James Percy is the founder and managing director of Verasseti, an IT firm focused on the development of software, websites and web and mobile apps. Having established the company in 2002, it grew organically and can now boast numerous FTSE 100 firms as clients. In the following piece, Percy discusses some of the shortcomings with Excel.
If you look at our online description of ourselves—and obviously, I hope you do, at least now and again!—we describe ourselves here at Verasseti as an award-winning software development company that’s all about building bespoke web applications to automate business processes.
And while that's very true, and brilliant fun most of the time, something that kind of drives me up the wall is that over and over again, the business process we’re called in to automate is saving a great company from Death by Excel.
This is all front of mind this week, because yet another project to do this has come in. It almost seems to be our bread and butter: replacing a complex, messy, nearly out-of-control Excel-based solution with a proper web one. And yet again, the Excel file we need to coax back from the world is responsible for managing millions and millions of pounds of sales for the company it’s lurking in; once again, someone’s let Excel (and I should immediately say I’m starting to get emails from concerned Google Sheets corporate owners) get used for something it’s just not meant for.
Yup… yet once more with feeling, I’m having to gently persuade business process owners that for all its amazing qualities, what they are doing with it isn't ‘always’ what it was built for, and they are running a big risk letting things get to this situation.
‘The computer for the rest of us’… maybe not the app for everything, though?
A lot of people can relate to this, I am sure. I love Excel—the spreadsheet is an indispensable business tool, has been for a generation at least. But it also just too, too easy to build more and more customised and arcane internal spreadsheets you end up depending on. It’s one thing to run your household finances and plan your wedding off one, and they are great for managing a small team or for a project… BUT, if you end up running the whole company on one spreadsheet, shall we just say that’s probably not best practice?
Let’s be honest: we all know why this happens. Business people have historically been suspicious of IT teams who take six months to do anything; they like to crack on and get stuff done, and the very original promise of things like VisiCalc and Lotus 1-2-3 was that this was business software anyone could (and should) use (check out this hilarious, but pretty on-point, Apple Mac TV ad from 1984 with a certain Mr W Gates III if you don’t believe me!).
And so they should continue to do so. But there is a line that is so easy to cross where what had been a fantastic tool for speeding up your business somehow becomes a bespoke bit of magic full of self-written formulae that no-one now knows how to change, even if they dared. By definition, that’s not safe for the business: if you are running any significant/large processes in your business on Excel, then there is a risk that Excel will go wrong, will get corrupted, all sorts of technical issues could throw it out of whack (I give you, in respectful silence, someone’s decision to run an allegedly £37 billion Track and Trace COVID strategy on an out-of-date version with a table size limit no-one knew about, for example).
There are plenty of other Excel horror stories like this in the public domain, but I have seen a lot of it in private; people that have been successful CEOs of medium-sized businesses who made one screw up on their one-version-of-the-truth cash flow spreadsheet and the business basically went out of business. There are other business problems here just waiting to snare you beyond machine or user error, of course; one of your team could take the customer spreadsheet and walk out the building with it over to a competitor; auditors would lose their guano if they knew you ran on Excel, as it's just not auditable. There’s also a HR problem waiting to bite you, too: a financial services company I know ended up running off one massive, complicated spreadsheet that only one very clever person understood—and that person suddenly woke up one day and realised he would never be able to get another job because they needed him so much.
And he didn't like that situation. Luckily, he was a friend of a friend, so we got asked to replace it with a proper business system. Now, let me say that there's a whole market out there of companies that will help you use Excel in a much more reliable and controlled way; we have such a partner company ourselves, and they’re great. If you’re wedded to Excel and you are convinced Excel is the right tool for the job, then that's great.
But quite often, Excel isn't the right tool for what you’re using it for, or you outgrow it. It's great for quickly prototyping ideas or managing a small thing, but if it ever gets anywhere near mission-critical, you have to think about all the kinds of security, operational risk and control issues we’ve talked about here.
‘Useful little systems that are small but perfectly formed’
So: what do you do about this? In some companies, there might be a beautiful relationship between IT and the business where there isn’t a shadow IT issue, and managers nicely step in and take spreadsheets off a line of business manager once it’s getting too critical and some proper accounting engine gets installed. (I’m not being sarcy; I am sure there are some places!) But IT is often very stretched, so often, this kind of gets left until some bit of budget suddenly becomes available and we (or another great software house like us, if there are any, ha ha!) get called in to what I call the ‘nice-to-have’ work we are so good at, like this Excel porting work.
That’s because a small and agile company like us can be very good at quickly building, in a cost-effective way, useful little systems that are small but perfectly formed, comply with all the standards and constructed as if their own IT team would, if they’d had the time. We also score well here as, after so many years of this kind of engagement, we have a process and a technology that allows us to turn a spreadsheet into a great, secure, but also easy to use online application in next to no time. We aren’t smuggling in our own product here, either; there's no ongoing support cost unless you want there to be, there's no commitment, no tie in: for a fixed fee, we will take your spreadsheet and we'll turn it into a web application, we'll document it, we'll evolve your IT department in as much, or as little as you want us to, hopefully with their blessing. If you want, sure, we'll host it and support it, as we have done for dozens of FTSE 100 companies, now, but the offer is to create something that you own as much as that internal spreadsheet, but with just a bit more of a guard-rail or two!
So, I’m happy to take on this latest Excel migration: it’s important work, and we know how to do it well. But I think the message to the entire spreadsheet industry here is, you've got to support business users better. How about a warning about too many bespoke variables in a formula, for instance: you’re pretty good at telling me my grammar’s wrong, after all?
I’ll leave you with that cheeky thought!
