Launch / Campaign Setup
At launch, the only unknown should be concept performance. Product flow, offer, tracking, creative, and campaign configuration must already be verified.
1
Final structure
- Use separate Testing and Winning campaigns, both ABO.
- At initial launch, only Testing delivers.
- Each testing ad set contains one persona + angle + offer.
- Keep unproven concepts out of Winning.
2
Campaign QA
| Level | Confirm |
|---|---|
| Campaign | Correct account, app, objective, acquisition mode, ABO, attribution, naming, start time, and no campaign-level budget. |
| Ad set | One concept, broad US/CA/UK/AU targeting, exclusions, OS, event, equal budget, LTV floor, delivery bid, placements, and naming. |
| Ad | Correct identity, creative, copy, CTA, destination, UTMs, placement version, captions, safe zones, real UI, substantiated claims, and no drafts. |
3
Product-journey QA
Ad → App Store or landing page → install → signup → onboarding → paywall → trial or purchase.
- Verify the destination, screenshots, copy, price, trial, redirects, and mobile experience.
- Create a fresh account and complete every onboarding step.
- Verify the correct paywall, product, price, duration, terms, restore flow, and entitlement.
- Do not launch when the ad and paywall present different offers.
4
Tracking and billing validation
| System | Confirm |
|---|---|
| MMP | Install, source, campaign, ad set, and creative attribution. |
| PostHog | Signup, onboarding start/completion, and paywall view. |
| RevenueCat | Trial, purchase, product, price, currency, and entitlement. |
| Ad platform | Mapped events arrive once with the intended meaning. |
| Internal dashboard | Acquisition, product, and subscription data join correctly. |
- Use a new test user and fresh install.
- Confirm timing, identity, source, product, price, currency, and environment.
- Separate trials from purchases, sandbox from production, and refunds from positive revenue.
- Verify payment method, balances, access, 2FA, currency, time zone, and spending limits.
5
Launch sequence
- Build the complete Testing campaign before enabling delivery.
- Submit every ad for review and resolve the actual cause of rejection.
- Launch during monitored hours in the account’s time zone.
- Activate all intended testing concepts together.
- Confirm spend, delivery, links, countries, currency, and that no unintended campaign is active.
- Validate early clicks, installs, product events, subscriptions, and platform postbacks.
6
Pause rules
| Pause now | Do not panic over |
|---|---|
| Wrong destination, app, market, budget, event, offer, or price; broken signup/paywall/purchase; missing or duplicated events; privacy or policy issues. | Uneven ad delivery, early CPM or CPI volatility, slow starts, immature trials, or reporting latency. |
Fix technical failures immediately. Once the system works, protect the initial test from unnecessary audience, budget, bid, placement, concept, offer, or creative edits.
7
Record the launch
- Date, time, and account time zone.
- Platform, campaign, event, bid strategy, markets, and OS.
- Concepts, ads, budget per concept, total budget, and LTV assumption.
- Tracking test result and every post-launch change.
Takeaway
A clean launch removes avoidable uncertainty before money is spent.