Every recorded setter call from the Pilot-2 flight (Jun 25 to Jul 6), transcribed and scored against the current first-call script, automatically. Binary standard: each script concept either ran, in any wording, or it did not; no partial credit. Process sections only, product conversations are never evaluated. Scored retroactively, so gaps are expected; most of these calls predate the script being formalized. Rubric v2 · run 2026-07-22.
Script adherence · 18 discovery calls
Each concept of the first-call script, checked against every discovery conversation. A concept counts as run only if it demonstrably reached the prospect, in any wording; verbatim transcript evidence sits behind every bar.
Opens with who + whyfirm named, tied to their own request
17 of 18
"Not a sales call" framedefuses selling in the opening
11 of 18
Proactive acknowledgmentcredits them for taking action
0 of 18
Goal questionideal monthly income
15 of 18
Research historywhat they tried, what stopped them
8 of 18
Account inventorywhere the money sits, with numbers
17 of 18
Paycheck vs playcheckneeds and wants, with numbers
6 of 18
Gap named + agreed"that's a $3,000 hole, right?"
7 of 18
Pain questionwhat happens if you miss it
3 of 18
Scale questionsat least two of the five 1-to-10s
6 of 18
Next meeting on the calendara specific day and time, locked in
10 of 18
Expectations for that meetingvideo, spouse, duration, channel
14 of 14 discussed
Bar = share of the 18 discovery calls where the concept ran. No partial credit: a half-attempt that never landed counts as skipped. The last row counts only the 14 calls where a next meeting was discussed.
The three coaching flags this sample produces
1 · The emotional spine of the script is being skipped.The acknowledgment ran in 0 of 18 calls, paycheck vs playcheck in 6, and the pain question in 3. The facts get gathered; the feelings get skipped. The script's whole design is that the facts only matter after the prospect feels the gap.
2 · The gap gets named and agreed in only 7 of 18 calls, and when it lands, the call visibly turns.Every strong call in this batch pivots the moment a concrete monthly number is agreed to. Make the named gap the non-negotiable center of every discovery call.
3 · Only 10 of 18 calls end with a meeting you could write on a calendar.Yet when a next meeting is discussed at all, expectations get set 14 out of 14 times. The skill is there; the discipline of locking a specific day and time before hanging up is not. "I'll text you to figure it out" is where booked meetings go to die.
What good looks like (from this batch)
Best call of the flight: goal question, account inventory, a named-and-agreed gap, and a dated next step, all in one call.
"What would be your ideal retirement? What would that look like as far as income?"
"You have a pension about $1,000 a month... you have a 401K about $300,000."
"That means there's a $3,000 a month gap... Does that sound about right?"
"Friday? 1 p.m. central time? July 3, 1 p.m. You're going to receive a text message."
The rest of the flight's data
The sample at a glance
223recorded calls ingested
61reached a voice (50+ recorded words)
50conversations transcribed and scored
18discovery calls
28/50conversations where the rep held control
13/18discovery calls followed the script's order
Show risk follows booking distance
The same engine reads the calendar. Bookings happen minutes after the survey submit (median 4.6 minutes; 41 of 52 within 10 minutes), and from there, show rate tracks how far out the slot was booked:
Booked <24h outshow up on their own
5/7 · 71%
Booked 24-48h outthe trough
1/6 · 17%
Booked 48h+ outthe risk window
10/24 · 42%
Bar = show rate of due Pilot-2 appointments booked that far out, counting appointments with no pre-slot conversation. 46 of 52 due appointments carry a resolved verdict; the 6 unresolved are excluded and each is being chased as a data defect, not written off.
The 48-hour-plus window is a solvable nurture problem: tight booking windows so fewer slots land far out, plus a heavier pre-slot cadence on the bookings that do. Booking timestamps are verifiable in your GHL: contact created date vs appointment booked-on date.
Every conversation, scored for intent
All 103 real conversations across the setter lines, classified by how they ended and by where the lead was in their lifecycle at the moment of the call:
Before bookingn = 53 conversations
57% positive
Booked, before meeting 1n = 19 conversations
58% positive
After meeting 1n = 31 conversations
55% positive
positive (advanced or open)logistic (callbacks, screeners, scheduling admin)negative (resistant or terminal)
Lead quality holds at every step: a real conversation ends positive 55-58% of the time and negative at most 10%, wherever the lead is in the pipeline. The few objections that appear are the known three: wants someone local, distrust from past experiences, already has an advisor.