When your business needs a web app instead of a website

Short answer first. A website exists so that people find you and leave an inquiry. A web app becomes necessary when the real work starts after that inquiry, and today that work is carried by spreadsheets, messaging apps and the memory of your managers. If money leaks out after a client has already agreed to work with you, a new website will not close that hole.

How a web app differs from a website

A website shows the same information to everyone. A web app processes the data of one specific person and stores the result of their actions. The difference becomes obvious the moment a user signs in and sees what nobody else sees: their orders, their balance under a contract, their documents, their price.

Both open in a browser and often live on the same domain. The difference is in what happens after someone presses a button. On a website the button sends an email to a manager, and every step after that is done by a human being. In an app the button changes the order status, recalculates the price under the contract, reserves stock and passes the data to your accounting system.

There is one more practical difference. After launch, a website mostly needs updates and small edits. An app changes together with the processes of the company, so its growth is budgeted right away.

Signs that your business needs a web app

Almost every request for an app comes with the same background. The company grew and the process inside stayed manual. Here are the signs that come up most often.

• The work lives in spreadsheets. Orders, stock, debts and schedules sit in files that people forward to each other. Somebody is always working with an old version, and nobody knows which copy is correct.
• One action repeats hundreds of times a month. A manager calculates the cost by the same formula every day and prepares identical quotes.
• Clients call to ask about status. Questions like “Where is my order?” and “When will it ship?” take up hours of the sales team’s day.
• Data is entered twice. A request arrives by email, a manager copies it into the accounting system by hand, then into a spreadsheet for a report.
• Several types of people take part in the process. A dealer, a retail customer, a manager, the warehouse, the accountant. Each needs their own slice of the same information.
• Every client has their own terms. Individual discounts, contract prices, deferred payments, personal quotas.

One sign from that list means nothing on its own. When several of them add up, it is worth calculating the cost of manual work and comparing it with an automation budget. That expense hides inside salaries and never appears in a report. How to price that hidden cost and pick the first process to automate is covered in our article on where to start automating routine work.

When an ordinary website is enough

Some companies that come to us asking for an app hear back that their task has a simpler solution. When the problem sits in the site structure and slow pages, a custom-built system will not solve it.

A corporate site or an online store covers the job in these conditions.

• The sale ends with a conversation, a meeting or a normal checkout in a cart.
• You have few enough clients that one manager keeps them all in mind.
• Your client has no need to see personal data, documents or individual terms on their own screen.
• Everything rests on a single accounting system, and nobody outside the office needs access to it.

In that case the money is better spent on structure, copy, speed and advertising. We compared a single-page and a multi-page site in our article on landing pages versus multi-page websites.

What a web app is built from

The estimate for an app is based on user scenarios, and the number of pages explains almost nothing. Four parts appear in nearly every project, and they set the budget.

User accounts and roles

A role answers the question of who sees what and who is allowed to do what. A dealer sees their own prices and shipments, a manager sees the clients assigned to them, the owner sees everything. The more roles there are, the more expensive development becomes, because every screen is tested separately for each of them.

Business logic

These are the rules by which the system calculates and makes decisions. How the price is set for a particular buyer, when a request moves to the next status, who gets notified about an overdue payment. The logic is written down before development starts, and an owner often sees the whole process for the first time there.

Integrations with what already runs

An app rarely lives on its own. It gets connected to your accounting system, payment providers, delivery carriers, email platforms and telephony. Every integration is a separate piece of work, and its cost depends on how willingly the other service hands over data. With some the exchange is set up in a few days, with others you write an intermediate layer first.

Admin panel and reports

The part that people remember last, although your employees use it every day. Search, filters, bulk actions, exports and a change log that shows who edited the data and when. Saving here at the start turns into constant requests to the developer over small things an employee could have done in a minute.

How much development costs and how long it takes

We calculate the exact figure after the processes are described, but here are the ballpark figures.

