01Since we took over
May 19 to August 31, 2026, every day since takeover, compared with the same dates last year.
The headline
We took the account over on May 19. Every day since then through the end of August, ad spend went from $8,270 to $21,046 and attributed sales went from $48,997 to $110,316, up 125%. Bookings rose 90%, from 532 to 1013. ROAS moved from 5.9x to 5.2x, which is what happens when you scale spend into colder audiences, and that period still includes August, when ads were paused most of the month at your request.
How to read ROAS
ROAS is a ratio, not revenue. At low spend, ads reach only your warmest audience, so the ratio looks high while the dollars stay small. Scaling reaches colder audiences: the ratio comes down and the total goes up. Last year, $8,270 at 5.9x produced $48,997. This year, $21,046 at 5.2x produced $110,316. Average booking value rose too, from $92 to $109, up 18%. And the FareHarbor checkout means some bookings never reach Meta, so over any full season these figures are a floor, not a ceiling.
02This year vs last year, month by month
Attributed sales since takeover, each period against the exact same dates last year. August is low by design, ads were paused most of the month at your request.
03What Meta reports, and what actually landed
Meta only counts a booking when it can follow the guest all the way through the FareHarbor checkout. Your FareHarbor totals include everything: repeat guests, walk-ins, phone bookings and organic. Every month where both numbers exist is shown below, full calendar months on both sides.
| Month | Meta-attributed | FareHarbor actual | Difference | Meta share |
|---|---|---|---|---|
| May 2025 | $25,010 | $36,000 | +$10,990 | 69% |
| June 2025 | $15,921 | $52,000 | +$36,079 | 31% |
| July 2025 | $13,282 | $24,000 | +$10,718 | 55% |
| August 2025 | $13,907 | $19,700 | +$5,793 | 71% |
| Summer 2025 May to Aug | $68,120 | $131,700 | +$63,580 | 52% |
| October 2025 | $48,282 | $59,000 | +$10,718 | 82% |
| November 2025 | $47,926 | $55,900 | +$7,974 | 86% |
| December 2025 | $69,980 | $82,900 | +$12,920 | 84% |
| Holiday season 2025 Oct to Dec | $166,188 | $197,800 | +$31,612 | 84% |
| May 2026 | $50,959 | $40,000 | -$10,959 | 127% |
| June 2026 | $46,244 | $73,000 | +$26,756 | 63% |
| July 2026 | $32,502 | $35,000 | +$2,498 | 93% |
| August 2026 | $7,192 | $19,400 | +$12,208 | 37% |
| Summer 2026 May to Aug | $136,897 | $167,400 | +$30,503 | 82% |
| Combined, last two seasons 7 months | $303,085 | $365,200 | +$62,115 | 83% |
| Combined, all 11 months incl. summer 2025 | $371,205 | $496,900 | +$125,695 | 75% |
Across the last two seasons, Meta's figure came to 83% of what FareHarbor recorded, holding steady at 84% and 82%. Including summer 2025 it is 75%, because ads ran on $9,625 that summer against $24,281 this summer, so far less of the business came through Meta.
- Last summer ads ran on $9,625 against $24,281 this summer, so far less of the farm's revenue came through Meta and its share was lower.
- May 2026 includes 18 days before we took over the account on May 19. June 2026 includes an extra week of summer camp, roughly $10,000.
- FareHarbor figures for September 2025, January 2026 and September 2026 are not in yet, so those months are left out rather than estimated.
Why the two never match exactly
They measure different things. Meta records a sale on the day someone clicks or views an ad, and counts it even when that guest is a repeat visitor or a referral who might have booked anyway. It also misses people who saw an ad and later booked by phone or after a Google search. FareHarbor records every booking, on the day it is made. So Meta's figure is the platform's own estimate, and the FareHarbor total is what actually landed.
04The scorecard
| Month | Ad Spend | Bookings | Attributed Sales | ROAS | Same Period, 2025 | Change vs 2025 | What We Did |
|---|---|---|---|---|---|---|---|
| May 19-31, 2026 from takeover | $3,324 | 185 | $24,378 | 7.3 | $5,887 | +314% | Took over May 19. Full account audit and cleanup. Compared with the same 13 days last year. |
| June 2026 | $9,221 | 409 | $46,244 | 5.0 | $15,921 | +190% | Founder-led video ads launched. Audiences built. |
| July 2026 | $7,325 | 328 | $32,502 | 4.4 | $13,282 | +145% | Goat Yoga video test. Retargeting added. |
| August 2026 | $1,176 | 91 | $7,192 | 6.1 | $13,907 | -48% | Ads paused most of the month at client request. |
| Sept 1-28, 2026 partial | $5,343 | 330 | $25,497 | 4.8 | $26,428 | n/a, partial month | Event budget moved from Goatstravaganza to the core events. Expanded targeting into Houston. Built the Pumpkin Carving, Baby Goat Snuggles and Cow Cuddles campaigns. |
The 2025 column is the same period one year earlier. For May that is the same 13 days from the 19th, the day we took over. For June, July and August it is the full month. For September it is the full month of September 2025 set against 28 days of 2026, which is why no percentage is shown there. September is also the one period left out of the totals above.
05The season ahead
What October through January produced last season, all Meta-attributed. This is the window we are planning for now, and it opens with Pumpkin Carving with Goats on October 9.
The season opens October 9
Last season Oct to Jan delivered $226,761 on $26,401 of spend, an 8.6x return. It is the single biggest revenue window of the year, and this year it opens with Pumpkin Carving with Goats, Fridays to Sundays from October 9. Campaigns need to be live and out of the Meta learning phase before that first weekend, not during it.