From DM to closed deal: a 5-step coach sales funnel (with conversion benchmarks)
The five steps every high-ticket coach funnel should have: capture, qualify, book, show, close. Median conversion at each stage: 35% qualified, 80% book, 75% show, 30% close. End-to-end: ~6 closes per 100 DMs at the $1,500–$3,000 offer level.
The five steps
- Capture — get the DM in your inbox
- Qualify — confirm they fit your offer
- Book — get them onto your calendar
- Show — get them on the call
- Close — convert to paid
Median benchmarks per step
These are aggregated from 500+ coaches running DMForge across fitness, business, coaching and course niches.
| Step | Median rate | Top quartile |
|---|---|---|
| Capture → Qualified | 35% | 50% |
| Qualified → Booked | 80% | 92% |
| Booked → Show | 75% | 88% |
| Show → Closed | 30% | 50% |
End-to-end: median ~6.3 closes per 100 raw DMs. Top quartile: ~20 closes per 100.
For a $1,500 offer, that's $9,450 (median) or $30,000 (top quartile) per 100 DMs.
Where coaches lose the most leads
The biggest leak is almost always Capture → Qualified (the AI script). If your number here is under 25%, your qualification questions are wrong (likely too long or badly ordered). Fix the script and the rest of the funnel lifts with it.
The second-biggest leak is Booked → Show. Fix this with:
- WhatsApp reminders 24h and 2h before
- A pre-call resource (a 5-min video) that the lead has to watch to "lock in" the call
- A confirmation message the lead has to reply to 12 hours before
The third leak is Show → Closed, but that's your sales skill on the call, not the funnel.
How to instrument it
In DMForge, the dashboard tracks all four conversion rates per agent. You'll see which step is leaking within a day of going live.
If you're on a tool that doesn't track this (most flow-builders don't), you can't fix what you can't see. Switch.
What "great" looks like
Top quartile coaches consistently hit:
- 50% qualified rate by writing tight, specific qualification questions
- 90% book rate by having the AI propose 3 specific slots, not "let me know what works"
- 88% show rate with double reminders + a pre-call homework asset
- 50% close rate by offering a clear single offer with a single price, not a menu
These are not theoretical numbers. They are what the top 25% of our user base achieves every month.
A real example
A fitness coach we work with runs ads to a quiz funnel that ends with a DM CTA. Numbers for last month:
- 1,840 raw DMs (most via comment-to-DM)
- 644 qualified (35%)
- 580 booked (90%)
- 481 showed (83%)
- 144 closed (30%)
- $1,500 offer → $216,000 monthly revenue
- Cost of running it: $39/mo for DMForge + $4,000/mo on ads = $4,039
- Net: $211,961
That coach used to do this with two human setters costing $4,000/month combined and was closing 1/3 of that volume.
Where to start
Don't try to optimize all five steps at once. Pick the worst one, fix it, ship, measure. Repeat.
For most coaches, the worst step is Capture → Qualified, which means: rewrite your qualification script. Use the 7-question framework. Run it through the DMForge live simulator. Tune until you've felt your script work in the role-play. Then connect Instagram.
Related: The 7 qualification questions, Comment-to-DM automation, The hidden cost of per-message pricing.
Build your own AI DM setter in 60 seconds.
Free forever. No credit card. No Meta approval.
Try DMForge →