Most congregations and small nonprofits that commission an app should not have. The requirements, announcements, events, giving, media and a directory, are met by established products at a fraction of the cost of building, and the money saved does more good elsewhere. Custom is justified rarely and specifically, and it is worth testing that honestly first.
Price the total cost, not the build
A custom app costs the build, then annual platform compatibility work, then store account fees, then someone's time to manage releases, for as long as it exists. Compare that against a subscription product over five years rather than against the build quote alone. For most organisations the comparison is not close, and making it explicitly is the most useful thing a board can do before approving the project.
When custom is genuinely justified
When the organisation runs a programme no packaged product models, such as a structured course with progress tracking or a volunteer coordination workflow with real complexity. When an existing system of record must be integrated and no product will. Or when the app is part of the mission rather than administration of it. Those are real reasons. Wanting your own branding in the store is not, and it is the usual one given.
Giving is the part to handle carefully
Donations mean payment handling, and the safest architecture keeps card data out of your systems entirely using a provider's hosted fields or tokenisation, so the PCI Data Security Standard applies as narrowly as possible. Recurring giving adds failed payment handling, card expiry and a way for donors to change or cancel without phoning someone. Receipts and record keeping for donors matter too, and are frequently an afterthought.
Keep it maintainable by whoever comes next
Volunteer and small staff organisations have turnover, so require documentation written for someone with no context, your own store accounts and repository, and an express written copyright assignment, since under 17 USC 101 commissioned software is not automatically a work made for hire. An app only one departed volunteer understood is the most common way these projects end, and it ends with the app quietly disappearing from the store.
Questions people ask about church app development
Should our church or nonprofit build an app?
Usually not. Announcements, events, giving, media and a directory are met by established products at a fraction of the cost. Compare a five year total against a subscription, including annual platform work and release management, before approving a build.
When is custom justified?
A programme no packaged product models, a system of record that must be integrated and cannot be, or an app that is part of the mission rather than its administration. Wanting your own branding in the store is not a sufficient reason.
What should we require if we do build?
Documentation written for someone with no context, your own store accounts and repository, and an express copyright assignment. Turnover is high in these organisations and an app only one departed person understood will disappear.