No code or hire an agency? Decide on what you are holding in six months.
Both routes end with an app. They end with very different situations around it. An agency sells you judgement and a finished deliverable. A builder sells you the ability to change the thing yourself on a Tuesday afternoon without raising a ticket.
Cost is the thing people compare and it is the least stable part of the decision, because it changes with scope and nobody publishes real numbers. What does not change is who can modify the app after launch, and that is usually what people regret.
Compare what you end up holdingThe short version
You are not choosing a price. You are choosing who can change it next month.
Apps are not finished at launch. Stores change rules, operating systems move every autumn, and the first real users ask for things nobody predicted. So the question that matters is what happens the first time you need a change.
If the answer is a statement of work and a two week lead time, that is a real cost that does not appear in the original quote. If the answer is that you do it yourself, that is a real capability that does not appear either.
What are you actually buying from an agency?
Judgement, mostly. An experienced team has seen the failure modes: the permission that gets refused, the store rejection reason, the thing that is cheap now and expensive later. That is genuinely worth money and it is the part that is hard to replace.
You are also buying absorption of risk. If it does not work, that is their problem to fix within the contract, which matters when the app is load bearing for a business rather than an experiment.
What you are not necessarily buying is ownership in the sense people assume. Whether you receive the source, who holds the developer account, and what happens if you stop paying are all contract terms rather than defaults. They are worth settling before the work starts rather than after.
The clearest way to price judgement is to notice what you are not asking. If you already know what the app must do, who uses it, and what the rules of your domain are, you are buying hands rather than judgement, and hands are the part that has got cheaper. If any of those three is still open, that is what the agency is for. Building it yourself against hiring a developer is the same question at individual rather than firm scale.
What do you end up holding in each case?
With a builder, the thing to check before you commit is whether the output is a codebase or a configuration. A real React Native and Expo project can be opened, read and handed to any mobile developer. A configuration inside a platform account can only be changed inside that platform, and only while you keep paying for it.
That difference decides what your options are later. If a builder stops being the right fit, a codebase is a handover and a configuration is a rebuild. It is the single most consequential thing on this page and it is rarely on the comparison tables.
With an agency the equivalent question is simply asked out loud: do we receive the repository, and is the developer account in our name. A reasonable agency will say yes to both. The answer being unclear is itself the answer.
There is a middle position that often gets missed. You can own the repository and still pay someone to work in it, which keeps the judgement available without the dependency. That arrangement is normal and most agencies will do it if asked at the start. How agency app development usually works covers what the arrangement looks like in practice.
Who is allowed to submit it to the App Store?
This is the rule that surprises people on both routes. Apple's guideline 4.2.6 says apps created from a commercialized template or app generation service will be rejected unless they are submitted directly by the provider of the app's content, and that these services should not submit apps on behalf of their clients.
So if an agency builds your app on top of a template or generation service, the agency submitting it for you is the thing the rule is aimed at. It is a fair question to ask any agency: what is this built on, and whose developer account does it go out from.
On the builder route the same rule points the same way. You submit, under your own account, as the owner of the content. That means an Apple Developer Program membership at 99 USD a year, or 25 USD once for Google Play, and a person on your side who has done it. Budget for that step rather than assuming it is included.
Budget for the store side separately whichever route you take, because it is nobody's default inclusion. The Apple Developer Program is 99 USD a year and Google Play is 25 USD once, and somebody on your side has to hold those accounts. What app development actually costs breaks down the parts that recur against the parts that do not.
So which should you choose?
Hire an agency when the app is load bearing from day one, when being wrong is expensive, when the domain has rules you do not know, or when nobody internally will own it afterwards. Those are the cases where judgement and absorbed risk are the actual product.
Use a builder when you are the person who understands the problem, when the first version is a question rather than a commitment, and when you expect to change it often. Being able to make a change in an afternoon is worth more than it sounds, and it is the thing an agency relationship cannot give you at that cadence.
The combination that works well in practice is building the first version yourself to learn what it actually needs, then paying someone to do the part that turned out to be hard. That is a cheaper way to buy judgement, because you know what you are buying it for.
If you are leaning towards building and the question is which tool, the filter is the one from the table rather than the feature list. Choosing a builder as a non technical founder is organised around what you can do on your own afterwards, which is the property that decides whether the decision ages well.
The questions that actually separate the two
| Question | Agency | Builder, codebase output | Builder, platform configuration |
|---|---|---|---|
| Who can change it next month | Them, on their schedule | You, or any developer | You, inside that platform only |
| What you hold at the end | Whatever the contract says | A repository you own | An account you rent |
| Who submits to the App Store | Ask: 4.2.6 may forbid it | You, under your account | You, under your account |
| Where the judgement comes from | Bought with the work | Yours, or hired later | Yours, or hired later |
| Cost shape | Quoted per project | Subscription plus your time | Subscription plus your time |
If you are leaning towards building it
The check that matters is the one in the table: codebase or configuration. With Newly you describe the app in plain English and get a real React Native and Expo project you own, which keeps the agency option open rather than closing it.
That is the genuinely useful position to be in while you are still deciding. Build the first version, find out what is actually hard about your problem, and then either carry on or hand a working repository to someone who does this for a living.
Newly is a paid product from $25 a month with no free tier. The store fees are separate and are yours either way.
Questions people ask when choosing between the two
Per project, usually, but the comparison is not like for like. An agency quote includes judgement and someone else carrying the risk. A subscription does not include your time, and your time is the largest line item on the build it yourself route. Neither of those is published as a number you can compare.
Decide on what you will be holding
If you want to keep both routes open, start with something you own: describe the app in plain English and get a real Expo project you can hand to anyone.
Start building