The State of Spreadsheets in 2026

Nearly everyone on earth uses a spreadsheet. The average office worker spends about half their week in one. And decades of research show that most real-world spreadsheets contain errors. Here's the data — sourced — and what's finally changing.

We still run the world on spreadsheets

More than 45 years after VisiCalc, the spreadsheet is still the most widely used analysis tool on the planet. Estimates put the number of Microsoft Excel users somewhere between 1.1 and 1.5 billion worldwide, and a widely-cited market study found 54% of all businesses globally use Excel. Google Sheets has added roughly another 1.1 billion users on top. Whatever the exact figure, the conclusion is the same: spreadsheets are how the modern economy actually does its sums.

And we spend a huge share of our working lives in them

A survey of 1,000 office workers by Acuity Training found the average respondent spends around 20 hours a week working in Excel — and that a spreadsheet is involved in about 38% of their total work. Roughly 1 in 25 office workers spend 80% or more of their working time in a spreadsheet. For finance, operations, and analytics teams, the spreadsheet isn't a tool they use; it's the room they live in.

But the data underneath is quietly broken

Here's the uncomfortable part. Decades of research by Professor Raymond Panko, summarising real-world audits, found that about 94% of operational spreadsheets examined contained at least one error, with an average cell error rate of around 5%. These aren't typos in someone's personal budget — they're the live models that price products, forecast cash, and inform board decisions. Human beings make small mistakes in roughly 2–5% of complex cognitive steps, and a spreadsheet faithfully preserves every one of them.

When a spreadsheet breaks, it can be catastrophic

The risk isn't theoretical. In October 2020, Public Health England lost 15,841 COVID-19 test results because case data was being collated in an old .xls file, which silently stops at roughly 65,000 rows. The overflow simply vanished — and as a result, an estimated 48,000 potentially infectious contacts went un-traced during a pandemic. No malice, no hacking: just a spreadsheet doing exactly what spreadsheets do when they hit a limit nobody checked.

What's changing in 2026

Three things are finally shifting the picture:

• AI moved into the grid. Plain-English questions over your own data — "which region is churning fastest?" — now return answers and charts without a single formula. The error surface shrinks because humans write fewer fragile cell references.
• No-code analysis went mainstream. A new class of tools lets you upload a spreadsheet and get a dashboard, forecast, or report in one click — keeping the file you already have, but moving the logic somewhere auditable and repeatable.
• Spreadsheets became a system of record, not a system of analysis. The smart pattern in 2026 is: keep your data in the sheet, do the analysis somewhere built for it.

The takeaway

Spreadsheets aren't going anywhere — nor should they. But "everyone uses them, all the time, and most contain errors" is no longer an acceptable status quo when the fix is finally easy. The teams pulling ahead this year are the ones that kept the spreadsheet as their source of truth and stopped doing their most important analysis inside it.

FAQs