KS

Kero Sports Financial Model

• Whitelabel SDK embedded into clients' (sportsbooks, teams, leagues, networks) existing apps.
• Proprietary ML models generate curated in-game bets on ~30–60 second intervals.
• Two modes: real-money sportsbook betting and free-to-play (FTP) leaderboard/points mode to reach casual fans.
• Social chat layer between bets to drive engagement and retention.
• Funnel logic: All Sports Fans → Those Watching → FTP Users → Sportsbook Clients - Kero claims to serve the entire funnel.

Not explicitly stated in deck. Inferred from context:

• B2B licensing / revenue-share with sportsbook, team, league, and media network clients - standard for whitelabel igaming SDK vendors. Rationale: product is described as a "whitelabel SDK deployed in the apps of our clients" with no direct-to-consumer monetization cited.
• Possible data/analytics upsell on first-party fan behavior data generated through the platform. Rationale: slide 5 references "first party data" as a conversion driver for sportsbook clients.
• FTP mode likely monetized via referral/affiliate fees to sportsbooks when FTP users convert to real-money accounts; consistent with industry norm for FTP-to-real-money funnel products.

Sample in-game numbers visible on slide 3 are illustrative UI mockups (e.g., "YES $27,790 / NO $19,270" bet pool sizes; "$9.47" win amount) - not traction figures.

• Moat claim: proprietary ML models for real-time in-game bet curation; first-party behavioral data flywheel.
• Differentiation framing: incumbents (sportsbooks) are positioned as failing to engage the next generation; Kero positions itself as the engagement layer they lack.
• Client-side pain points addressed: sportsbooks need differentiation beyond price; teams/leagues/networks need fan monetization and engagement strategies.

Team:

• Tomash Devenishek - Founder & CEO; 15+ years senior tech exec; previously built Coachella app gamification engine; bootstrapped Kero for 2 years; formerly founded a blockchain P2P sports betting exchange.
• Tom Gray - VP of Revenue; 10+ years sports & media sales; joined from OpenBet; formerly Senior Manager of Sports Partnerships at Sportradar.
• Rustin Domingos - Chief Data Scientist; UC Berkeley EPS PhD Candidate; MIT Physics PhD Candidate; 2 peer-reviewed publications; built predictive algorithms for NBA, MLB, NHL.

• Archetype + why: B2B SaaS / revenue-share SDK model. Kero's economics are driven by enterprise client contracts (sportsbooks, leagues, networks) - not direct consumer volumes - making a contract-count × revenue-per-client ARR model the right spine. A secondary FTP-to-real-money conversion affiliate layer can bolt on as a separate revenue line.

• Forecast horizon & granularity: 3 years monthly (Year 1–2 monthly detail; Year 3 quarterly roll-up). Pre-revenue / pre-launch stage warrants monthly granularity to track client onboarding pipeline and burn clearly.

• Key drivers & assumptions:

*Client acquisition & contract*

• Number of signed B2B clients (sportsbooks, teams/leagues, media networks) by quarter - zero disclosed; start from 0 with ramp assumption.
• Average annual contract value (ACV) per client type - sportsbooks likely higher ($200K–$500K ACV range) vs. teams/media ($50K–$150K); no deck data.
• Revenue-share % of GGR (gross gaming revenue) generated through SDK - industry range 10–25% of incremental GGR; alternative to flat ACV.
• Sales cycle length ~6–12 months for regulated gaming operators; affects revenue recognition timing.
• Client churn rate ~10–15% annually; typical early-stage B2B SaaS.

*FTP / affiliate layer*

• FTP monthly active users (MAU) - not in deck; modeled as a function of client installs × engagement rate.
• FTP-to-real-money conversion rate ~5–10% industry benchmark.
• Affiliate CPA per converted bettor ~$100–$200 per acquired depositing player; standard igaming affiliate rate.

*Cost structure*

• Headcount: engineering (ML/data science heavy), sales, client success - 3 named team members currently; model headcount ramp as clients sign.
• Cloud / infrastructure costs for real-time ML inference at 30–60 second intervals - significant; scale with active users per game event.
• R&D / model training costs.
• No COGS for physical inventory (pure software).

• Scenarios (Base / Bull / Bear - which variables flex):
• Bear: Slow client onboarding (1–2 signed in Year 1); lower ACV; FTP conversion below 5%.
• Base: 3–5 clients in Year 1 growing to 10–15 by Year 3; ACV mid-range; FTP layer contributing ~20% of revenue by Year 3.
• Bull: Major sportsbook anchor client signed early (DraftKings / FanDuel tier); rapid distribution via league deals; FTP flywheel accelerates conversions.

• Required sheets / outputs:

1. Assumptions dashboard (all drivers clearly flagged, easily toggled per scenario)
2. Client pipeline & ARR build (client count × ACV, monthly new / churned / expansion ARR)
3. FTP/affiliate revenue model (MAU funnel → conversion → CPA revenue)
4. P&L (Revenue, Gross Profit, OpEx by category, EBITDA, Net Income)
5. Headcount plan (by function, with salary + benefits)
6. Cash flow & runway (monthly burn, cash balance, implied runway to next raise)
7. Summary KPIs: ARR, clients, burn rate, months of runway, LTV/CAC (once data available)