About the client
Overgear is a global e-commerce marketplace for digital gaming services, connecting players with professional service providers across multiple games and international markets. PPC activity supports multiple game ecosystems, service lines and purchase journeys, requiring disciplined segmentation, rapid experiment cycles and close efficiency control.
NDA naming policy
Game and project names are anonymized as project-*; direction names are anonymized as service*. Operational campaign suffixes—including _all, _retention, _test and _control—remain visible because they identify the test group, not the NDA entity.
Metric definitions
Transactions is the standard term used for purchases and items purchased. First purchasers are users completing their first transaction. Total revenue standardizes revenue, purchase revenue and transaction revenue.
CPA is cost per Google Ads conversion; CPO is cost divided by transactions; CAC is cost divided by first purchasers; ROAS is total revenue divided by cost. Conversions and All conversions remain separate Google Ads reporting metrics.
All monetary values use a comma as the thousands separator, a period as the decimal separator and a trailing dollar sign. N/A means a metric cannot be calculated or does not apply; an em dash means no source value was provided.
Bid increase tests
Main goal: to increase share of voice and amount of clients.
- Bid increase — project-gammaHypothesis: Increase Search impression share, top-of-page visibility, CR and AOV while maintaining or reducing CPO. Traffic split: 70/30 for service3; 50/50 for other directions.
- Bid increase — project-betaHypothesis: Buy more impression share, improve CTR and increase conversions while lowering CAC. Traffic split: 50/50.
- Bid increase — project-delta service5Hypothesis: Raise Search impression share and top positions to increase traffic and purchases despite higher CPC. Traffic split: 50/50.
- Bid increase — project-beta service3 C1Hypothesis: Increase impression share and top positions, adding traffic and purchases despite CPC growth. Traffic split: 50/50.
RSA-related tests
Experiments with ads for best texts optimization.
- Pinned headlinesHypothesis: Pinned USP headlines improve CTR and CR, reducing CPA despite potential CPC growth. Traffic split: 50/50.
- Ad quality upliftHypothesis: Improve Top %, Abs. Top %, CTR and CR while reducing CPA through better ad relevance. Traffic split: 50/50.
- RSA simplification — project-gamma service7/service8Hypothesis: Reduce RSA headline count and tighten pins to lift relevance and CR without materially hurting position. Traffic split: 50/50.
- RSA simplification — project-gamma service3Hypothesis: Use fewer, more relevant headlines to improve query fit, CTR, CR, CPO and CAC. Traffic split: 50/50.
Other cases
- Exact match keyword expansionHypothesis: Exact Match improves CTR and CR through more precise intent, while Phrase Match maintains reach and lower CPC. Traffic split: None; Exact and Phrase ran together for four weeks.
- Low-bid split for high-volume keywordsHypothesis: Separate high-frequency terms into low-bid campaigns to cut CPC by 30–50% while preserving traffic for lower CPA. Traffic split: None; pre 08.01–12.02 vs post 13.02–20.03.
- Generic service semanticsHypothesis: Reach new non-game-specific audiences and generate additional purchases at acceptable ROAS, CPO and CAC. Traffic split: 50/50.
- Landing page test — project-gammaHypothesis: Replace the standard URL with a CTA-style URL to increase commercial relevance, CTR and purchases. Traffic split: 50/50.
- Retention split into separate campaignsHypothesis: Separate warm audiences from broad acquisition to lift CTR, CR trans, CR FP and ROAS. Traffic split: None; retention vs all-traffic baselines across 2 projects × 2 directions.