UA team and data stack by spend level — what to staff and what not to build
Anas Maqsood's view of subscription-app UA teams — ExpressVPN's ~$8–9M/month app setup run by 3–4 performance people plus creative, BI and CRM pods; a 3-person team at $50–200K/month; and what not to build under $50K/month (MTA, MMM, data warehouse, CDP, in-house paywall tool).
UA team and data stack by spend level
TL;DR
- At ~$8–9M/month (ExpressVPN apps): only 3–4 performance marketers across Google, Meta, TikTok and Apple Ads, heavily AI/agent-assisted. Alongside: 2 creative strategists + 2 designers, 1 BI/data person, plus separate product and CRM/lifecycle teams.12
- At ~$50–200K/month revenue: three people. (1) Performance + analytics + creative briefs, (2) motion designer for videos/graphics, (3) growth manager for onboarding, paywall, CRM and ASO.2
- Very early: the founder can do everything. Cal AI’s CEO handled finance, product and performance in its early days, as Anas saw while working with them.3
- Under ~$50K/month, don’t build: multi-touch attribution, media-mix modelling, your own data warehouse or CDP, or an in-house paywall/subscription tool.4
- Once past that: CAPI setup, your own LTV and churn-prediction models, backend quality signals (via a subscription platform or Snowflake/Google Cloud). Then keep working the funnel (install→trial, trial→paid, trial length, freemium) and the ATT prompt.45
Details
ExpressVPN structure (Anas, Sep 2026). Desktop was separate. On apps, the performance side was 3–4 people. Anas managed campaigns, and “AI agents” did much of the routine work. Creative production sat in its own small team. Product (PM, QA, iOS and Android engineers) was separate. A CLM/CRM team owned onboarding, paywall, trial-to-paid, winback, remarketing and email.12
Why “don’t build it yet”. Anas has seen several VPN apps try to rebuild a paywall/subscription platform in-house. Not impossible, but too early for an app making ~$50K/month. It makes sense at $5–10M/month.4
Stack minimum. An MMP if you run paid social (see SKAN, AEM and CAPI — how app conversions reach Meta after ATT), a subscription/paywall platform, and a clean optimisation event (see Which event should Meta optimise on? Qualified trials and enriched events).
Disagreements / caveats
- Nebojša Radović describes a similar three-pillar team (media buying, creative, analytics) in gaming at much larger budgets (How mobile UA changed (2015 → 2026) — the remaining levers, team structure and channel mix by budget). Both point to small, senior, AI-assisted buying teams.
- MMM/incrementality: Anas’s “not under $50K” agrees with the “last-click is good enough when small” view (When last-click attribution is good enough (and when to invest in incrementality)).
- One practitioner’s experience. Spend thresholds are approximate and mix revenue and ad spend loosely as he spoke.
See also
- Planning Meta creative volume, budget and team — creative-team sizing on Meta
- Anas Maqsood
Footnotes
-
Why Your Meta Campaigns Attract Users Who Never Pay | Anas Maqsood @ 28:10 — Anas Maqsood ↩ ↩2
-
Why Your Meta Campaigns Attract Users Who Never Pay | Anas Maqsood @ 29:45 — Anas Maqsood ↩ ↩2 ↩3
-
Why Your Meta Campaigns Attract Users Who Never Pay | Anas Maqsood @ 31:18 — Anas Maqsood ↩
-
Why Your Meta Campaigns Attract Users Who Never Pay | Anas Maqsood @ 34:30 — Anas Maqsood ↩ ↩2 ↩3
-
Why Your Meta Campaigns Attract Users Who Never Pay | Anas Maqsood @ 36:07 — Anas Maqsood ↩