Checklist: Geo-Targeted Views

The 5 signs you're scaling the wrong market, the metric that tells you before you burn the budget, and the scale-or-cut decision rule

THESIS: Most "our growth stalled" problems are actually "we scaled the wrong geo" problems wearing a disguise. A cheap-install market feels like a win at the top of the funnel — low CPI, rising volume, a dashboard that looks like it's working — right up until you notice the users never activate, never pay, and never come back. The trap is that geo quality is invisible at the metric everyone optimizes (CPI) and only becomes visible three stages down (activation, purchase, retention). By then you've scaled spend into a market that structurally can't return it, and blended CAC is quietly rotting while the volume chart points up. This checklist makes geo quality visible early: the five signs you're in the wrong market, why each one leads the CAC damage, the fix (organic traffic from creators who actually live there), and the scale-or-cut rule that ends the argument.

CONTEXT: The core distortion is that CPI measures the wrong thing. CPI ranges from under $1 to $26+ by category and geo, and the cheapest geos are cheap because the audience is low-intent, low-ARPU, or hard to retain — the market prices that in. So a low CPI isn't a discount, it's often a warning. The only honest read is downstream: install → activation → purchase → retention → blended CAC vs LTV. And the fix isn't "buy better traffic in that geo" — it's earning qualified traffic from inside the market: local creators, local device networks, real localization. A native local creator's content converts a market that translated global creative can't touch — the difference between reaching an audience and merely reaching a geo. Score each sign ✅ healthy / ⚠️ watch / ❌ wrong-geo; the decision rule is at the end.

═══

SIGN 1 — CPI is low, but activation is weak

• What to look for: installs reaching the core product moment (the aha — first search, first scan, first level, first log) at a rate comparable to your core markets.
• Red flags: cheap installs that never activate; a big gap between install count and activated-user count in the geo; users opening once and never hitting the core action.
• Why it leads the damage: activation is the first honest signal of intent, and it's invisible at CPI. A market that installs cheap and activates weak isn't a bargain — it's tourists clicking a free thing. The fix: the message that drove the install has to match the product's actual value in that market; native-local content recruits users who install for the product, not for the novelty of the ad.

═══

SIGN 2 — Purchase / subscription rates collapse after install

• What to look for: purchase, subscription, or paid-conversion rates in the geo within range of your core markets (adjusted for local pricing).
• Red flags: affordable traffic, near-zero monetization; installs that never see a paywall conversion; a geo that looks great on volume and empty on revenue.
• Why it leads the damage: low CPI markets are frequently low willingness-to-pay markets — the price you saved on the install is exactly the intent you didn't buy. The fix: match monetization and price to local ARPU (a Tier-1 price wall in a Tier-2 market converts nobody), and recruit through creators whose audience actually pays for this category — intent transfers from the creator, not the CPI.

═══

SIGN 3 — Retention drops faster than in your core markets

• What to look for: D1/D7/D30 retention curves in the geo tracking your core-market shape.
• Red flags: users install, try it, and disappear faster than anywhere else; a retention curve that falls off a cliff after D1; the geo dragging blended retention down as it scales.
• Why it leads the damage: retention is the truest signal of product-market fit in a geo, and it compounds — a market that doesn't retain will never return its CAC no matter how cheap the install. The fix: retention problems are usually fit problems — the product's core use case may not match local behavior. Native creators surface which use case the market actually wants (the India lesson: the same product, framed for local behavior, retains where the global framing bounced).

═══