As wild and ridiculous as the news cycle has been this year, I hope you might still remember the recent story of some 15,841 positive Covid-19 tests going unreported for a number of days (if not, stick with me - I’ll give you the jist).

Initially it was blamed on a mysterious and vague “IT error” - obviously it wasn't quite that simple. Let’s take a closer look at what happened, and more importantly see what learnings we can take away from it.

It was down to a combination of human error and IT, that is a human using IT. It’s fundamentally an IT error but there is a bit of human error involved in this too.
Word Salad from Public Health England, via The Guardian

So… what happened?

The Covid-19 testing programme here in the UK came together quickly, with various public and private labs being enlisted to cope with the huge demand for sample testing. Public Health England is responsible for handling the data received from these individual labs, collating it and reporting the daily totals.

Some of these labs were providing their data in CSV format (Comma Separated Values - one for your pub quiz, there) - an incredibly versatile file type which uses commas to separate the individual data fields, it’s widely used and supported just about everywhere - in short, it’s an excellent way to provide the raw data.

But here's where it all went wrong...

PHE saw fit to do their data collation in Microsoft Excel.

Excel comes with file size limitations, specifically how many rows your sheet can have. The original XLS sheet format could handle up to 65,536 rows while the newer XLSX format increased that to 1,048,576. Try and import a file containing more rows than that, and anything over the limit is missed out.

Oh, did I mention this process is part of a system that cost a cool 12 billion? Yup. It also makes me wonder what other government systems rely on this kind of method, and what else is being missed?!

Captain Picard Facepalm ©Paramount

Wrong tool for the job

You’re probably thinking “This really doesn’t apply to my business” or “We don’t handle that kind of data” but fundamentally, the learning here goes beyond the nature of the data.

PHE's mistake here was selecting the wrong tool for the job. Excel, a spreadsheet program, was never intended to be used as a centralised database, there's no surprise that led to failure.

Making the most of data

Here at Eleven, we work closely with our clients to manage their data, and maximise its potential. That takes a number of forms:

  • Implementing best-fit systems for the business size and needs
  • Reworking and improving data to allow for detailed segmentation and targeting
  • Creating and managing effective campaigns

Think about your data and consider these questions:

  1. Where is your data is stored, and are those systems fit for purpose?
  2. How that data is managed and shared between those systems?
  3. Are you maximising the potential of that data for productivity / promotion?

If any of those are unclear, be sure to take the time to properly review before you have your own Excel moment!

Excel isn't all bad!

While it's clear Excel is most definitely not a database, it is very good at what it's built for - spreadsheets, and pop it in the hands of a 70-something Japanese fella and it can produce some pretty wonderful art!

Until next time!

- JB

Data concerns?

If you'd like to talk data and see how we could help, get in touch today.

Abstract lines in front of a night cityscape