Development guides / New apps and first releases
Your first app release: complete one useful task from start to finish
The first release should let its intended users finish a useful task. That includes saving the data and giving staff a way to handle what was submitted.

Start with the task a user needs to finish
You may already know why you need an app: customers want to request a service, staff need to record work, or an existing service needs mobile access. The harder part is choosing a set of features that people can actually use to finish that job.
For example, a customer may submit a request successfully on their phone, while staff have no way to view, assign or respond to it. Plan both sides of the journey before deciding which features deserve investment in the first release.
Compare a mobile website, an app and a prototype
For occasional forms, information lookup or booking, compare a mobile website with an app. If offline work, device features or frequent use matter, specify the required app capabilities. Decide whether platforms should share code based on functionality and maintenance needs.
Separate features needed to complete the main job from later improvements. Permissions, essential administration, error handling and persistent data affect the whole journey; screen count alone is a poor scope estimate.
Clickable prototype
Use simulated data to test whether people understand the screens and steps. A clickable prototype does not establish that data storage, permissions or external connections work in production.
Mobile website
Worth evaluating when people occasionally open a link, submit a form or look something up. Device features and offline requirements still need testing in the intended browsers.
Production app
An app may suit frequent use or work that depends on device features. Confirm the target platforms and distribution method, and include administration, testing and release preparation in the plan.
Choose the first version of a service-request app
The example below illustrates an approach. It is hypothetical, not a client case study or a claim of delivered results.
Consider a service-request app: customers register, submit a request and track progress; staff receive, assign and update it. Connect that journey in the first release. Decide on referral rewards, social features and complex reports after observing real usage.
Three things a first release often leaves out
Who handles the submission?
Staff are users too. Decide who can view and update requests, and when the customer receives a reply. Some early steps can remain manual, provided the handover is clear. Otherwise, the screen may say “in progress” while nobody has picked up the work.
What happens when the connection fails?
The app may be sending, the server may have saved the request without a reply reaching the phone, or nothing may have been sent. Show the user what to do in each case. Interrupt the connection and retry during testing, then check the records for duplicates as well as the screen message.
Can you operate what you receive?
Before development, agree who controls the source code, developer accounts, hosting and external services. Include instructions for releasing updates. An installation link is only part of the handover; account access and release instructions also help with maintenance and changes of developer.
A demo, a pilot and a store release are different milestones
- Run the full journey as a customer and an administrator; verify permissions, saved data and state after signing in again.
- Test network interruptions, repeated submissions, missing data and third-party failures. If payment is required, reconcile payment and order states.
- Agree separate milestones for a testable build, a submission-ready build and public launch. Store review is controlled by the platform; finished code does not mean store approval.
With a tight deadline, what should you reduce?
When comparing app quotations, check design, frontend and backend, permissions, test devices, integrations and handover. Existing screens or AI-generated interfaces can be a starting point, but data handling, failure cases and administration still need checking.
If the date is tight, consider a smaller release, staged testing or an internal trial. An external connection that is not ready can stop the whole task from working. Allow time for store review and any changes it requires when planning public launch.
Use the first release to decide what comes next
Watch whether people finish the main task, where they stop and whether they use the app again when they need it. If service requests receive no reply, adding social sharing will not fix the problem. Improve the steps that lead to a completed task first.
Ask someone to carry out an actual task. Watch where they stop, then ask why. Record the difficulties they encountered and the features they would like separately. Compare these with workload, maintenance costs and business needs when setting priorities for the next release.
Put the need into a one-page brief
Provide the intended users, the most important journey, any designs or sketches, data sources and developer accounts you control. Allow for design decisions, content, integration testing, acceptance and store responses. Handover should cover source code, accounts, deployment instructions and maintenance.
Questions before you start
Do AI-generated screens mean most development is finished?
Screens make discussion easier, but working behaviour determines readiness. Walk through the journey as both customer and administrator, checking storage, permissions, failures and deployment. That reveals what can be reused and what still needs implementation.
Can the first release target fewer platforms?
Evaluate that against your actual users and working context. Start with the necessary platform or pilot distribution, document excluded needs, and decide on expansion using real use and the ongoing maintenance cost.

