⚠️ Critical drop-off between 3-sec plays (2,521) and 25% watched (1,305) — 48% of initial viewers bail in the opening seconds. Another 57% bail between 25% and 50%. Move the hook, the pain point, and the CTA to the first 15–20 seconds of the video.
Three active ads ($131+ combined) are clicking but generating zero bookings. Check Meta Events Manager and the landing page booking flow. If the event isn't firing, pause those ads immediately and fix before restarting.
V2 is generating leads at $8.18 CPL vs V1's $24.28 — nearly 3× cheaper. They're the same format, different cut. V2 gets more budget now, or at minimum run a proper head-to-head with equal spend.
93% of viewers never reach the end of V1. The hook, pain point, and lead form prompt need to appear in the first quarter of the video. Even a rough re-edit that front-loads the offer will likely improve CPL significantly.
The algorithm needs ~50 conversions per week per adset to learn. With 30+ ads splitting impressions, nothing exits the learning phase. Pick the 3–5 strongest concepts, pause the rest, and give real budget to the winners.
The r/AusMechanics-style Reddit post is the cheapest traffic driver in the whole account at $4.21 CPM and $0.22/LPV — about half the cost of everything else. Push it from $5.77 to $30–50 to validate the signal before scaling further.
This was the highest-CTR traffic ad at 3.15% before it was paused. If it was paused without a good reason, reactivating it could immediately improve traffic efficiency vs the currently-running alternatives.