The most engaging creative in the account produced nothing

Six creatives ran in this account. Creative 1 produced 64 leads on $1,238.01 at $19.34 each, the lowest hook rate of any creative that produced a lead at 5.50%, and a 25.88% hold rate. Creative 2 produced 24 leads on $742.50 at $30.94 each. Creative 3 produced 2 leads on $327.96 at $163.98 each, and had the highest hook rate in the account at 24.72%.

Creatives 4, 5 and 6 produced no leads at all, and between them they spent $556.06: $322.58 on Creative 4, $20.16 on Creative 5 and $213.32 on Creative 6. Creative 4 held the highest hold rate in the entire account, 50.26%, and still produced nothing. That $556.06 should have gone to the two creatives already shown to be working, Creative 1 and Creative 2, days into the flight rather than after it ended.

Grouped by what they produced rather than by any engagement number, two creatives brought in 88 of the account's 90 leads on $1,980.51, at $22.51 per lead. The other four brought in 2 leads on $884.02, at $442.01 per lead. Hold rate and hook rate ran the other way entirely: the group that produced almost nothing held 32.68% of its viewers and hooked 11.51% of them, both higher than the group that did almost all the work, at 24.73% and 7.75%.

Ranked by hold rate or hook rate, the wrong group wins. Ranked by leads, the right one does. The hold rate gap between the two groups is not adjusted for video length, and it does not need to be to make the point: the group that held and hooked better is the group that produced almost nothing, and in this account the only ranking worth trusting is the one that counts what a creative actually produced.

90 leads became 66 qualified leads, a 73% qualification rate the clinic reported from its own records. Those 66 qualified leads became 21 new clients, a 32% close rate on qualified leads, also the clinic's own figure. New client count is derived as leads times qualification rate times close rate, then rounded, rather than a raw headcount from the CRM.

Cost per new client came out at $136.25. Two creatives out of six did essentially all of this funnel's work: the other four are what turned a $22.51 group cost per lead into a $31.83 account-wide average.

Pair this account with chiropractic clinic D, elsewhere in this set. There, hold rate correlated with cost per lead at -0.665, a real and useful signal: the creative that held attention longer bought cheaper leads. Here, the creative that held attention longest, at 50.26%, bought no leads at all. Both findings come from unmodified platform data in the same portfolio.

The honest conclusion is not that hold rate is useless or that it is reliable. It is that engagement metrics predict conversion in some accounts and mislead in others, and an account has to establish which one it is before building any rule on top of them. That is exactly why hold rate and hook rate cannot be the thing a New Zealand clinic optimises towards without first checking what they actually correlate with in that clinic's own data.