---
title: "Does My Small Business Need an App, or Just a Website?"
description: "Somebody told you your business needs an app. Name the repeated customer task first, test the cheap version, and know what an app commits you to after launch."
url: "https://arlingtonwebsitedesigner.com/blog/does-my-business-need-an-app-or-just-a-website/"
lang: "en"
published: "2026-04-28"
modified: "2026-04-28"
author: "Cal Hewitt"
tags: ["what-goes-on-a-site","what-it-costs","hiring-someone"]
image: "https://arlingtonwebsitedesigner.com/og/blog-does-my-business-need-an-app-or-just-a-website.jpg"
---

# Does My Small Business Need an App, or Just a Website?

[Cal Hewitt](https://arlingtonwebsitedesigner.com/about/cal-hewitt/)Published April 28, 202611 min read

- what goes on a site
- what it costs
- hiring someone

![Does My Small Business Need an App, or Just a Website?](https://arlingtonwebsitedesigner.com/images/post-does-my-business-need-an-app-or-just-a-website.jpg)

Somebody has told you that you need an app. A competitor launched one, or a developer raised it, or a customer asked whether you had one, and now the thought is sitting there. You cannot tell from the outside whether it is a real gap in your business or an expensive idea that sounds modern.

The reason the question is hard is that it is being asked in the wrong shape. App or website is a choice between two products, and you are not ready to choose a product yet. What decides this is whether there is a task your customers repeat often enough, and in a way a browser cannot carry, to be worth asking them to install something and come back to it.

That is a question you can answer yourself, this week, before anybody quotes you anything.

### Key Takeaways

#### Name the repeated task first

An app earns its cost when a customer does the same thing often, signs in, and needs something a browser cannot reach. Without that task named in one sentence, there is nothing to build against.

#### Work outward from the smallest option

The existing site, a mobile-first page, a booking or payment service and a browser-based experience all come before installed software.

#### Two platform charges are published and fixed

The Apple Developer Program is $99 per membership year and full Google Play distribution has a $25 one-time registration fee. Everything else about the price depends on your scope.

#### Launch is the start of the obligations

Both stores require privacy declarations that must stay accurate as your practices change, and security maintenance continues for as long as the app is available.

## What Job Would the App Actually Do?

Write one sentence naming the person who opens it and what they are trying to finish when they do. Not the features, the job. "A customer who books us every fortnight wants to move an appointment without ringing" is a sentence you can build from. "Customers want an easier experience" is not, because nothing about it can be tested or refused.

Then add the four facts that decide everything downstream. How often does that person do this, is it once a week or twice a year? Do they sign in, or would they have to create an account for the first time? Can you identify and reach them already, through an existing customer list, or would you be starting from strangers? And what do they do about it now, because whatever the current workaround is, your product has to beat it.

The US Small Business Administration frames this as ordinary [market research](https://www.sba.gov/es/guia-de-negocios/planifique-su-empresa/investigacion-de-mercado-y-analisis-competitivo): establish demand, market size, location, saturation and pricing before you commit, and use direct research when the question is specific to your own customers. That is the work in front of you, and none of it requires a developer.

## Which Questions Tell You Whether an App Has Work to Do?

Five conditions tend to separate the businesses that get value from an installed app from the ones that end up maintaining an empty one. An app earns its place when people use it repeatedly rather than once, when they sign in and it remembers them, when it needs the phone itself for a camera, a scanner or background location, when notifications are part of the job rather than marketing, and when it has to keep working with no connection.

A field crew logging work at sites with no signal meets several of those at once. A restaurant menu meets none of them, and a mobile page does the same job with nothing to install. Most businesses sit somewhere in between, and the useful exercise is counting how many of the five you can honestly claim. One is usually not enough. Three or more is a real conversation.

If none of them fit, that is not a disappointing answer. It means the money is better spent making the thing customers already reach for work properly on a phone, which is faster to do and easier to change.

Which condition does your case meet?

