Two stores, one budget
Mobile App Development for Small Business: iPhone, Android or a Web App First
Two native apps mean two codebases, two sets of store rules and two things to keep working, all for a company that may have one person. There is a cheaper question to ask first: does this job need to be in a store at all? The answer changes the bill by thousands.
November 8, 2026 · 9 min read
The short answer
A small business has three routes to the phone: a web app that runs on iPhone and Android from one build, from $575 depending on complexity; one native app, from $3,750; or two native apps, which add to that price. Start with the web app unless the job needs what only a native app does well. Google Play charges $25 once; Apple charges $99 a year, as of November 2026.
On this page
- Start with the cheapest thing that fits in a pocket
- Three routes to the phone, priced side by side
- Android's rent has a different shape from Apple's
- Build one screen for one person, then watch what happens
- The walls a web app runs into on an iPhone
- Ask who is holding the phone before choosing a store
- When to put the checkbook back in the drawer
Start with the cheapest thing that fits in a pocket
Getting onto your customers' phones does not require an app. Pew Research Center's fact sheet, published October 19, 2026, says 91% of U.S. adults own a smartphone. All of those phones already have a browser, and browsers do not ask anybody to visit a store, make a password, wait for a download and find room on a crowded screen. Each of those steps loses people, particularly people who need you once.
So the question for an owner with one budget is not iPhone or Android. It is which job, for which person, how often. A customer booking twice a year belongs on your website. A crew member logging eight jobs a day may belong in an app. Between those lies a large middle ground, where a web app does the work for a fraction of the cost, and most small companies should begin there. The page on whether a small company needs an app gives you the tests.
Three routes to the phone, priced side by side
Here is the comparison as DataCram prices it, with the store charges beside it, as of November 2026. The native price is for a focused app. A back end and admin screens, or a long feature list, add to any route.
| Web app | One native app | Native on both | |
|---|---|---|---|
| Build | From $575, depending on complexity | From $3,750 | More than one native app; two codebases add to the price |
| Runs on | Any phone with a browser | iPhone or Android | iPhone and Android |
| Store account | None | Apple, $99 a year, or Google Play, a one-time $25 | Both |
| Store review | None | One store's rules | Two stores' rules |
| Updates | Change it once, everyone has it | Submit, wait, users update | Do that twice |
| Good at | Forms, lists, dashboards, booking, portals | Camera, offline work, location, notifications | A paying audience on both platforms |
| Bad at | Reliable reach to iPhone users through push; a store presence | Anyone who will use it once | Being cheap to keep running |
Two details on the store accounts. Google says it does not refund the $25 if the identity information you submit is judged invalid, so have your business details straight first. And Apple's yearly charge has no exemption for sole proprietors, as the iOS page for small companies explains along with the rest of Apple's rules. Our mobile app development service gives the general picture, and the limited budget plan shows how to start with one user and one job.
Android's rent has a different shape from Apple's
If you plan to charge for anything inside the app, read this part slowly. Google says it has no single service fee, and its table takes some reading. Google Play's fee table, read November 6, 2026, shows the following for users in the US since its June 30, 2026 rollout. Check it for your own program and install type before you put a number in a business plan.
| Rate as printed | Source | |
|---|---|---|
| Google Play, auto-renewing subscriptions | 10% plus a 5% billing fee | |
| Google Play, other transactions on new installs | 20% plus a 5% billing fee | |
| Google Play, the row labeled First $1M (USD) of annual earnings | 10% plus a 5% billing fee | Google; the row does not spell out its conditions |
| Apple Small Business Program, paid apps and in-app purchases | 15% | Apple, for developers new to the store or with up to $1 million in prior-year proceeds |
Google also says 97% of developers distribute on Play at no charge, and that 99% of those subject to a service fee are eligible for 15% or less through its programs. For a very small company, the practical lesson is shorter than the table. If the app books real-world jobs or works as a tool for your crew, these fees may never apply to you, so ask before assuming. If you plan to sell a subscription through the app, build these numbers into the price from the start. Our micro-SaaS page deals with that kind of product.
Build one screen for one person, then watch what happens
The usual way to spend too much is to imagine the finished product. The better way is to pick the most annoying thing that happens in the business, and make a screen for it. Say a two-van pest control company in Waco, invented for this example, loses paper service tickets. The first version is one screen: pick the job, tap what was treated, take a photo, send. That is the whole app, and it should stay that way until someone complains.
- 1
Write down who holds the phone and what is in their other hand
A person in gloves on a roof needs big buttons and few of them. A person at a desk does not need an app.
- 2
Name the data, and where it already lives
If jobs live in your field service tool, connect to it instead of rebuilding it. Integrations go live the same day.
- 3
Build it in the browser first
A tool from $575, depending on complexity, is a cheap way to discover that the screen you imagined is the wrong one.
- 4
Use it for a real month
Watch whether your people open it unprompted. If you have to remind them, the app has a problem that going native will not fix.
- 5
Go native only for a reason you can name
A camera workflow, working with no signal, or a push notification that must arrive. Without a named reason, stay where you are.
The walls a web app runs into on an iPhone
Web apps are the cheap route, and fairness requires listing what you give up. Most of the limits show up on iPhones, and they matter more for customers than for staff, because you can stand next to your crew member and show them how.
