RESEARCH
the table ranked them
backwards.
a conversion report pulled in the first weeks with a 10-person sdr team, then checked against what happened to those ten people. ranked by acceptance rate it put the one who got let go 4th of 10, above one of the two who went on to be the strongest.
you have this table.
meetings booked by rep, and how many an account executive accepted. it sits in every crm, it takes four minutes to pull, and people get promoted and managed out on the strength of it.
this one came from a B2B e-learning marketplace, pulled 25 August 2025, two weeks into an engagement that ran 8 August to 19 December 2025. it covers the quarter before anything changed. 1,003 meetings, 10 reps, one number each.
what makes it worth publishing is not the table. it is that i know what happened to all ten of those people afterwards.
64.0%
MEETINGS ACCEPTED BY AN AE (SAL)
642 of 1,003. 95% CI 61–66.9%.
43 pts
BETWEEN BEST AND WORST REP
47.3% to 90.3%, on the 8 reps with 30+ meetings.
2 of 10
REPS WITH TOO FEW MEETINGS TO RATE
volumes of 21, 9.
WHAT HAPPENED AFTER
four reps, four ways it was wrong.
- ranked 4 of 10 · 73.6%
- let go. he was known at the time for booking meetings that should not have been meetings, and this number puts him fourth of ten, above the rep who was promoted. if acceptance were a quality gate he would sit at the bottom of the list.
- ranked 5 and 2 of 10
- the two who went on to be the strongest reps on the team. both had started weeks before this was pulled, which is why one sits mid-table and the other has only 9 meetings behind an 88.9% that means almost nothing. they had the fewest habits to unlearn, so when the handoff was standardized they were the fastest to run it. every careful reading of this table sets the second one aside for lack of volume, including mine.
- ranked 1 of 10 · 90.3%
- top of the board, and her own sdr manager said of the account executive taking her meetings: "Aaron, I think would be a lot more lenient on her cells." the best figure on the table came with an asterisk from the person who managed her.
act on this table the day it was pulled and you remove the wrong person, overlook both future top performers, and promote from a figure the sdr manager already doubts.
the whole team, and how little the table knows.
each dot is a rep. the bar through it is the range the true rate could sit in. the hollow dots are the reps with too few meetings to say anything about, and the length of those bars is the point.
SCROLL SIDEWAYS TO READ
one thing this table is not: a measure of how anyone ended up. it was pulled to diagnose a motion, not to grade the people running it, and nothing in it tracks whether a rep got better afterwards. every one of them did. that is the part a conversion report cannot show you and the reason it should not be read as a verdict on a person.
the eye goes to the 43-point gap between best and worst and reads it as a talent ranking. it is not one. the hollow dots carry intervals running most of the width of the chart, which is the honest way of saying nobody knows where those reps sit, and one of them was weeks from being the best on the team.
WHY IT LIED
it was measuring the handoff, not the rep.
why do meeting-to-SAL rates vary so much between reps?
going through the team one at a time, every rep had invented their own way of handing a meeting over. some sent a calendar invite with no description. some wrote detailed notes to their account executive. some joined the call. some sent a follow-up email covering what was discussed and introduced the AE inside it. nothing was standard, and the reps doing all of it had the highest acceptance rates.
it shows up in a single pair of confirmation emails from that week. the rep at the top of the table wrote: "hey, great conversation. in regards to this, this, this and this, i'm gonna have my partner Aaron branch talk about this, this and this." she was reselling the meeting. two reps further down, asked what they sent, both said the same four words: "i sent a template."
so the spread is real and it is not talent. it is ten people each running a process they invented, reported as though it were a property of the person.
is a SAL a reliable measure of meeting quality?
no, and the team knew it. the sdr manager, on that same call: "if they're not meeting the criteria and they're getting saled, that is going to be a huge coaching issue." a meeting becomes a SAL because an account executive accepts it, and an account executive is a person with a relationship, a workload and a mood.
the rep with the reputation for over-eager qualification still had 73.6% of his meetings accepted. that is not a filter doing its job.
what happens after the meeting is booked?
the figure that reframed all of this did not come from this table at all. a separate report showed that roughly half the intro calls the sdrs booked, they were not attending. they set the meeting, sent the invite, and left the account executive to run it cold. the internal estimate at the time put about $175 of value on each one.
a booked meeting was being treated as a finished job, and the conversion table reports the damage as a rep's personal batting average.
