of the week, say so out loud and ask for the hardest one instead. Reps who flag only safe calls are telling you the cadence feels like performance management, not coaching.
-
4
Score both calls on the four-dimension rubric before the meeting
The 30-minute meeting is for conversation, not scoring. Both the manager and the rep score each of the two calls on all four dimensions independently, in the shared review log, before the meeting opens. The meeting then starts with a side-by-side look at the two scorecards and spends most of its time on the dimensions where the manager and rep diverge by two or more points. Divergence is the signal: it means either the rubric anchor is unclear, the rep is not seeing the moment the manager is seeing, or the manager is pattern-matching off a prior call that is not on the screen. Working the divergence is where the real learning happens.
- Lock the scoring window to the 24 hours before the meeting so neither party rushes it ten minutes before.
- Require a one-sentence written justification per dimension, not just a number, so the score is defensible.
- Flag dimensions with a two-point or greater divergence as the first agenda item for the live meeting.
- Use Strkr AI to pre-summarize divergence patterns across the trailing month so systemic rubric drift surfaces before the review, not after a quarter.
Tip: If manager and rep scores agree perfectly every week, the rubric is too lenient or one party is anchoring on the other's score. Rotate who scores first each week to break the anchor.
-
5
Anchor every meeting on two or three timestamped moments, not full calls
A 30-minute review that tries to walk two calls end to end is a 30-minute review that teaches nothing. Pick two or three timestamped moments per call before the meeting, cue them up with a shared clip link, and spend the live time on what the rep could have done differently at exactly that beat. Timestamped moments force specificity: 'at minute 14, when the CFO asked about renewal pressure, you answered with a feature instead of asking a quantified follow-up' lands and is remembered. 'Work on your discovery' does not. Reps retain feedback tied to their own words and their own buyers in a way they never retain abstract advice, so the whole cadence is built on this specificity.
- Share the clip links in the review log 24 hours before the meeting so both parties watch them with context.
- For each clip, name the rubric dimension it illustrates so the moment ties back to the scorecard, not just an opinion.
- Alternate between buyer-moment clips and rep-moment clips so the rep practices reading the room and refining their own delivery.
- Cap the meeting at three clips total; if there are more moments worth discussing, log them for next week rather than compressing all of them into 30 minutes.
Tip: If the rep is listening to a clip for the first time during the meeting, the prep broke down. Pause, reschedule, and come back next week with pre-watched clips rather than running a half-prepared review.
-
6
Close with one written behavior commitment for the next five calls
Verbal commitments evaporate between Friday and Monday. Written commitments logged in the same place both parties see every day do not. End every 30-minute review with one specific, testable behavior the rep will apply on their next five calls, phrased as a one-sentence change: 'ask one quantified business-impact question before any product talk' or 'confirm the economic buyer by name before proposing the next step'. Log that commitment in the shared review record and open the following week's review by replaying one call where the commitment should have shown up. The commitment-replay loop is what turns a conversation into a system.
- Phrase the commitment as a behavior change, not a skill label; 'better objection handling' is not testable, 'name the objection out loud before responding' is.
- Log it in the CRM or shared review template, not a manager-only tool, so the rep sees it when they open their deal list.
- Open the next week's meeting by playing one clip where the commitment should have landed and score it together on just that dimension.
- Track commitment follow-through as a separate metric from rubric lift so execution gaps and skill gaps show up as distinct signals.
Tip: If a rep misses the same commitment three weeks in a row, pause and ask whether the commitment is wrong for the deals they are actually running. More often the behavior is miscalibrated to the segment than the rep is coasting.
-
7
Rescore the rubric monthly and pair the trend with pipeline math
Weekly scores are noisy, monthly trends are not. Every 30 days, rescore each rep on the same three to five calls per rep methodology used in step four and plot the trend across the four rubric dimensions. Pair each dimension with the pipeline metric it should move: discovery depth with discovery-to-demo conversion, question quality with meeting-to-next-step rate, objection handling with late-stage stall rate, close ask with proposal-to-close velocity. Expect lag; rubric scores lift in weeks and pipeline math shifts in months. If scores climb for a full quarter and the paired pipeline metric does not, the rubric is scoring the wrong behavior for the segment and the cadence needs a redesign, not more coaching intensity.
- Rescore every rep on the same calendar day each month so the trend is comparable across the team.
- Chart each rep's four-dimension trend separately rather than averaging into one score; averages hide the dimension that is actually moving.
- Pair each dimension with a single named pipeline metric and review the paired trend in the quarterly business review.
- Segment the trend by tenure so new-hire lift does not mask tenured-rep plateau or the other way around.
Tip: If a rep's scores plateau for 60 days on a dimension, pause the current commitment and ask whether the behavior is wrong for the buyer type they are running. Most 60-day plateaus are segment mismatches, not effort problems.
-
8
Protect the cadence through the busy quarters, or lose it
Call recording review is the first ritual that gets cancelled when the quarter heats up, and the one that is hardest to restart once it slips. Treat the 30-minute slot the same way a forecast call is treated: it does not move for a buyer meeting under 60 minutes, it does not get shortened to 15, and it does not become a pipeline review in disguise. If forecast or firefights crowd the slot three weeks in a row, the problem is calendar design, not time management. Separate the review cadence from pipeline review into two different meetings with two different agenda templates so the review surface is never where forecast questions go to live.
- Label the calendar hold as 'Call review' not '1:1' so the protected intent is visible to everyone who sees the invite.
- Keep pipeline review as a separate weekly or biweekly meeting with a different agenda template and a different note location.
- Require 24-hour notice to reschedule the review slot, the same bar used for forecast calls, so the cadence is defended structurally.
- If the review has slipped three weeks running for one rep, pause the manager's own next skip-level and use that time to restore the cadence; it is the single highest-leverage move in the quarter.
Tip: A review that slipped three weeks almost never fully resumes without an explicit reset. Have the reset conversation on the fourth missed week, not the sixth, before the cadence quietly dies.