- Nobody is prompted to install it. MDN says the automatic install prompt is not supported on iOS. Users have to find the Share menu themselves, which on iOS 16.4 and later works in Safari, Chrome, Edge, Firefox and Orion.
- Notifications take effort. WebKit said on February 16, 2023 that Web Push on iOS and iPadOS 16.4 requires the site to be added to the Home Screen, a manifest with display set to standalone or fullscreen, and permission asked in response to a direct action. A customer who never added you never hears from you.
- You do not appear in the stores. If people will look for you there, a web app will not be found. Someone with a leak is probably not browsing an app store for a plumber, but check for your trade.
- Heavy camera or offline work gets harder. A form with a photo works fine. A full day in a basement with no signal is what native apps are for.
The requirements for installing a web app are modest: MDN lists a manifest and a secure origin, and a service worker is not required. The iPhone comparison of native and web has a decision table. If a customer-facing app is what you had in mind, also read about a portal in a browser, which skips the install question altogether.
Ask who is holding the phone before choosing a store
Choosing one platform is easier than it sounds when you sort by who the user is. With a crew, you control the answer. With customers, you do not, so you look at the numbers instead of arguing about them.
Staff and crew tools
- Count the phones, and pick the platform most of them use
- Consider buying the same phone for three people, against paying for two builds
- Skip public listing where the store allows it
- See employee scheduling, inspection checklists and inventory
Customer apps
- Read your site's device report before you decide
- Expect few installs unless customers return weekly
- A browser reaches both platforms with one build
- See loyalty, portals and booking
Before you build a field tool, also price what already exists. Jobber, Housecall Pro and ServiceTitan have spent years on the job screens you are imagining. Our page on field service apps compares building with renting, and renting often wins. What we would then build is the integration that connects your site to the tool you chose.
When to put the checkbook back in the drawer
These are the moments we would tell an owner to spend nothing. None of them is rare, and none of them gets mentioned at the app developer's conference booth.
- The main reason is that a competitor has one. Ask the competitor how many people open it.
- A rented tool does the job at a monthly fee less than the upkeep of your own. Rent it, and revisit in a year.
- You cannot name the person who will test it after each phone update. That person will be you, at 10 p.m.
- Your website does not yet take bookings. Fix that first, since it reaches everyone. The website package starts at $1,500 and includes it.
- The job is really an estimate and an invoice, and accounting software you already pay for has a screen for that.
Questions owner-operators ask
Often not. A web app runs on both from one build, starting at $575 depending on complexity, and works for most forms, lists and dashboards. Two native codebases add to the $3,750 starting price and to the upkeep forever. Build both only when you have a paying audience on both and a job a browser handles badly.
Google charges a one-time US$25 registration fee, as of November 2026, and does not refund it if the identity information you submit is judged invalid. Service fees on sales are separate and vary by program and install type, so read Google's fee table for your case. Apple's account costs $99 a year by comparison.
Yes, with conditions. WebKit says iOS and iPadOS 16.4 added Web Push for Home Screen web apps. The user must add the site to the Home Screen and grant permission after a direct action. That works for a crew you can show how. It works poorly for customers who will never take those steps, and that is a fair reason to consider native.
One screen for the most annoying thing that happens every day, for the person it happens to. Skip accounts, menus and settings until someone needs them. Build it in the browser, use it for a real month, and let what your people actually do decide what comes next. Our limited budget page walks through this.
You probably already have one, since the tool likely ships its own. If your crew needs something it does not do, we can build a small screen that talks to it through its API, rather than rebuilding your schedule. That integration goes live the same day. Start with the field service comparison to see what the existing tools cover.
Keep reading
- One owner, one phoneWeb Design for One-Person Businesses: What You Need and What to RefuseWebsite design for one-person businesses: the short page list, what to refuse to pay for, and fixed prices from $1,500, or when a builder plan is better.
- The build sideSmall Business Website Development: Who Holds the Keys When the Builder LeavesWebsite development for independent contractors: who owns the code, domain and accounts, what custom means at $1,500, and what you must maintain yourself.
- Probably not. Sometimes yes.iOS App Development for Small Businesses: When an iPhone App Earns Its KeepWhen an iPhone app makes sense for a very small company, what from $3,750 buys, Apple's fees and rules as of 2026, and what to try before you build.
- Small, and yoursCustom Software for Microbusinesses: Small Tools That Retire the SpreadsheetDashboards, portals and integrations from $575, for owners who run the business from a phone and a spreadsheet, and the rented tools to try first.
- Narrow searches, small billsSEO for Microbusinesses: A Small Program for a Small Site, and When to Skip OursSEO for a small business site: an hour-a-week plan you can run alone, when our $575 program is worth it, and the sales lines to walk away from.
- One town at a timeLocal Marketing for Microbusinesses: One Town, One Truck, One ProfileLocal marketing for one-truck and home-based businesses: the free profile first, reviews within Google's rules, and what our monthly plan from $575 covers.
- Everything in MicrobusinessSee the section
Pick the cheapest route that works
Tell us who would use it, what phones they hold and what they do with it. We reply with the route we would take and its starting price, including when that route is a tool you can rent.
Which of these sounds like you? (pick any)
Rather talk now? Call (210) 346-0848.