Calls all upcomingheld 30dno-shows
Synced from GoHighLevel and verified against Fireflies. brief ✓ means the pre-call brief is built — it's inline on the profile. A booked call is never counted as held until Fireflies sees it happen.
One ranked list for the whole pipeline — the OS's recommended move for every open prospect, plus referral and review asks and data verdicts. Drafts wait in Gmail for your one-click send — nothing sends automatically. Done hides a row 7 days; snooze 3 or 7; skip removes it. Click a name for the profile.
Since you last looked
Fires when a quoted prospect replies, a call books, a payment lands, a no-show is detected, or a new /apply arrives. Last 48 hours shown; click a kind for the full history.
Open deals
Every open opportunity, one table. Win % blends stage (Fireflies-verified show→quote and quote→close), recency of real contact, engagement and source. At risk = money-stage deals past the safe touch window (quoted >7d, showed >14d, booked >10d). Held, not quoted names the specific blocker for each held call. Value in italics is imputed from the open-quote average. Click a column to sort, a row for the profile.
Closed all deals
Live from TaxDome via Zapier. Green = counts toward this year's goal (closed on or after the goal start).
Conversion Fireflies-verified
Held (grey) vs quoted (green) · weekly
Show→quote is the conversion the whole autopilot exists to lift: only about a third of held calls reach a quote. GHL auto-advances every form fill to "Discovery Call Held", so every rate here is built on evidence (Fireflies, calendar, survey), not the GHL stage.
Cohorts where each month's leads leaked
Everyone whose first lead arrived that month, measured on the same clock so young and old cohorts compare fairly. Dimmed = still maturing, not a real zero. Click a month for its people and their actual dates.
Delivery capacity Mark's side
| Month | Disc held | Opp-rpts | Med. days | Packs | Signed | $ |
|---|
Discovery calls held → opportunity-report calls held (median days between) → quote packs built → signings. A red opp-report count means a month of discovery calls produced none. TaxDome onboarding turnaround is not instrumented (no API).
Where clients come from How they heardForm source
How they heard is the /apply survey answer — the real channel. Form source is the GHL form or import that created the lead. UTM-level attribution lights up once the /apply redirect passes the query string through.
Referrals the compounding engine
| Referred | Status | Revenue | Referrer |
|---|
Referral is the #1 true source. The engine drafts the referral ask 60 days post-signature (one per client), Google-review asks post-close, and referrer thank-yous — all wait in Gmail. ✓ = referrer thanked. Drafted asks appear in The queue.
Campaigns
Click a campaign for its audience — the non-won rows are the retarget list. Spend, ROAS and per-video cost per application appear here once Google Ads access lands and the /apply UTM fix ships. The Marketing Analyst posts weekly scale / cut / retarget calls above the table.
What the gap demands — reverse funnel
Avg fee, show→quote and quote→close prefill from your real data; the other rates are editable estimates.
Outbound bench sending paused
Warm-up plan
| Week | Sends / mailbox / day | |
|---|
The bench is enriched, scored and sequenced. Nothing sends until dedicated cold domains exist and a 3-week warm-up has run — never from prosperlcpa.com. Suppressed addresses are honored permanently.
Send queue
One person per company (highest ICP), verified email, not suppressed, no company touched in 30 days — the exact order the send agent will work. Triggers are mined from real employment data (new role = fresh equity or comp jump); they're the personalization hook.
Sequences — per-segment templates
First touch + follow-ups per segment, all driving to /apply. Merge fields {{first_name}} {{company}} {{sender_name}} fill at send time; the platform appends the physical address and one-click unsubscribe.
Belt
Scored mechanically off your own Fireflies transcripts by sales-os/training/bx.py, not by self-assessment. Rank is your trailing-10-call average, so one good call doesn't promote you and one bad call doesn't demote you. Every point is a behavior you control on the call — whether they bought is deliberately not scored. Top 1% of consultative sellers live at BROWN.
This week
One behavior per week, from the 12-week study guide. You can hold two or three things in attention on a live call — a week where you tried five things is a week you measured nothing. The pass condition is a number, not a feeling.
Where the points go
Aggregated over your last 20 scored calls. The lowest row is where the next point is cheapest to win. Note the weights come from your own corpus, not from a sales book: exploring an emotional handoff is worth 20 because it is the only metric that separates your lost deals from your committed ones, and talk share is worth only 8 because in your data it runs the wrong way.
Scored calls
Click a row to open that prospect. "Feel" is emotional handoffs explored over handoffs offered — the openings a prospect gave you to go deeper, and how many you took. Recording failures (single-speaker transcripts, silent recordings) are excluded rather than scored as bad calls.
Manuals
The training manuals, served through this dashboard's own authenticated API rather than as public files, because they quote real client calls by name. Push updates with python bx.py docs after editing the markdown in sales-os/training/.