The challenge

Targon is INFINIA’s defense vertical: naval systems sold to government procurement agencies across NATO and MENA. When I joined in June 2025, the marketing function was effectively zero: no funnel, no measurement stack, no inbound channel. Sales cycles run 12–18 months, deal sizes are six to seven figures, and the buying committee includes procurement officers, naval engineers, defense attachés, and ministerial advisors. None of them respond to retargeting.

The hypothesis

If I treat the NATO + MENA naval-procurement universe as a finite, named list (it is) and run patient, intelligence-led ABM against the ~80 accounts that actually buy this category, I can produce qualified, sales-ready conversations at a CPL that performance benchmarks would say is impossible in defense, because the denominator is small enough to make the math work.

The approach

1. 01

   Defined the account universe. Built a named list of ~80 tier-1 naval procurement offices and primes across NATO and MENA, then mapped 4–6 buying-committee roles per account.
2. 02

   Built the asset library before the ads. Capability briefs, technical whitepapers, and short demonstration videos in EN and TR, gated behind a low-friction form (work email + role only).
3. 03

   Ran LinkedIn ABM as the primary channel. Account-based audiences with role-level targeting, refreshed weekly. No broad targeting, no lookalikes, just the 80 accounts.
4. 04

   Layered Google Ads on category-defining intent. $20K monthly budget on naval-systems and capability search terms, with manual bid strategy and aggressive negative-keyword hygiene.
5. 05

   Architected the measurement stack. GA4 + GTM + Looker Studio unifying ad, web, and HubSpot CRM, so every MQL is attributable to account, channel, and asset.

The results

Over two quarters, the programme produced 320+ MQLs at an average $42 CPL against accounts on the named list, a benchmark that, in defense B2G, behaves more like B2B SaaS economics than the category norm. 38% of those MQLs converted to SQLs in sales handoff, indicating the qualification logic upstream was working: we weren’t paying for traffic, we were paying for the right conversations with the right roles inside the right accounts.

In parallel, the $20K monthly Google Ads budget, applied with a strict test-and-learn cadence on bid strategy, landing-page variants, and audience refinement, produced a 22% reduction in CPA and a 15% lift in conversions in Q3 2025. Multilingual SEO with structured data and topical content hubs (EN/TR, across Targon, Marisonia, and XPZONE) drove non-branded organic sessions +215% YoY and page-one rankings for 45+ commercial keywords.

Cross-functional reach

Weekly coordination across engineers, designers, and sales teams in INFINIA’s Ankara, Istanbul, Amsterdam, Manchester, and Doha offices, translating technical product features into channel-specific marketing assets and lead-gen funnels without losing engineering accuracy.

Channels & stack

The challenge

Marisonia is INFINIA’s marine vertical: a subscription product for boat owners and charter customers across the Aegean and Mediterranean. The DTC paid-acquisition motion was scaling, but bumping against unit-economics ceilings. CAC was creeping up as the addressable consumer segment narrowed, and the product’s seasonal usage pattern (April to October peak) made retention harder than annualized subscription benchmarks. We needed a second acquisition channel that wasn’t paid social.

The hypothesis

Boatyards and charter operators are the natural distribution channel for marine subscriptions. They already have the trust relationship with boat owners, they handle adjacent services (storage, maintenance, charter management), and they cluster geographically in the same ports we already target with paid social. If I structure a B2B2C partnership with revenue share, we can offload most of the acquisition cost while protecting margin, and the DTC paid budget can be repositioned for brand and category education.

The approach

1. 01

   Built the partner economics model first. Worked through CAC arbitrage with finance: at what revenue-share rate did partner-acquired customers beat paid-acquired ones on contribution margin after factoring expected retention curves? Got to a defensible ratio before writing the first outreach email.
2. 02

   Identified ~50 target partner accounts. Filtered by partner type (boatyards, charter operators, sailing schools), geographic concentration in our existing paid hotspots, and adjacency fit with their existing service offering.
3. 03

   Designed co-branded landing pages with partner-specific UTMs. Each partner got a dedicated landing page with their branding and a unique tracking parameter, so partner contribution was attributable to within ~5%.
4. 04

   Repositioned DTC paid as the category-education channel. Reduced direct response intent, raised brand and education content. Goal shifted from “convert immediately” to “be the obvious choice when a boatyard mentions us.”
5. 05

   Migrated lifecycle email to handle partner cohorts separately. Different onboarding flow, different upsell timing aligned with partner relationships.

The results

12+ partners onboarded across Aegean and Mediterranean ports in two quarters. Blended Meta + Google ROAS held at 3.2× across €69.99 to €159.99 plans through the pivot, meaning DTC paid stayed efficient while the partner channel scaled in parallel. The acquisition mix shifted without breaking unit economics.

The challenge

One marketing function, three verticals: Targon (B2G defense), Marisonia (B2C marine), and XPZONE. Different audiences, different sales cycles, different stacks, different languages. Without infrastructure, the marketing role would collapse into reactive triage: produce a brief, get it translated, ship the campaign, scramble to assemble reporting, repeat. There would be no time for strategy.

The hypothesis

The marketing function’s leverage comes from infrastructure, not heroics. Build the production system first: the content workflow, the attribution stack, the dashboards. Once they exist, you can run three verticals with the same effort that ad-hoc work would require for one.

