Mobile AppsWeb Design

Should a Small NZ Business Build a Mobile App, or Is a Website Enough?

Kage Works10 August 20269 min read

Apple answers this in writing before you get a quote. Guideline 4.2 of the App Store Review Guidelines reads: "Your app should include features, content, and UI that elevate it beyond a repackaged website. If your app is not particularly useful, unique, or 'app-like,' it doesn't belong on the App Store." Guideline 4.2.2 adds that apps "shouldn't primarily be marketing materials, advertisements, web clippings, content aggregators, or a collection of links" (Apple).

The introduction to those same guidelines sends you somewhere else: "For everything else there is always the open Internet. If the App Store model and guidelines or alternative distribution and Notarization for iOS and iPadOS apps are not best for your app or business idea that's okay, we provide Safari for a great web experience too."

That rules out the app most small business owners picture, which is their website with an icon on it. An app has to give a customer something the mobile site can't, and Apple's reviewer decides whether it does. Everything below is about what clearing that bar costs in New Zealand, and the two dates in 2026 that turn an app into an ongoing obligation rather than a one-off build.

Apple and Google gate the account before anyone looks at your app

You need a developer account on each store, and neither one hands them out on the spot.

Google Play splits accounts into personal and organisation. An organisation account requires a D-U-N-S number, the nine-digit business identifier Dun & Bradstreet issues free, along with legal business documents, an identity document for the account owner, and verification of your website through Google Search Console (Play Console Help). The number costs nothing, though it doesn't arrive the same week you ask, and the Search Console step means you need a website before you can publish an app.

Take the personal account route instead and you hit a different wall. Personal accounts created after 13 November 2023 must run a closed test with at least 12 testers opted in continuously for 14 days before Google will grant production access, and Production stays disabled in Play Console until you meet it (Play Console Help). Google cut the requirement from 20 testers to 12 in December 2024. Twelve people with Android phones and Google accounts, all installing and keeping your test build for a fortnight, is a real errand for a five-person business.

The fees run in opposite directions. Apple's Developer Program costs US$99 a year, every year (Apple). Google Play charges US$25 once for the life of the account. Over a decade that's roughly US$990 against US$25, before anyone writes a line of code.

Apple's review itself is quicker than its reputation suggests. Apple states that "on average, 90% of submissions are reviewed in less than 24 hours," and that "over 40% of unresolved issues are related to guideline 2.1: App Completeness," which covers crashes, placeholder content and broken links to support or privacy pages (Apple). The delay sits earlier, in the account setup, and the rejection risk sits in 4.2.

Two 2026 deadlines make an app a subscription, not a purchase

A website you leave alone for three years still loads. An app you leave alone for three years stops being shippable, and then stops being visible.

Since 28 April 2026, Apple requires that "apps uploaded to App Store Connect must be built with Xcode 26 or later using an SDK for iOS 26, iPadOS 26, tvOS 26, visionOS 26, or watchOS 26" (Apple). Changing your phone number in the app means rebuilding on a current toolchain. Whoever built it needs to still hold the signing certificates and a machine running the current Xcode.

Google's rule costs you more, because it applies whether or not you touch the app. From 31 August 2026, existing apps must target Android 15 (API level 35) or higher. Apps that don't "will only be available on devices running Android OS that are the same or lower than your app's target API level" (Android Developers). Your app doesn't get deleted. It quietly stops appearing for anyone shopping on a newer phone, while existing users keep it. Developers can request an extension to 1 November 2026. New apps and updates submitted from 31 August 2026 have to target Android 16 (API level 36).

Neither deadline is a one-off. Both companies raise the bar once a year. Budget an app as a yearly recurring engineering cost with a build attached at the front, and ask any agency quoting you who pays for the compliance rebuild in year two. Our questions to ask before hiring a web design agency covers how to pin that down in writing.

Push notifications stopped being an app-only feature in 2023

The strongest argument for an app used to be the notification. Apple removed it. In the WWDC23 session on web apps, Apple's engineers state: "In iOS and iPadOS 16.4, we added notifications to Home Screen web apps, using the Web Push standard with support for badging" (Apple). A site added to the Home Screen can now ask permission, send push notifications that appear on the Lock Screen and a paired Apple Watch, and set a badge count on its icon. Android has supported web push for years.

The catch is the install step. Safari has no automatic install prompt, so an iPhone user has to tap Share and then Add to Home Screen, and developers have been asking Apple for a programmatic prompt (Apple Developer Forums). That friction is real, and it's smaller than the friction of getting found in a store. You also lose the App Store listing itself, which matters if customers search the store for your category.

The frequency test decides it

An app earns a slot on someone's home screen or it gets deleted. Count how many times a year one customer transacts with you, and the answer usually falls out.

A roofer or a conveyancer sees a given customer once, maybe twice in a lifetime. Few people install software for a purchase they make that rarely. The mobile site, the quote form and the Google Business Profile carry the whole job, and our post on whether a website is still worth it covers what that looks like by trade.

A gym, a physio, a childcare centre or a café with a card sees the same person weekly. An app can hold something the browser can't hold as well: a saved payment method, a membership state, a booking history, a points balance, a door code. The test to run is narrow. Name the one thing a customer would otherwise ring you or queue for, count how many times a month they'd do it in the app, and multiply by how many customers would install it. If that number doesn't reach a few thousand actions a month, the app has no way to pay back a yearly rebuild.

Staff apps are the case worth taking more seriously than customer apps. Job sheets, timesheets, site photos and stock counts run many times a day, the audience already works for you, and store discovery stops mattering because you tell them to install it. A home-screen web app usually covers it without either store getting involved.

The quotes you'll find online span fifty times over

Search for NZ app development costs and the results quote anything from about $4,000 for a simple single-platform build to $200,000-plus for something mid-complexity, all under the same label of "a small business app." The pages come from New Zealand and offshore agencies alike, and none of them publish the hours or the rates behind the figure. They exist to start a sales conversation. A fifty-fold spread gives you nothing to put in a budget.

Costs you can pin down without a quote:

Platform fees. US$99 a year to Apple, US$25 once to Google.
Two codebases or one. StatCounter's New Zealand data puts iOS at about 56% of mobile traffic and Android at about 44% as of July 2026. Neither platform is small enough to skip, so an app means shipping and testing on both, where a website means one build.
The annual compliance rebuild. Driven by the two deadlines above, and separate from any new features you want.

Ask for quotes broken into build, first-year support and annual compliance, priced as separate lines. If an agency won't split them, you're looking at a number rather than a plan. The same logic applies on the web side, and our breakdown of what a website costs in New Zealand shows the ranges there.

What to build first

Build the mobile website. Make it fast, make booking or quoting work with a thumb, and add it to the Home Screen path if you want notifications. Run it for a year and read the analytics: how often the same person comes back, and whether anyone asks you for an app unprompted. You get that data for nothing, and it answers the question better than any agency pitch.

Then, if the frequency test clears and the repeat behaviour is in your numbers, build the app for the one job it does better. Ordering, membership, loyalty, a customer portal. At Kage Works we build the web side first for this reason: it's the cheaper way to find out whether the app has a job. If you're still choosing a platform for that site, our comparison of Squarespace, Webflow and custom builds covers the trade-offs.

One last piece of small print, for whenever you do build it. If your app lets customers create an account, Apple's guideline 5.1.1(v) requires you to "also offer account deletion within the app" (Apple). Plenty of first submissions fail on that line alone.