Hover or tap a row to highlight it.

| Condition | What it looks like in practice | Website can usually carry it |
| --- | --- | --- |
| Repeat use | The same customer does this weekly or monthly | Yes, if they do not mind typing the address |
| Signed in and remembered | Accounts, saved details, history | Yes, with a proper login |
| Needs the device | Camera, scanner, background location | Often no |
| Notifications are the job | A dispatch, a status change, a shift | Partly, and less reliably |
| Works with no connection | Reading offline is one thing, changing data offline is harder | No |

ConditionRepeat use

What it looks like in practiceThe same customer does this weekly or monthly

Website can usually carry itYes, if they do not mind typing the address

ConditionSigned in and remembered

What it looks like in practiceAccounts, saved details, history

Website can usually carry itYes, with a proper login

ConditionNeeds the device

What it looks like in practiceCamera, scanner, background location

Website can usually carry itOften no

ConditionNotifications are the job

What it looks like in practiceA dispatch, a status change, a shift

Website can usually carry itPartly, and less reliably

ConditionWorks with no connection

What it looks like in practiceReading offline is one thing, changing data offline is harder

Website can usually carry itNo

## How Do You Test the Need Before You Commission Software?

The cheapest test is asking the people who would use it. The SBA names surveys, questionnaires, focus groups and in-depth interviews as direct research methods, and for a small business the in-depth version is the one that pays. Ten conversations with real customers about the moment you have described will tell you more than any amount of internal discussion.

There is a second test that costs nothing and is more honest than a survey. Build the crude version of the path on your existing website, put it in front of people, and count how many complete it. A booking form, a payment link, a page that lets somebody check a status: these can be running within days. If nobody uses the free version of the task, an installed version of the same task is unlikely to change that.

Write down, before you start, what result would make you go ahead and what result would make you stop. Setting that threshold afterwards is how projects continue past the point where the evidence stopped supporting them.

![A notebook page listing one customer task in a single sentence, with tally marks beside it counting how often it happens each week](https://arlingtonwebsitedesigner.com/images/post-does-my-business-need-an-app-or-just-a-website-1.jpg)

## What Does a Responsible Build Look Like From the Inside?

If the evidence supports building, the shape of the work is predictable and worth knowing before you agree to it. Discovery comes first and it ends with a written scope, not a price: who uses it, what they are trying to finish, which devices and languages are in scope, what it must connect to, what accessibility and privacy obligations apply, and what counts as done. A prototype comes next, something you can tap through on a real phone with nothing working underneath, because that is where a misunderstanding costs an afternoon instead of a month.

Then the build, in reviewable pieces, tested against the acceptance criteria you already agreed. Security belongs in this stage rather than after it: [NIST's secure development guidance](https://csrc.nist.gov/projects/ssdf) treats preparing, protecting, producing and responding to vulnerabilities as continuing practices, and the last of those does not stop at launch.

Submission is its own piece of work. Apple requires [App Privacy information](https://developer.apple.com/help/app-store-connect/manage-app-information/manage-app-privacy/) covering what is collected and how it is used, including the practices of any third-party code you have integrated, and Google requires a completed [Data safety form](https://support.google.com/googleplay/android-developer/answer/10787469?hl=en) for published apps. Both mean somebody has to inventory every analytics, advertising, payment, booking and chat component in the product and describe honestly what it does with data.

### The gates, rather than a schedule

1.  1

    #### Discovery ends

    When you can state the customer, the task and the test that would disprove it.

2.  2

    #### Prototype ends

    When you have tapped through the path on a real phone and accepted it.

3.  3

    #### Build ends

    When the acceptance criteria you agreed in writing actually pass.

4.  4

    #### Submission ends

    When the store accepts the release, which is not the same day you submit.

5.  5

    #### Maintenance

    Runs for as long as the app is available, and is agreed before launch, not after.

Notice that none of those gates is a date. Duration depends on how many paths, how many integrations, how many platforms, how ready your content is and how quickly decisions come back from your side. A schedule offered before the scope exists is a guess with a calendar attached to it.

## What Can You Price Now, and What Has to Be Scoped?

Two numbers are published and will not move with your project. The [Apple Developer Program](https://developer.apple.com/programs/enroll/) is $99 per membership year for distribution through the App Store, and [Google Play](https://support.google.com/googleplay/android-developer/answer/6112435?hl=en) charges a $25 one-time registration fee for full distribution. If you are distributing a private app to your own staff rather than to the public, the [Apple Developer Enterprise Program](https://developer.apple.com/support/compare-memberships/) is $299 per year for eligible organisations.

