<b>Technology News:</b> Spreadsheet error undermines government IT tenders. By John Stokdyk

Kashflow logo
Share this content

A new entry at number 88 in the catalogue of Excel infamy maintained by the European Spreadsheet Risks Interest Group (EuSprig) points to a spreadsheet error that undermined the UK government's IT procurement programme.

As originally reported by Microscope, the error cropped up in an Office Government Commerce spreadsheet used to monitor the accreditation of IT suppliers in its new Catalist procurement programme.

The Catalist agreements cover 11 contracts to supply hardware, software, networks and printers to public bodies such as the NHS, local authorities, police and charities. The assessment spreadsheet was maintained by the OGC's Buying Solutions team, which was forced to review all of the tenders it received.

After discovering the error, Buying Solutions wrote to affected suppliers, one...

Please Login or Register to read the full article

The full article is available to registered AccountingWEB.co.uk members only. To read the rest of this article you’ll need to login or register. Registration is FREE and allows you to view all content, ask questions, comment and much more.


Please login or register to join the discussion.

By Anonymous
21st Apr 2006 15:18

Agree with Dennis ....

Tend to agree with Dennis; although it really boils down to the right tool for the job. Over complexity & multiple nested IF statements (over 2 or 3 deep) etc. probably mean that one is using the wrong tool.

Excel seems to have become the 'de facto' standard for professional firms for a number of reasons - including power, minimal learning curve & the one solution fits all approach. Unfortunately aligned with 'minimal leaning curve' is a lack of underlying knowledge (understanding) and because of this the world is made up of instant IT experts

You are quite right in saying '..it is the users who screw up..' but this is surely the result of lack of knowledge or IT discipline and the ease with which Excel lets you get away with errors

No spreadheet documentation and the difficulty of incorporating an audit trail or even a 'code review' in Excel is a contributing factor as is the inevitable 'scope creep' when others in an organisation take on a useful speadsheet and modify it accordingly

Also having a customisable tool such as Excel means that users do not really need to address their actual requirements because anything can be modified on an 'ad hoc' basis. Quite frankly, there are hundreds of users re-inventing the wheel every day and whilst it is actually quite fun, the validity of this approach needs to be questioned

Dennis is right - the horror stories continue to emerge

Thanks (0)
21st Apr 2006 15:40

I am not necessarily disagreeing with Dennis! I too am somewhat cynical of how Excel is used, but I would question whether all of the errors that are reportedly ascribed to Excel are actually Excel errors. Often the blame properly lies with the users - they just choose to blame Excel.

What is also interesting is that the problems are not always a result of complex formula. I have seen examples where it was formula overwritten, or the wrong units used, or relatively simple spreadsheets where the logic was flawed, or even spreadsheet output rekeyed into an application incorrectly.

Equally valid of thought is that there are errors arising all the time where Excel plays no part at all. For example, no one can blame Barings on Excel!

Thanks (0)
By neileg
24th Apr 2006 11:59

Old fashioned?
I know I'm getting long in the tooth (big 50 this week!)but there are two simple techniques that help spot spreadsheet errors.

The first I learned at school when using slide rule or log table, which is to estimate the answer. You should understand the question well enough to be able to predict at least the order of magnitute of the result you are computing. If you don't have that understanding, or the spreadsheet doesn't tally with your expectations, you can't rely on the answers you get.

The second I learned as an student in the firm in which I trained, reconcilliation. The opening figure plus the movements should equal the closing figure.

Too elementary? Perhaps, but these two tecniques did allow me to spot a £1m error in a spreadsheet based financial model (not created by me!) that had been reviewed and approved by a number of other people. They also allow me to spot my own errors, before they are passed on to others.

Thanks (0)
21st Apr 2006 01:18

You really don't want me to...
rant about this but I feel I have to. why do we have a vibrant Excel channel when the evidence continues to mount that Excel is positvely dangerous?

Thanks (0)
21st Apr 2006 13:15

oh, come on Dennis!
there is nothing wrong with excel (it is probably the least buggy bit of software you can buy, oops, I probably mean licence) - in every case it is the users who screw up, and users are just as capable of screwing[***] up other apps. excel (and spreadsheets) get the publicity because they are used the most.

Thanks (0)