Moving an existing subscription app onto a new platform
Switching subscription infrastructure touches billing, entitlements, analytics, and paywalls at once, so the real question is not whether a migration is possible but what gets carried over and what you still own afterward.
This page walks through the migration program as described, the deliverables it bundles, and the checks worth running before you sign up.
What the migration program bundles
The program describes three deliverables that arrive together. First, the program describes a discounted first-year price as offsetting migration cost. Second, the program describes white-glove migration support in which the vendor's team helps move subscribers from your current setup. Third, the program describes a paywall rebuild with tailored experiments aimed at improving conversion after the switch.
Those three pieces map to different owners inside your company. Pricing is a finance decision, customer migration is an engineering and data decision, and paywall experiments are a product and marketing decision. Treating them as one package is convenient, but each still needs its own acceptance criteria before you commit. The exact discount terms, the scope of migration support, and the number of paywall rebuilds are program claims that should be confirmed in writing before you sign up.
A pre-migration checklist
Before scheduling a migration call, confirm the following on your side:
- A current inventory of products, subscription groups, and promotional offers across every store you sell through.
- A list of active entitlements and how each maps to a customer-visible feature.
- The identity model you use today (anonymous IDs, account IDs, or a mix) and how it will translate.
- A named owner for the customer migration step and a named owner for the paywall rebuild.
- A rollback plan if the first paywall experiment underperforms.
If any of those items is missing, the guided migration has less to work with. The program describes moving customers on your behalf, but only you can confirm which entitlements those customers are supposed to keep.
One SDK across stores, and paywalls without app updates
The pitch is a single SDK that covers StoreKit, Google Play Billing, and other tooling, kept current so you do not run maintenance sprints each time a store changes its APIs. On top of that, paywalls are built from templates or from scratch, published remotely, and rendered in SwiftUI and Jetpack Compose.
The practical implication is that paywall copy, layout, and pricing display can change without shipping a new app binary. That is a real operational shift: it moves paywall iteration from an engineering release cycle to a product workflow. It also means your release process needs a review step for remote paywall changes, because a mistake there reaches users immediately.
Analytics you get on day one
Three analytics surfaces are called out. The program describes a near-real-time dashboard with sub-one-hour latency pending owner confirmation. Cohort Explorer breaks retention and revenue down by signup month, week, or day. LTV Predictor forecasts twelve-month revenue using machine learning.
For a migration, the cohort view is usually the most immediately useful, because it lets you compare pre- and post-migration cohorts on the same axis. The LTV forecast is a planning input, not a substitute for your own historical revenue data, so keep your existing finance reporting alongside it during the transition.
Worked example: reading a paywall experiment result
A published customer case cited by the program, from Pixelcut, references a variant that produced a 16% increase in subscribers pending owner permission. Treat that as a dated reference point for what a paywall experiment can surface, not as a forecast for your app.
Here is how to read a result like that responsibly. Confirm the baseline period and the test period are the same length. Confirm the metric is subscribers, not trials started. Confirm the variant ran long enough to clear a full billing cycle. Then decide whether the lift is large enough to justify keeping the variant live, and whether it holds when you segment by store and by country. A single headline number is a prompt to look at the underlying cohort data, not a conclusion on its own.
What actually happens during an outage
There are two different failure modes worth separating. If the vendor's backend is unreachable, the SDK's locally cached entitlements keep working, so paying users retain access to what they already bought. That is the resilience story, and the program describes an uptime SLA on enterprise plans pending owner confirmation.
A different failure mode is a misconfigured entitlement or product mapping. Local caching does not help there, because the device is faithfully serving the wrong access rule. Migration planning should therefore include a post-launch audit of entitlement mappings, not just a reliance on cached state.
Build it yourself, or use StoreKit 2 directly?
Two objections come up repeatedly. The first is managing subscriptions in-house. The counterargument is that store updates break custom code, maintenance becomes an engineering tax, and you still end up without analytics, paywalls, or experiments unless you build those too.
The second is using StoreKit 2 on its own. StoreKit 2 operates on the local device; a cross-platform backend, real-time data, experiments, and a user management interface are the additions a platform layer brings. If your app is single-platform and you genuinely need none of those, direct StoreKit 2 may be sufficient. If you sell on more than one storefront or want experiment infrastructure, the build cost is the thing to weigh.
Data portability and lock-in
Events can be streamed to your own warehouse through webhooks or integrations, and raw CSVs can be exported at any time. That addresses the usual lock-in worry at the data layer.
What is not covered by an export is the configuration layer: paywall definitions, experiment assignments, and entitlement rules live in the platform. Plan for how you would reconstruct those if you ever left, and keep a documented copy of your entitlement model outside the vendor's system.
What the signup form asks for
The migration signup form collects first name, last name, email, and company name, plus a hidden field that is not shown to visitors. The submit control is labeled "Get started." Nothing in the form asks about your current billing provider or app count, so expect that context to come up in the follow-up conversation rather than at signup.
Frequently asked questions
Why not just manage subscriptions ourselves?
You can, but each store update can break custom billing code, and maintenance becomes ongoing engineering work. You would also need to build analytics, paywalls, and experimentation separately, since those are not part of a bare in-house implementation.
How does this compare with using StoreKit 2 directly?
StoreKit 2 works on the local device. A platform layer adds a cross-platform backend, real-time data, experiments, and a user management interface. If you are single-platform and need none of those, direct StoreKit 2 can be enough; the trade-off is the build and maintenance work for the rest.
How does it compare with other subscription platforms?
The program cites a fully native paywall builder, four-way experiments, predictive LTV, virtual currency support, and scale as its differentiators, and references third-party SDK adoption data as supporting evidence. Treat the superiority claim as a program assertion pending owner substantiation, and compare those specific capabilities against your shortlist rather than relying on a general claim.
Is my data locked in?
Events can stream to your warehouse via webhooks or integrations, and raw CSVs can be exported at any time. Configuration such as paywall definitions and entitlement rules is not part of that export, so keep your own copy of the entitlement model.
What happens if the platform goes down?
The SDK caches entitlements locally, so users keep access to what they already purchased. The program describes an uptime SLA on enterprise plans. Note that local caching protects against backend outages, not against a misconfigured entitlement mapping.