Excel to Web App: When to Automate or Replace a Spreadsheet

Many spreadsheets should stay spreadsheets. Turning an Excel file into a web app pays off when several people depend on the same data. It also pays off when the rules matter (who may change what, and a record of who did) and when the work has to connect to other systems. If none of that applies, tidying the workbook, automating its weekly chores or moving to an off-the-shelf tool costs less and takes less time. We build software for a living and would still tell you that first.

When it is time to move, our bespoke software service has two published prices: Discovery & design from £4,000, then a Bespoke build from £15,000. Both are agreed at a fixed price before work starts. This article covers the warning signs, four routes from Excel to a web app (keep, automate, buy, build), published tool prices, how a conversion works, and how to choose.

Key takeaways

• Many spreadsheets do not need replacing. Automating the weekly report, or moving a simple tracker to an off-the-shelf tool, solves many problems for much less.
• Move to a bespoke web app when several people edit the same data, you need roles and a history of changes, or the work has to connect to other systems.
• Low-code builders publish per-user prices and plan limits, so test your real data against those limits before you commit.
• Cost alone rarely justifies a build. The case rests on what the spreadsheet can no longer do safely.
• With us the bespoke route starts with Discovery & design from £4,000, followed by a build from £15,000.

Should your spreadsheet become a web app?

Only when the spreadsheet has become a system rather than a document. A document is read and occasionally edited by one or two people. A system is something the business stops working without. The four routes below run from the cheapest to the most involved, and you should take the earliest one that fixes your actual problem.

Route	Cost	Best when
Keep Excel and tidy it	Your time	One or two users, simple sums
Automate the chores	Your time, or a small project	Reports and re-keying hurt
Off-the-shelf or low-code app	Per-user subscription	A simple tracker, small team
Bespoke web app	Ours: from £4,000, then from £15,000	Roles, audit trail, integrations

The sections below take each route in turn. The warning signs come first, because they tell you which row you are in.

Is there a free Excel web app?

Yes, for the spreadsheet itself. Microsoft offers free access to Excel in the browser through Microsoft 365 for the web. Its page says files must be saved in and shared from OneDrive for real-time collaboration. You get the same kind of workbook in a browser tab. It does not become an application, so the single-file, single-owner problems described below stay where they are.

You will also find converter products that publish a workbook as a web form or calculator. They suit one calculator with one author. Before you rely on one, check that its results match your original formulas on real inputs, and ask where it stores the data.

If the sheet holds personal data, the ICO’s data security guide explains that UK GDPR requires personal data to be processed with appropriate security, using appropriate technical or organisational measures. That is not legal advice, but it is a good reason to know exactly where your data lives.

Warning signs your spreadsheet runs the business

If losing the file, or the person who maintains it, would stop work, the spreadsheet is already a business system. These are the usual signs:

• Version chaos. Several copies circulate by email, and nobody can say which one is current.
• A single owner. One person understands the formulas, so every change waits for them.
• Copy-paste reporting. Someone rebuilds the same report each week from other exports.
• Overwrites and over-sharing. Two people edit at once, or everyone with the file sees columns that only some should.
• No history. You cannot say who changed a price or a status, or when.
• Re-keying. The same figures are retyped into accounting, a CRM or a supplier portal.

Size is rarely the first limit. A worksheet in Excel holds up to 1,048,576 rows by 16,384 columns, according to Microsoft, so the limits that bite first are usually about people and process.

Spreadsheet risk is also documented in research. A 2001 paper by Chadwick and Sue in the European Spreadsheet Risks Interest Group (EuSpRIG) proceedings starts from the high incidence of errors in spreadsheet models used in industry. Every extra person who edits the file, and every formula only one person understands, adds to that exposure.

Our rule of thumb is simple. With one or two signs, tidy the file and agree an owner. With three or more, compare the routes below.

Exit one: automate the weekly report and data entry

If the pain is the repetitive work around the spreadsheet, keep the spreadsheet and automate the work. This is the cheapest fix that still removes hours, and it keeps everyone in a tool they already know.

Excel can do part of this itself. Its Power Query feature can import data from a wide variety of sources, and web pages are among them. That answers the common question of how to make Excel pull data from a website. For repeated clean-up steps, Office Scripts let you record your actions and replay them. Microsoft’s documentation currently says to use Power Automate to run them on a schedule.

When the job is bigger than a script, such as a report assembled from several systems, the problem is usually the data underneath rather than the spreadsheet. A data audit from £3,000 shows where your data stands. The warehouse and reporting tier, from £5,000, takes it from source systems to dashboards.

If people re-key figures between tools all day, AI and process automation is the relevant service. Our guide to where automation pays off helps you judge which tasks to start with.

