Web App Development in 2026: SPA, PWA or Server Pages, and What It Costs

In 2021 this post named the web app trends it expected to stay: single-page apps, progressive web apps, AMP pages, animation, the MEAN stack and chatbots. Five years on, our advice for a new web app is simpler. Build the logged-in screens as a single-page app in React and TypeScript. Render the public pages on the server so search engines and slow phones get them fast. Make it installable as a PWA if people will use it every day on their phones. Skip AMP, and plan for accessibility from the first screen.

The 2021 trends, five years later

2021 trend	What changed	What we would do now
Single-page app (SPA)	Still the normal shape for logged-in screens	SPA for the app itself, server-rendered pages for anything public
Progressive web app (PWA)	Since iOS 26, every website added to the Home Screen opens as a web app by default	A good first step before paying for store apps
AMP pages	Google’s current Search guidance asks for good Core Web Vitals, which any fast page can reach	Don’t build AMP for an app; hit the speed targets instead
Animation	Heavy animation can slow the response to taps and clicks	Keep it small and use it to show what changed
MEAN stack (MongoDB, Express, Angular, Node.js)	Relational databases lead: PostgreSQL is used by 57.9% of respondents, MongoDB by 19.0%	TypeScript, Node.js and PostgreSQL as the default
Chatbots	Language models can now answer from your own content, not only from a script	One clear job, with a person approving where a mistake costs money

The database numbers come from the Stack Overflow Developer Survey 2026: the share of respondents who did extensive work in each database over the past year. In the same survey, 41.5% of respondents to the web frameworks question did so in React.

What changed in 2025 and 2026

PWAs got easier on iPhone. Apple’s WebKit team announced with Safari 26 in September 2025 that on iOS 26 and iPadOS 26, every website added to the Home Screen opens as a web app. It no longer depends on a manifest file. For a founder, that makes “add to Home Screen” a real way to test a product on phones before any store review.

Accessibility became an ISO standard. On 21 October 2025 the W3C announced that WCAG 2.2 is now ISO/IEC 40500:2025, which lets more countries adopt it formally. Building to WCAG 2.2 from the first screen is easier than fixing a finished app.

React kept moving, with a neutral home. The React blog lists React 19.2 and React Compiler 1.0 in October 2025, the React Foundation under the Linux Foundation in February 2026, and React 19.3 in September 2026. The most used front-end library now has a home outside one company, which makes it a safer long-term pick.

Speed targets to write into the scope

Google’s Core Web Vitals page recommends that site owners reach good Core Web Vitals. The targets:

• Largest Contentful Paint within 2.5 seconds: the main content shows up fast.
• Interaction to Next Paint under 200 milliseconds: taps and clicks get a quick response.
• Cumulative Layout Shift under 0.1: the page doesn’t jump while it loads.

Put these in the acceptance criteria for public pages. Then speed becomes a check that passes or fails.

Pick the shape of your web app

Your product	Shape	Why
A tool people log into: SaaS, client portal, admin panel	SPA, with sign-in and marketing pages rendered on the server	Fast screens after login, findable public pages
Public listings plus accounts: marketplace, booking, catalogue	Server-rendered pages for listings, app screens for accounts	Each listing can rank in search
Something people open daily on their phone	PWA	Installed from the browser, with no store review
Needs in-app purchases, heavy camera use or a store listing	A store app; see mobile app development	Some features and sales channels need the stores
An internal tool for a few staff	SPA, or a ready-made tool	Search doesn’t matter; cost does

What we have built

Our web app development page counts 30 published case studies that include a web app. Two show the range:

• IPOPBOOKS is a bookkeeping platform with one portal for business owners, bookkeepers and admins, each seeing only what they need. The first working demo came 3 weeks in. It launched 10 weeks after kickoff, with a 733-case QA checklist and a live card payment and refund tested first, on the client’s own AWS account.
• Daisy’s Guide started as a printed guide for couples. We turned it into a web app they add to their phone’s home screen. It reached beta with real couples about 2 months after kickoff. The client owns plain Flutter code that can become iPhone and Android apps later.

What a web app costs

A first version typically costs $5,000 to $15,000 with us and takes 2 to 4 weeks. Bigger platforms take longer; IPOPBOOKS took 10 weeks. The price depends on how many user roles and screens the app has, whether it takes payments and which systems it connects to.

For a market view: across 95 agencies, the median published MVP starting price is $13,234, and the median timeline agencies promise for their own MVPs is 6 to 12 weeks (Rocketech MVP Price Index 2026, H2 edition).

How we would estimate it

1. Our project estimate turns your brief into user stories. You confirm what is in scope, and every hour in the budget traces back to a story you kept.
2. We check the hours the model suggests against what our team actually books on finished projects, then multiply by our published rate of $39-55 an hour.
3. You get an hours range, a budget range and a number of weeks. It is a planning range, not a fixed quote.

Here is an example run of our estimate’s math with sample inputs. For a B2B web app with teams, roles and subscription billing, it came to a budget of $6,600 to $13,600 and 5 weeks. Every input in that run was an assumption; only the arithmetic is ours. To see the numbers for your own idea, get a free project estimate.

What we would do

1. Decide which pages must be found on Google. Those get server rendering; the rest can be app screens.
2. List the user roles and what each one may see and do.
3. Start in the browser. Add PWA install if people will open it daily on their phones.
4. Write the three speed targets into the acceptance criteria for public pages.
5. Write WCAG 2.2 checks into the acceptance criteria for every screen.
6. Launch on a cloud account in your company’s name.
7. Before launch, test the payment path with a real card and a refund.

FAQ

Is a single-page app still a good choice in 2026?

Yes, for the screens people use after they log in. Render public pages, such as the home page and listings, on the server so search engines and slow phones get them fast.

Should I build a PWA or a mobile app first?

Start with a PWA when people will use the product in the browser or daily on their phone and you don’t need in-app purchases. Since iOS 26, every website added to the Home Screen opens as a web app by default.

Is AMP still worth it for a web app?

Not for a new app. Google’s current guidance for Search recommends good Core Web Vitals: main content within 2.5 seconds, a response to taps under 200 milliseconds and a layout shift under 0.1.

How much does a web app cost in 2026?

A first version typically costs $5,000 to $15,000 with us and takes 2 to 4 weeks. Across 95 agencies, the median published MVP starting price is $13,234.