Why can acquisition costs differ so much between a mini-program and an app?

A mini-program is not inherently cheaper to promote. Costs are comparable only for the same channel, user outcome, and observation period. A mini-program can open from a QR code or ecosystem entry while an app usually adds a store download, but fewer steps do not make traffic free. Wavesteam puts media, incentive, channel operation, creative, landing, installation or authorization, retention, and revenue into one funnel rather than quoting an unsupported “three to ten times” ratio.

Stage	Mini-program	App	Common outcome
Reach	Private channels, QR, content, ads; still costs media and operation	Store search, web, ads, referral, offline	Genuine exposure or click by a qualified target
Product entry	Fast open; may require permission/login and face scene limits	Store page, download, install, first open	First completed core task, not “open”
Re-engagement	User return and platform-permitted messaging	Push, deep link, update; can be denied or uninstalled	Authorized effective return
Retention and value	Smooth for light tasks, often used once	Higher install friction, potentially more value for frequent/device work	Same cohort retention, margin, or LTV
Attribution	Shares, QR, and in-platform paths cross contexts	Store, ad, and web definitions differ	Deduplicated user and explicit window

When comparing platform capabilities, constraints, and switching costs, also compare How much does custom software cost to operate, and is it much more expensive than SaaS? and How different are the app and mini-program versions of the same product?; the linked guidance adds context that should be considered in the same decision.

Sharing is not free: incentives, content, community work, support, and fraud loss count. App acquisition is not only paid ads; search, brand, web, and existing-user referral can be organic. Channel composition follows brand, industry, user, and region, not the container.

Define one success event for both, such as a first appointment completed without refund after the agreed window. Use the same deduplication, attribution, included media/incentive/labor, and cohort start. Track exposure, click, entry, core task, return, payment or margin, and refund. Keep unassignable acquisition separate rather than force allocation.

Apple's App Store Connect Acquisition and Google Play's store performance guidance expose source and conversion under their own definitions. Their counts cannot be divided directly by WeChat analytics without aligning denominators.

Use randomized or at least comparable audiences, channels, and creatives, controlling discounts. Report uncertainty when samples are small. CAC divides attributable investment by qualified new users, and payback uses contribution margin rather than gross sales. Low short-term CAC with next-day abandonment may be inferior to a costlier install that produces sustained value.

One-time lookup, in-store scanning, and rare booking often suit mini-program or mobile web. Frequent tools, offline, reliable push, hardware, and complex creation may justify an app. Many businesses let a mini-program complete the first service without forcing an app. Ask users to migrate only when the app has clear additional value and repeated demand.

Wavesteam unifies campaign, source, user, and core event across clients. The dashboard shows channel CAC, completion, retention, margin, refund, support, and incentive cost. Marketing owns media and creative; engineering owns tracking, deep link or QR, event quality, and privacy controls. Run a bounded comparison and scale only after a pre-agreed unit-economic gate. If a second client cannot demonstrate value, Wavesteam does not recommend building it solely for promotion.