The brief
Category leadership needed more than reach.
The operator's ambition was straightforward: become a category leader in U.S. online sweepstakes gaming. That meant the acquisition system had to produce volume while steadily improving player economics and brand recall.
- Grow users: scale acquisition across eligible U.S. jurisdictions.
- Protect acquisition efficiency: keep cost per first-time purchaser under control while spend scaled.
- Build a repeatable creative system: create native and gameplay-first executions that could scale.
The system
Test, measure, refine, repeat.
We placed Meta at the centre of the acquisition strategy. Creative tests, audience split tests, lift studies and structured A/B experiments turned the account into a continuous learning system.
Audience experiments
Split audiences to isolate which segments produced valuable paying customers.
Creative iteration
Refine concepts and executions continuously instead of relying on one winner.
Incrementality
Use lift studies to separate true acquisition impact from attributed conversions.
Platform collaboration
Use advanced platform tools and measurement support to sharpen every decision.
The media system was built around incrementality, audience quality and cost per paying customer. At this spend level, even a 1% efficiency gain created meaningful acquisition upside.
Performance
Scale only counts when the economics hold.
From August 5 to September 3, 2026, the account invested $862,981 across five campaign lanes. Broad acquisition supplied the volume, registration retargeting delivered the strongest efficiency and creative testing remained deliberately constrained while new routes were validated.
- $862,981
- Total campaign spend
- 19,840
- First-time purchasers
- $43.50
- Blended cost per purchase
- 92,710
- Registrations
- 21.4%
- Registration to first purchase
| Campaign | Delivery | Spend | Registrations | First purchases | CPA | CTR |
|---|---|---|---|---|---|---|
| US_Web_CBO – Broad Scaling | Active | $388,341.47 | 45,433 | 9,200 | $42.21 | 2.37% |
| US_Web_CBO – Evergreen Prospecting | Active | $215,745.26 | 22,299 | 4,871 | $44.29 | 2.45% |
| US_App_CBO – Lookalike Scaling | Active | $129,447.83 | 11,520 | 2,718 | $47.63 | 1.86% |
| Retargeting – Website Visitors | Active | $86,296.11 | 7,858 | 2,432 | $35.48 | 4.20% |
| ABO_Web_Creative Angle Testing | Off | $43,150.33 | 5,600 | 619 | $69.71 | 1.26% |
Campaign performance: sales objective, first-time purchase, USD, America/New_York. CBO, highest volume, 7-day click / 1-day view, Advantage+ placements.
The brief, answered
The campaign was measured against the three priorities set at launch: player growth, acquisition efficiency and a creative system built for continued testing.
Objective 01: Grow users
- 92,710 registrations
- 19,840 first-time purchasers
The acquisition engine delivered meaningful player volume across eligible U.S. jurisdictions.
Objective 02: Protect acquisition efficiency
- $43.50 blended CPA
- $35.48 from registration retargeting
Acquisition economics remained controlled while monthly campaign spend moved beyond $860K.
Objective 03: Build a repeatable creative system
- 2 creative families
- 6 modular executions
Creator-led and gameplay-first routes created a repeatable framework for new hooks, games and offers.
Creative system
Two families. Six executions. One testing logic.
We paired native creator-led hooks with gameplay-first product proof. Human stories built attention and trust, while game-led executions made the experience tangible. Both closed through the same modular offer structure.
Family 01 · Native social: creator-led acquisition
Three ways to move from human attention into product value without losing the native feel of the feed.
- Lifestyle-led UGC: a conversational selfie opening earns attention before revealing the game and welcome offer.
- Direct-response creator: a concise direct-to-camera pitch simplifies the value proposition before a clean game-specific end card.
- UGC-to-gameplay proof: a personal recommendation transitions into real gameplay, then closes with a clear game-led call to action.
Family 02 · Product experience: gameplay-first discovery
Three executions that make the game itself the hook, using pace, reward progression, localisation and recognisable mechanics as proof.
- Reward momentum: immediate gameplay, escalating feature states and fast visual progression keep the product at the centre.
- Game-led concept: a distinct visual world carries the viewer from recognisable game mechanics into a direct play-now close.
- Reward escalation: a jackpot sequence combines localised language with rising reward states to sustain attention through the full loop.
Expansion
How we turned creative signals into scale.
The executions were starting points, not isolated ads. Each signal expanded into a matrix of hooks, creators, games, offers and audience states that could be tested without rebuilding the whole concept.
- Hook variants: change the opening line, face and visual pattern while keeping the game and offer constant.
- Creator profiles: test different creator archetypes and delivery styles against the same value proposition.
- Offer architecture: test value framing, introductory rewards and urgency as separate variables with compliant copy.
- Game-led concept families: give each flagship title its own visual system, hook bank and audience hypothesis.
- Geo-localised variants: adapt language, cultural cues and seasonal moments without rebuilding the core concept.
- Lifecycle retargeting: separate registered non-payers, first-time payers and lapsed players with stage-specific messages.
Days 1 to 7: Map the matrix
Concepts, audiences, formats, offers and measurement rules.
Days 8 to 21: Find repeatable signals
Kill weak routes quickly and deepen the strongest concept families.
Days 22 to 30: Scale with control
Increase spend only when player quality and unit economics remain intact.
At scale, creative learning compounds faster than media spend. The model behind scale was straightforward: precise measurement, structured experimentation and relentless creative iteration.
