Excel: Workers cling to the software despite shift to AI

In the 1990s some computer games had a “boss key” that allowed staff to call up an Excel spreadsheet if they needed to look like they were working.

Now bosses might frown upon a worker caught labouring over a spreadsheet. Excel, owned by Microsoft, is 40-years-old. Among some tech leaders it’s seen as, at best, a blocker to smoother digital workflows and AI, at worst, an accident waiting to happen.

Excel is certainly ubiquitous in the business world. According to research by Acuity Training, two-thirds of office workers use Excel at least once every hour.

Excel’s persistence is partly down to the way it remains embedded in technology education, along with Word and PowerPoint, says Tom Wilkie, chief technology officer of data visualization firm Grafana.

“Excel is just a really good tool. If you want to look at a small dataset, try an idea, or make a quick chart for a presentation, there’s nothing better for quick and easy analysis,” he says.

The problem is that people and businesses fail to distinguish between data processing and data analysis and visualization, says Prof Mark Whitehorn, emeritus professor of analytics at Dundee University.

“There are all these small departments where data comes in, goes into a spreadsheet, is run through macros, and it spits out the other end,” Whitehorn says.

A macro can be thought of as a short cut. It automates a series of steps within the spreadsheet, so that those instructions can be done with one click – for example formatting the data in a particular style or making calculations.

Spreadsheets are often poorly documented and maintained, continues Whitehorn, “and the guy who wrote the macros has gone and the people in the department don’t know how to run them.”

More practically, he says, it means data within an organisation is not centrally controlled. This makes it hard to secure and move data around the organization, or to extract it for broader analysis and to fuel AI.

This can result in critically important operations relying on fragile spreadsheets.

Last year, it emerged that Health New Zealand used an Excel spreadsheet as its “primary data file” for managing and analysing its financial performance.

This made collection and consolidation of data difficult, led to discrepancies and errors, and made it hard to gain a real time overview.

In the UK, the recruitment process for anesthetists was plunged into chaos in 2023 by spreadsheet confusion, while the Afghan data scandal resulted from the sharing of an Excel spreadsheet.

But getting teams and individuals off Excel is a challenge.

“It’s hard for an external vendor or an external tool provider to just provide something for the organization that it could use across all those different teams,” explains Moutie Wali, director of digital transformation and planning at Canadian telecoms firm Telus.

He has overseen a drive to shift hundreds of staff members off Excel and onto a custom planning system.

The aim is to smooth data integration and management, increase automation, and to incorporate AI.

But people wanted to keep their existing Excel setups and simply download information from the new system, Wali explains. “I said absolutely not. You have to force it by not allowing the spreadsheet to coexist with your [new] applications.”

In defence of its software, a Microsoft spokesperson says: “Over four decades, Excel has evolved from a basic spreadsheet into a versatile platform used by everyone.

“It is more widely used today than ever before, with monthly usage growing consistently over the past six years, and remains the default tool for data analysis, modelling, and reporting across industries.”

It’s not just large organizations that can benefit from rethinking their reliance on Excel. Kate Corden operates a bike fitting business, Hackney Bike Fit, which means managing two data streams – personal information about a customer, such as height, weight, flexibility – and information about the bikes.

She is an expert user of Excel, from her days as a business development manager in the corporate world. But, she says, “It’s too easy to lose data. It’s easy for data to be altered.”

Corden switched to LinkSpace, originally designed as a case management tool, which can be adapted for complex workflows. “It’s just having a complete data management system where you’ve got everything, instead of having multiple excels, which is going to really help me as I grow.”

And the potential benefits of dropping Excel can extend beyond easier data management.

Julian Tanner, a PR executive in London, is also treasurer for a local charity. He switched the charity’s accounts from Excel to an online accounting package that extracts information from invoices.

The package’s built-in AI means it can produce customized reports at the touch of a button.

It also meant they could dispense with the services of a book keeper, saving over £6000 a year. “It was a big expense for a charity,” says Tanner, “Which you always try to avoid.”

For bigger enterprises, the benefits include smoother workflows and aggregated, standardized data that can be poured into AI or machine learning systems.

Apart from smoother data integration and faster planning cycles, Wali expects to save C$42m ($30m; £23m) a year by eliminating misaligned capital. Other teams in Telus are now considering following suit.

But that means users being prepared to relinquish some control – or at least the illusion of control.

“People will say ‘well I’m taking my data and I’m doing this,'” Whitehorn explains. But, “It isn’t your data, it’s the company’s data.”

Excising Excel completely may be unrealistic. So, Whitehorn suggests, the boss key may have to be redesigned in the future to cover up spreadsheet use.