Everything else is scope. The build price depends on the paths, the integrations, the platforms, the content, the data involved and the testing, which is why a figure offered before the scope is written is not a quote. What you can do is insist that any proposal separates the one-time work from the recurring work, because the second one is where budgets fail. One-time covers discovery, design, build, content, testing, launch, legal review and account setup. Recurring covers hosting, subscriptions, platform memberships, third-party services, monitoring, support and the changes you will want in year two.

The SBA makes the same split when it walks a business through [costing itself](https://www.sba.gov/counseling/plan-your-business/), and it notes that some costs are well defined while others have to be estimated by asking vendors directly. Ask for both columns before you compare two proposals, because a cheap first column with an unexamined second one is the common shape of a project that becomes unaffordable to keep.

## What Should the Contract Settle Before Anybody Starts?

Ownership is the term that matters most and the one people assume is handled. Under [17 U.S.C. 201](https://uscode.house.gov/view.xhtml?req=%28title%3A17+section%3A201%28b%29+edition%3Aprelim%29), copyright vests initially in the author, and a qualifying work made for hire is owned by the commissioning party only where the statute applies or a signed written agreement says so. Paying an invoice is not, by itself, a transfer of every right, so the agreement has to say what you get.

Beyond ownership, read for the deliverables, the payment schedule, how changes are priced, what the acceptance test is, who is responsible if the store rejects the release, the support period, every recurring charge, and what happens at the end including handover steps. Ask specifically who holds the developer accounts, the source repository, the cloud account, the analytics, the payment account, the design files and the customer data, on day one and on the day you part company. Those should be in your name from the start, because they are painful to move later.

There are also platform terms you inherit. The Apple Developer Program agreement requires a privacy policy explaining collection, use, disclosure, sharing, retention and deletion of user or device data, and the [App Review guidelines](https://developer.apple.com/app-store/review/guidelines/) require that policy to be reachable from both the app and its listing. Google requires a [complete privacy policy](https://support.google.com/googleplay/android-developer/answer/11150561?hl=en) in Play Console and inside the app where personal or sensitive data is involved. A Texas attorney should look at any agreement that matters to you.

![A printed contract on a desk with the ownership and handover clauses marked in pen, beside a phone showing an app store listing](https://arlingtonwebsitedesigner.com/images/post-does-my-business-need-an-app-or-just-a-website-2.jpg)

## Which Parts Can You Do Yourself?

More than you would expect, and they are the parts that decide whether the project is any good. Write the customer problem in plain language. Map the path a customer takes today, including the phone calls. Interview ten of them. Gather the service rules, the prices, the policies and the content. List who holds each operational account. Keep the wish list somewhere visible so it does not arrive as a surprise mid-build. Test the agreed tasks yourself on a real device, and keep a record of what you approved and when.

The line is where a change touches authentication, payments, personal data, security settings, production code, cloud access, store declarations or legal wording. You can supply every fact behind those, and you should not be the one guessing at how they are represented. Apple's privacy answers have to cover integrated third parties, Google's data declaration has to be accurate, and both stay your responsibility as the business even when somebody else fills the form in.

## What Do Arlington Resources and Texas Law Add?

Neither Arlington nor Texas requires an ordinary small business to run an app, so nothing local forces this decision. What is available locally is help making it. The City points businesses to its [small business assistance](https://www.arlingtontx.gov/Business) pages, and its resource round-up names the Arlington Economic Development Corporation, LiftFund-supported zero-interest loans and the Greater Arlington Chamber of Commerce. The Tarrant Small Business Development Center offers feasibility analysis, business-plan development and advising, which is exactly the kind of second opinion worth having before a custom software commitment.

Texas law can reach the product itself. The [Texas Data Privacy and Security Act](https://statutes.capitol.texas.gov/Docs/BC/pdf/BC.541.pdf), Business and Commerce Code Chapter 541, regulates how certain businesses collect, use and process consumers' personal data, and gives consumers rights including confirmation, correction, deletion, portability and opt-outs for targeted advertising and sale. Whether it reaches your business depends on thresholds and exemptions in the statute and on what data you actually plan to handle, so read it against your real data flow and take legal advice on the answer.