• Starter version, from $4,000, from 1.5 months. A working first version with the core features, to validate the idea with real users.
• Standard web app, from $8,000, from 3 months. A complete system with a user dashboard, roles and integrations with payment and third-party services.
• Business portal, from $12,000, from 4 months. A closed platform with several account types and deeper integrations with CRM, ERP and accounting systems.
• Complex platform, from $20,000, from 6 months. A system with payments, subscription plans, many users, an API and integrations for high-traffic projects.

Interface design is not included in these figures; we price the mockups separately.

For comparison, a corporate website costs from $1,000 to $3,500 and takes from two weeks to three months. The gap comes from the invisible share of the work in an app. Access rights checks, error handling, data consistency, the behavior of the system when dozens of people work in it at once. That invisible part holds your business up when the load grows. The scope of work and examples of our projects are on the web application development page.

Timelines are counted from an approved description of the processes. Without one, the first weeks go into writing it, and that is normal. The stages and what drives the schedule are covered in our article on how long website development takes.

Middle-ground options between a website and an app

Between an ordinary website and a full platform there are solutions that do the job for less money. Most clients who come to us asking for a system start there and grow into a large platform a year or two later.

• A user account added to your existing site. If you run a store on OpenCart or WooCommerce, a customer account is built on top of it. Order history, individual prices for wholesale buyers, repeat orders in two clicks, invoice downloads. The store stays where it is, and part of your managers’ manual work disappears. Work of this scope usually takes a few weeks.
• A separate module or integration. When one specific area is causing trouble, fix that first. Data exchange with your accounting system, automatic shipping calculation for national carriers, document generation, a bot that collects inquiries. We handle work of this kind under our custom web development service. A separate module or integration costs from $500 to $2,000, automation on Laravel from $3,000 to $8,000, and small tasks are billed at $25 per hour.

We covered a similar choice between an off-the-shelf product and a custom build in our article on ready-made versus custom online stores.

Where owners go wrong at the start

These mistakes repeat from project to project. Three show up almost every time.

Ordering everything at once

A wish list of thirty items turns the project into six months of development with no visible result in the middle. A working approach starts with the scenario that brings in money or saves the most hours, and leaves the rest for later. A starter version, which takes at least a month and a half to build, gives you live feedback from your staff, and part of the wish list loses its meaning after that. What goes into that version and what waits is covered in our article on what to include in an MVP.

Drawing screens before processes

A design made before the logic is described has to be redrawn as soon as it turns out that a supplier needs the same screen to look different. Processes and roles come first, the interface after.

Forgetting the people who will work inside the system

The owner orders the system, and the people who use it every day are the warehouse keeper, the manager and the accountant. If nobody asked them how they work today, after launch part of the sales team will keep their own spreadsheet running in parallel. We invite these people to the meetings where processes get described, and details surface there that exist in no document.

Before the first conversation with a contractor it helps to prepare four things.

• One complete order path, from the first client inquiry to payment and shipment.
• A list of the people and roles that take part in that path.
• A list of the programs and services the system has to exchange data with.
• Examples of the real documents and spreadsheets you keep records in today.

With that set in hand a contractor gives you an estimate in a few days instead of a month of questions. Which contractors are worth considering at all is covered in our article on how to choose a web agency. An app is a long collaboration, and a one-time discount at the start settles little if the team is gone twelve months later.

We have built integrations between websites and outside systems many times: credit services embedded in the cart of a children’s products store, order sync through BaseLinker for a cosmetics brand selling in Poland, stores connected to accounting systems. The data exchange takes the longest in these projects, so plan the budget around it.

Conclusions

The difference between a website and an app comes down to one question. Who works with your data, a person or a system? While forwarding an inquiry to a manager is enough, you need a website. Once the inquiry starts a chain of roles, statuses, prices and documents, manual work gradually becomes the most expensive line in the company budget.

• Three or more signs of manual control are a reason to price an app.
• The budget is set by roles, business logic and integrations, while the page count is secondary.
• The safer start is a first version that closes one profitable scenario.
• A user account on your current site or a single module often solves the task for less money.

If you are not sure how to approach your task, write to us. We will look at your processes and tell you whether you need web application development or everything is solved by improving the site you already run.