Slim weekly face from your 2026-09-09 notes. Keep the deep dashboard; refresh this weekly. Sep = MTD through Sep 8 run-rate projected to month-end.
Sep gold bar is run-rate projected. One weak ad set paused Sep 8. Remaining-month pace may overstate leads if spend undershoots.
| Month | Media $ | L | CPL | S | CPS |
|---|---|---|---|---|---|
| Jul 2026 | $823 | 334 | $1.99 | 32 | $26 |
| Aug 2026 | $12,686 | 3,069 | $3.21 | 226 | $56 |
| Sep 2026 (projected) | $13,177 | 3,049 | $3.25 | 304 | $43 |
Media = Meta caregiver campaign + Google Search homepage. Jul/Aug S = HubSpot signup_date (ex yopmail/test); Sep S from Sep 1–7 week pin ×30/7. Repeat users: not defined yet. Gold outline = projected.
| Horizon | Leads | Signups | Media $ |
|---|---|---|---|
| EOY 2026 (Sep–Dec @ Sep RR) | 12,195 | 1,217 | $52,709 |
| EOY 2027 (12× Sep RR) | 36,585 | 3,651 | $158,127 |
If September’s pace held. Signup run-rate is the softest input (nurture paused; age and date-of-birth capture on the match path).
| Annual spend | $20 CPS | $25 | $35 | $40 |
|---|---|---|---|---|
| $1000k | 50,000 | 40,000 | 28,571 | 25,000 |
| $750k | 37,500 | 30,000 | 21,428 | 18,750 |
| $500k | 25,000 | 20,000 | 14,285 | 12,500 |
| $250k | 12,500 | 10,000 | 7,142 | 6,250 |
Highlighted = Gerard’s sketch anchor ($500k @ $25 → 20,000). Live media plan ~$240k/yr ($20k/mo). Live CPS lately ~$35–60.
Levers only. Ordered path is in section 5.
Not a forecast. How we move from today’s economics toward the sensitivity cell you circled ($500k at $25 CPS → 20,000 signups/year).
Leads are cheap. Signups still cost too much. Cost per signup is the number that moves us toward the $25 cell.
Repair age and date-of-birth capture on the match flow. Restart Instant Form nurture (the Meta lead-form email follow-up) after bounce cleanup. Ship one server-side signup event (our system tells Meta, Google, and Reddit when someone actually signs up) so platforms optimize to signups, not Instant Form leads.
Why first: spending more at $43–56 cost per signup does not get us to $25. Conversion work lowers that number without raising budget.
Pause creatives that bring cheap leads but few signups. Hold weak video variants. Put budget on proven static ads. Once the signup event exists, point every platform at that event.
Why second: mix shifts CPS even when total spend is flat. Do not raise daily budgets until Phase 1 is live.
Restart Reddit after bot-filter and click-ID attribution are proven (so we count real people, not bots, and can tie clicks to signups). Advance Google Exception / certification for cancer intent where policy allows (a Google review path that can unlock blocked health keywords). Launch the next creative wave when the signup path is stable.
Why last: adding channels multiplies whatever cost per signup you already have. Get that number down first.
Numbers: Aug CPS from closed media ÷ 226 real HubSpot signups; Sep CPS from projected face. Live media plan today is nearer $240k/year than the $500k sketch row.