BRIEFS BUILT ON HOOKS AND ANGLES
Each brief named the angles to test and gave creators several opening lines to film, so the first three seconds always had options.
FIELD REPORT 08 · UGC AND INFLUENCER · MATCH GROUP
The creator
content engine
I scaled UGC and influencer campaigns across Match Group's female app portfolio to 10x the spend, at a lower cost per result and a higher ROI. Each creator shoot was cut into hooks, bodies and CTAs, so one shoot made many ads.
10xthe spend, at a lower
cost per result
cost per result
01 THE BRIEF
Scale on paid social runs out of creative before it runs out of budget.
At Match Group I owned app and web acquisition across Google, Meta and Apple Search Ads, roughly $2,000,000 a month in spend. UGC and influencer creative ran as paid campaigns across the female app portfolio.
Creator video fits the feed. It is shot on a phone, in a real person's voice, and it looks like everything else people scroll past. The catch is that any single ad tires as frequency climbs.
So scaling was a supply problem as much as a media one: a steady flow of new variants, and a fair, fast way to pick winners.
COMPANYMatch Group
(S&P 500)
(S&P 500)
ROLESenior Digital
Marketing Manager
Marketing Manager
SPEND~$2,000,000
a month
a month
WHENNov 2019 to
Mar 2024
Mar 2024
The bet: if every shoot produced a kit of parts, the test budget would find winners faster than fatigue could wear them down.
02 THE MECHANISM
One shoot, cut into parts, becomes dozens of ads.
Creators filmed hooks, bodies and CTAs as separate takes. Pick one of each to assemble an ad. The counter multiplies the columns.
● BUILD A VARIANTONE SHOOT · EXAMPLE LINES
@creator.handlePaid partnership
0:00 · HOOK HOOK RATE
POV: your best friend writes your dating profile.
0:03 · BODY HOLD RATE
Screen recording: she builds her profile, start to finish.
0:15 · CTA COST PER RESULT
Set up your profile tonight.
INSTALL4 HOOKS × 3 BODIES × 2 CTAS = 24 ADS FROM ONE SHOOT
ON SCREEN: H1 · B1 · C1, AD 1 OF 24
One more hook adds 6 ads. One more body adds 8. One more CTA adds 12.
Example lines written for this page, not the original ads. Module counts are illustrative. Each part was read on its own number: the hook on hook rate, the body on hold rate, the CTA on cost per result.
03 HOW IT RAN
Every variant got the same small test. Winners earned the budget.
NATIVE FILMING, MODULAR CUTS
Creators shot on their own phones, in their own voice. Every shoot was cut into hooks, bodies and CTAs, and the edit recombined them into variants.
ADS FROM THE CREATOR'S HANDLE
The strongest cuts ran as creator handle ads, partnership style, so the ad came from a person's account and read like the feed around it.
SMALL TESTS, THEN GRADUATION
Every variant got a small fixed test budget. Hook rate and hold rate read the creative. Cost per result and ROI decided what graduated into the scaling campaigns.
THE LADDER, AS I RUN IT TODAY
- 01 PREFLIGHTAn AI check scores the copy before launch.
Every script and caption gets a compliance score before a variant spends anything.
- 02 TESTEach new creative gets a fixed test allowance.
The same small budget for every variant, so the read is fair.
- CUTSpend passes k times the target cost with no result at target.
It stops there. Favorites get the same rule as everything else.
- 03 GRADUATEn results at or under target.
The variant moves into the scaling campaigns.
- 04 SCALEBudget follows the winners.
Fatigue watch starts the day a winner lands.
k and n are written down before launch.
THEN · MATCH GROUP
Small fixed budgets per variant. Hook rate and hold rate read the creative. Cost per result and ROI decided what graduated.
NOW
The same ladder, written down and run by rule, with an AI compliance check on the copy before anything spends.
04 FATIGUE
Every winner wears out. The refresh is part of the plan.
Cost per result for one creative drifts up as the same people see it again. The program stays flat because a new variant lands before the old one crosses the line.
● CREATIVE FATIGUEILLUSTRATIVE · SHAPE, NOT DATA
- ONE CREATIVE, NO REFRESH
- THE PROGRAM, REFRESHED
- REFRESH LINE
WATCHHook rate and frequencyHook rate slips first, before cost per result moves.
REFRESHBefore the lineA new variant goes live while the old one still performs.
SUPPLYThe next batch is filmingShoots are booked against the curve, not after it breaks.
Illustrative. The axis is an index, launch = 100. The shape is the pattern paid social creative follows as frequency builds.
05 THE RESULT
10x the spend, and each result cost less.
More budget usually buys worse results. Here cost per result came in lower and ROI came in higher at 10x the spend.
10xspend on UGC and influencer campaigns
Lowercost per result (CPA) at that spend
HigherROI at that spend
START1x
SCALED10x
COST PER RESULT ↓ LOWERROI ↑ HIGHER
Figures are from my own reporting at the time. The 10x is spend on UGC and influencer campaigns across the female app portfolio. Cost per result and ROI stayed with Match Group, so this page shows their direction only.
06 WHAT CARRIES OVER
If your brand lives on creator content, brief the parts and let the numbers pick.
01
Film parts. A new hook on an existing body is a new test with no new shoot.
02
Give every variant the same small allowance and a written cut rule, so favorites stop eating budget.
03
Run winners from the creator's handle. The ad reads like the feed it sits in.
04
Book the next shoot against the fatigue curve, before the current winners wear out.
SAME PLAYBOOK, LATERAt Newton Crypto in 2024 I built affiliate, KOL, UGC and influencer programs from scratch alongside paid.