The approach

1. 01

   Built an AI-augmented content workflow. ChatGPT and Claude for brief drafting, EN/TR translation, and editorial QA, with strict brand-voice guardrails. Defense-sector copy needed human review at three points (technical accuracy, tone, compliance); consumer copy needed lighter editorial.
2. 02

   Codified brand voice per vertical as reusable prompts. Documented voice (defense formal, marine consumer warm, XPZONE specific) so AI outputs hit the right register before human editing. Reduced rework time per piece by roughly half.
3. 03

   Architected unified attribution across all three verticals. GA4 + GTM + Looker Studio + HubSpot CRM with consistent UTM governance, so the same parameter conventions worked for B2G LinkedIn ABM, B2C Meta retargeting, and internal campaigns.
4. 04

   Built one Looker Studio dashboard collapsing three reporting workflows. Previously: three separate spreadsheets, three weekly reports, three stakeholder conversations. After: one dashboard with vertical-level views and a cross-vertical summary, one weekly meeting cadence.
5. 05

   Codified the production process. Briefs, asset library, approval workflows, publishing checklist, written down once so it didn’t have to live in my head.

The results & the second-order effect

~40% content cycle compression (briefs that used to take ten days ship in six, with quality controls intact). 6+ hours/week saved on reporting, reinvested into channel testing and strategy. 3 verticals on one unified system that scales with marginal complexity. The infrastructure is what made the Targon ABM programme possible: without it, there was no time for 80-account campaigns.

The challenge

Indie strategy games live or die on wishlist velocity before Early Access. Storefront conversion, the Steam algorithm’s “trending wishlists” surface, and launch-day visibility all depend on it. Hakan’s War Manager entered my hands with fewer than 500 wishlists, a generic store page, no creator outreach, and no community. Early Access was 13 months out.

The hypothesis

The wishlist isn’t a vanity metric. It’s a leading indicator with a measurable ceiling tied to store-page CRO. If I treat the storefront like a landing page and pair that with creator-driven discovery in language markets the genre over-indexes in, the same traffic should produce multiples of the wishlists.

The approach

1. 01

   Storefront CRO as foundation. Iterated capsule art, “About this Game” copy, trailer hook, and screenshot order, A/B-tested against control periods to track visit-to-wishlist lift independent of traffic spikes.
2. 02

   40-creator program in strategy-game niches. Mid-tier YouTubers and Twitch streamers in grand-strategy and war-management, with custom early-access keys and tiered post-coverage attribution.
3. 03

   Steam Next Fest as the anchor event. Coordinated demo build, livestream schedule, and creator coverage around the festival, the single highest-leverage moment for wishlist velocity in the Steam calendar.
4. 04

   Organic short-form on TikTok and YouTube Shorts. Hooked on in-game moments and dev commentary. 1.2M+ organic views, with the strongest converters re-cut into store-page trailer variants.
5. 05

   Localization into Turkish, German, and Simplified Chinese. Storefront and trailer subtitling, prioritized by historical Steam audience data, contributed 31% of total wishlists from non-English regions.

The results

The campaign closed at 22,000+ wishlists pre-Early Access, a 44× lift. Store visit-to-wishlist lifted from 8.2% to 14.6%, meaning the same traffic produced 1.78× more wishlists per visit. Community-driven inquiries that converted to direct sales doubled. The 1.2M+ organic short-form views ran on a near-zero paid budget. The unit economics of indie GTM made organic creator content the only viable acquisition path, and it worked.

The challenge

Turquoise Choice operates PB360.me, pickleball events and league-management SaaS for North American clubs and tournament organizers. The sport was scaling fast; the platform’s marketing wasn’t. Paid was at zero, the SEO engine didn’t exist, the apps had no lifecycle email, and Play Store sat at 4.1. I joined as the first dedicated marketing hire and owned the full funnel for web plus iOS and Android.

The hypothesis

Pickleball’s player base was growing faster than category SEO supply could keep up with. If I shipped 50+ high-intent articles targeting queries tournament organizers were already searching, organic would compound faster than paid CAC could be optimized, and the budget should be structured around that asymmetry.

The approach

1. 01

   SEO content engine at 15+ articles/quarter. 50+ articles in 9 months targeting tournament-organizer and league-manager search intent, with internal linking and topical clustering.
2. 02

   Meta and Google Ads scaled $0 → $8K/month. Systematic creative testing on hooks, formats, and audiences, with retargeting flows that doubled weekly signups and pulled CAC from $14.20 to $11.05.
3. 03

   ASO across iOS and Google Play. Keyword optimization, screenshot iteration, and listing copy refinement drove a 41% organic install lift and moved Play Store rating from 4.1 to 4.6.
4. 04

   Klaviyo lifecycle infrastructure. Welcome, activation, and re-engagement flows that lifted 30-day retention by 27% and strengthened activation-to-paid.
5. 05

   Weekly A/B testing on landing pages and email creative. GA4 + Hotjar insights driving MAU growth of +15% across the 9-month engagement.

The results

Over nine months: organic visitors grew from 2,100 to 28,000 per month, five high-intent commercial keywords ranked on Google’s first page, paid CAC dropped 20%, weekly signups doubled via retargeting, and 30-day retention lifted 27%. Play Store rating moved 4.1 → 4.6, organic installs lifted 41%, and MAU grew 15%.