## How Do You Avoid the Next Expensive Decision?

Keep a decision record, and keep it before the work rather than after it. For any new channel, write down the customer, the repeated problem, what they do instead today, the action you want, how you will test it, what data is involved, who owns it, the budget ceiling, the ongoing obligations and the condition that would make you stop. It fits on one page and it is the difference between a considered purchase and a persuasive conversation.

You will know the process worked when every field has a written answer, a named person has accepted the scope, account ownership and handover are documented, and the thing you built passes the task test you agreed in advance. Set that threshold before work starts. A success measure invented after launch tends to be whatever the launch achieved.

**Quiz: Would an app earn its place in your business?**

1. A customer wants to move a booking without calling. How many of the five conditions does that meet on its own?
   - All five, so build the app
   - Repeat use and signing in, so two, and a website can carry both **(correct answer)**
   - None, so the idea is worthless
2. Your crew records job details at properties with no mobile signal. What does that point at?
   - A mobile-friendly website, since the data uploads later anyway
   - An app, because changing data offline is the case a browser handles worst **(correct answer)**
   - Paper, because software cannot work offline
3. You have written a scope and want a launch date before signing. What is reasonable to expect?
   - A firm calendar date for public release
   - Gates you can both recognise, with store acceptance outside anybody's control **(correct answer)**
   - A date, with a penalty clause if it slips

## Frequently Asked Questions About Apps and Websites for Small Businesses

**Is a mobile website enough for my business?** For most local businesses, yes. A modern site on a phone can take bookings, take payments, hold an account and work offline to a degree. The cases it handles worst are the ones needing the device itself or letting somebody change data with no connection.

**Will customers actually download it?** That is the question to test before building, not after. Ask ten of them about the specific task, and run the crude version of it on your website first to see whether anybody completes it.

**Do I need both an iPhone and an Android version?** If it is for the public, assume yes. One shared codebase can serve both, and it does not remove the need to test on both, because the platforms behave differently and each store has its own requirements.

**Should I build an app before I have a proper website?** The site is the thing search engines and AI assistants read, so it does the finding-you work that an app cannot. An app serves people who already know you.

**What does it cost to keep an app alive?** The published platform charges are $99 a year for Apple and a $25 one-time Google Play registration. Beyond those, budget for updates as phones change, security response, privacy declarations that stay accurate, and the changes you will want.

**Can a web app do what I need?** Often. It is worth writing the exact action and its constraints before accepting that only an installed app will do, because the answer turns on the constraints rather than on the label.

- Native app
- Web app
- Scope
- Acceptance criteria
- Data safety form
- App Privacy

## Moving Forward

App or website is a product choice, and it is the second question. The first is whether a repeated customer task exists that a browser cannot carry, done by people you can identify and reach. Name that task in one sentence, count how many of the five conditions it meets, and test the cheap version before commissioning the expensive one. If the answer comes back as a website, that is a result rather than a setback.

Doing it in that order leaves you with something useful either way: a written decision record, ten conversations with real customers, and a threshold you set before anybody was invested in a particular outcome. That record is what stops the same expensive question arriving again next year in a different costume.

If you have been told you need an app and want a straight read on it, we can help you work through it. [Arlington Website Designer](https://arlingtonwebsitedesigner.com/) builds websites and custom software for businesses in Arlington, Texas and the surrounding Tarrant County towns, and for clients across the country, and the app conversation starts with the customer task rather than the platform. Send what you have been told and what your customers actually do repeatedly through [the contact page](https://arlingtonwebsitedesigner.com/contact/), and you will get an answer on whether this needs an app at all.

## More articles

- [How Long Before a New Website Shows Up on Google?](https://arlingtonwebsitedesigner.com/blog/how-long-before-a-new-site-shows-up-on-google/)
- [How Long Does It Take to Build a Website? Where the Weeks Go](https://arlingtonwebsitedesigner.com/blog/how-long-does-it-take-to-build-a-website/)
- [The Website Accessibility Letter, and What Reduces the Risk](https://arlingtonwebsitedesigner.com/blog/website-accessibility-and-the-demand-letter/)

## Thinking about a site that does this for you?

Tell us what your business does and where you want to be found. We will tell you what we would build and what it would take.

[Start a project](https://arlingtonwebsitedesigner.com/contact/)