Automation treats the chores. The workbook is still one file with one owner, so if the other warning signs apply, read on.

Exit two: off-the-shelf and low-code app builders

For a simple tracker, form or approval step used by a small team, an off-the-shelf product is often the right answer, and we would say so. At the time of writing (October 2026), the vendors’ own pricing pages show the following.

Tool and plan	Published price	Worth checking
Airtable Team	US$20 per seat per month, billed annually	50,000 records per base
AppSheet Starter and Core	US$5 and US$10 per user per month	2,500 rows per database
Power Apps Premium	£15.40 per user per month, paid yearly	Excludes VAT

Airtable prices the Team plan per seat in US dollars and lists 50,000 records per base. Its FAQ says that once you reach a record or attachment limit, you cannot add more until you upgrade to a new plan. AppSheet’s page prices both plans in US dollars and lists 2,500 rows per database on each. Microsoft prices Power Apps Premium in pounds and states that the price does not include VAT. Other builders, such as Glide, sit in the same category.

Here is an illustrative sum. Ten people on Power Apps Premium cost £15.40 × 10 × 12 = £1,848 a year, excluding VAT. That is the licence line only, without the time to build the app. A £15,000 build equals about eight years of that licence, so price alone seldom justifies bespoke software. The case for building rests on the ceilings below.

Test these in a free trial, using your real data:

• Row and record limits. Compare the plan’s limit with your data today and in three years.
• Per-seat pricing. Multiply the monthly price by your headcount, including occasional users.
• Permissions. Check whether you can control who sees which records, not only which screens.
• Calculations. Rebuild your hardest formula and compare the result with the spreadsheet.
• Connections. Confirm the builder supports the systems you need to connect.
• Vendor changes. Ask what happens to your app if the vendor changes its plans or prices.

If your list is short and none of these apply, stay with the builder.

Exit three: a bespoke web app with a real database

Build when the problems are structural. That means several roles, a need for an audit trail, workflows that span teams, integrations, or rules you must own. A bespoke web app replaces the file with a database and puts the logic in code that is tested, versioned and owned by you. This is the service we describe on our bespoke software page.

In practice, the move changes five things:

• A database, not a file. Records live in tables with validation at entry, so a date cannot be typed as free text.
• Logins and roles. Each person sees and edits only what their role allows.
• A history. Every change is recorded with who made it and when.
• Integrations. The app exchanges data with accounting, a CRM and other systems through their APIs instead of re-keying.
• Reports without copy-paste. Dashboards read the live data.

At the large end, our claims platform work replaced one client’s legacy case management, shared inboxes, spreadsheets and supplier portals with one platform, including full audit trails. A single business-critical workbook is a smaller job. The ideas are the same, with one database, defined roles and a record of every change.

Our terms are the same whatever the size. The scope and price are fixed and agreed before work starts. You pay by milestone as each stage is delivered, and you can cancel within the first 21 days and get your money back in full. All code and IP are yours once the work is delivered and paid for. Ongoing maintenance and support is from £400 a month.

For what moves a software price up or down, see our guide to software development cost in the UK. If your spreadsheet is mainly a list of customers and deals, our guide to bespoke CRM covers when to extend an existing CRM and when to build.

How to convert an Excel spreadsheet into an app

Converting a spreadsheet is mostly analysis, with the build coming second. The steps are the same whether you choose a builder or commission software.

1. Map what the sheet does. List the inputs (who types what), the calculations, the outputs (reports, quotes, exports) and every person who touches it.
2. Separate data from rules. Records become database tables. Formulas become rules written in plain English first, then in code.
3. Decide roles and history. Agree who may view, edit and approve each item, and what must be logged.
4. Rebuild and test against the sheet. Feed the same inputs to both and compare the outputs until they match. Rules that people apply from memory usually surface here.
5. Migrate and clean the data. Fix duplicates, mixed date formats and free-text values before import.
6. Run in parallel, then switch. Keep the old file read-only until the app has run a full cycle, such as a month-end.

The first three steps are the work a scoping stage covers. Ours is Discovery & design, from £4,000, which produces a scope and blueprint.

Two common spreadsheets have their own routes. If yours is a quote or price calculator, quoting automation starts from £5,000. We also built ChatQuote, an AI quoting assistant for service businesses that drafts a quote from a website enquiry for the team to review. If yours is really a database feeding reports, start with the data audit described above.

Your next step

Count the warning signs, pick the earliest route that fixes them, and test it on one real workflow. If you land on the bespoke row, start with Discovery & design from £4,000. It is a scoped, fixed-price look at your spreadsheet that ends in a blueprint before you commit to the build from £15,000.

If you land on a different row, that route is often cheaper than you expect, and we would rather you start there.