what moved it.
one standardized qualification and handoff, and an end to sdrs and account executives running as separate operations. some pairs had no standing meeting at all to talk through territory, requested meetings, or how either of them was tracking. there was no shared picture of what good looked like, so ten people had built ten of them.
by the last day the same measure read 84% against 64.0% in this baseline. that figure is here on my word. it was read off a Salesforce report at the time and i do not have a copy, so treat it as testimony rather than evidence, which is why there is no chart of it on this page.
two caveats belong with it and neither is small. the SAL criteria were tightened inside the same window, so part of any movement is the definition changing rather than the work. and the roster changed. i cannot separate those from the process, and anyone claiming to separate three overlapping effects from one before-and-after reading is selling something.
the two who improved fastest were the two who had started most recently. they had the fewest habits to unlearn, and the standard handoff was the only one they had ever been given.
every row, including the ones nobody should act on.
all 10 reps, best first. ids are assigned in rate order and identify nobody.
SCROLL SIDEWAYS TO COMPARE
| rep | meetings | sals | rate | 95% ci | what happened next |
|---|---|---|---|---|---|
| R10 | 103 | 93 | 90.3% | 83.0–94.6% | AE called lenient |
| R09 | 9 | 8 | 88.9% | 56.5–98.0% | became a top performer |
| R08 | 73 | 56 | 76.7% | 65.8–84.9% | |
| R07 | 87 | 64 | 73.6% | 63.4–81.7% | let go |
| R06 | 42 | 26 | 61.9% | 46.8–75.0% | became a top performer |
| R05 | 186 | 114 | 61.3% | 54.1–68.0% | |
| R04 | 141 | 86 | 61.0% | 52.8–68.7% | |
| R03 | 195 | 116 | 59.5% | 52.5–66.1% | |
| R02 | 21 | 10 | 47.6% | 28.3–67.6% | |
| R01 | 146 | 69 | 47.3% | 39.3–55.3% |
ordered by rate, best first. the last column is filled only where what happened next is mine to publish, so a blank there means withheld rather than unknown. the 2 reps under 30 meetings are shown, because leaving them out is the mistake this page is about, and the width of their intervals says how little is known. rates are the share of booked meetings an account executive accepted.
how this was counted.
- what this is
- a baseline. the export was pulled 25 August 2025, about two weeks into an engagement that ran 8 August to 19 December 2025. it is the quarter before anything changed, which is the only reason it is worth reading: it is the table a manager actually had in front of them on day one.
- population
- the 10 people carrying an sdr seat, and every meeting they booked, 1,003 of them. 2 more people appear in the source export and are out because they were in other roles, which defines who this is about rather than selecting among performers. no rep was removed for their numbers and no meeting was dropped.
- the measure
- a requested meeting counts as converted when the account executive accepts it as a qualified lead. that is the client’s own definition, taken from their crm, not one imposed for this study. this page argues the definition is weaker than it looks, and still reports it exactly as they kept it.
- channels
- not the subject here, reported for completeness: web 80.2% on 253 meetings, outbound 54.7% on 201, and 549 more at 59.9%. that last group is a bucket i built for the analysis, holding 72-hour leads, personal web flows and the crm default value. it is not evidence their tracking was broken, and an earlier draft of this page said that it was.
- intervals
- 95% Wilson score intervals. the normal approximation was wrong here: it puts the top rep’s upper bound above 100%.
- where this is weakest
- the account executive is not a field in this export, so a rep’s rate carries an unknown amount of their receiving AE inside it. the top rep booked across several AEs, which makes a pure leniency story hard to tell about her, and her manager still called one of those AEs lenient. both are true, neither is measured, and one more pull with the AE attached would settle it.
- what this is not
- one team, one company, one quarter. anyone quoting it as a cross-industry benchmark is quoting something it never claimed to be. what it is: a measured account of a table i acted on, published with what the table got wrong.
- the people
- reps are anonymized because the export names them individually and that is theirs, not mine to publish. the company is anonymized because permission to name it was not sought; publishing did not need to wait on someone else’s review cycle.
what does your table say, and who does it disagree with?
- pull meeting-to-SAL by rep, then write down who you would rank differently
- read four confirmation emails, one per rep, and see whether they are the same email
- count how many booked meetings the person who booked them actually attends
30 MINUTES WITH PATRICK