Another spreadsheet WTF

At the daily WTF of course.

This doesn’t sound much like a WTF to me, its completely standard practice isn’t it?

Spend a stack of time/money/effort creating a report that everybody ignores?

Thats the whole reason why the ‘whoooo – spreadsheets have errors’ scare story breaks down in the commercial world. Everybody ignores them, or at least very very few decisions (important or otherwise) get made on the strength of only one single reporting/analysis spreadsheet. (Financial traders are excluded from that generalisation).

I would argue a great many decisions are made on gut feel and/or wide ranging, hard to specify research and information gathering, and the spreadsheet is used as rational justification. See El Reg’s savage destruction of M Gladwell (author of Blink) here.

Everywhere I have ever worked I have seen multiple period reports issued that have been nonsense and they have got away with it too. I think the self healing nature of organisation is often underestimated.

I’d be interested in any research that compares gut feel to error ridden spreadsheets for decision quality.

I’d also be interested in any research comparing what people do when gut feel != system result.

If you have research in these or other related areas then you should tart it up and submit to Eusprig, then  we can meet in Paris for a Pastis in July.

Who else has seen supposedly ‘big’ errors in spreadsheets in companies that survived just fine? Or more to the point which companies failed because of them?

cheers

simon

Advertisements

4 Responses to “Another spreadsheet WTF”

  1. Gordon Says:

    Hmmm, see this all the time. Half the billing our company produces is based on such flawed data collection and processing that a 14 year old doing Excel at school would be able to pick it to bits. It says a lot about most of our managers, and those of our clients, that these things pass unnoticed and unfixed for month after month.

    When I was green, I gamely tried pointing out some of the errors, but of course the errors mainly seem to be in our favour so was quietly discouraged from making the same mistake twice!

  2. Bob Phillips Says:

    I am sure some catastrophic decisions have been made from spreadsheet analyses Simon, not on the basis on a considered judgement based on a faulty modeul, but more on an instant decision based upon a flashy chart extracted from a simplified cut of a spreadsheet model – we all know how macho it is making instant decisions.

    I have no idea how big this problem was or the impact, but back in my corporate days when I managed a project, I had to fill in a capital depreciation spreadsheet (necessary for the Bak to reclaim the VAT). Capital expenditure was depreciated over five years (later changed to 3), so the depreciation each month is a simple 1/60th of the capital spend. You would have thought that any fool could build that model, but when I moved to central IT, the first one I got calculated the depreciation based upon that month’s outstanding capital amount, that is 1/60th of what was left after the previous month’s 1/6t0th had been subtracted and so on. When I pointed it out to accounts I was told it can’t be worng, they had been using it for ever (sic!).

    And don’t forget the Russia crisis in 1997 (I think) when a major bank chairman told the BoE that it was not over-exposed to Russia, and then later found that one trader was dealing off-balance sheet, accounting it on an Excel spreadsheet – they were actually hundreds of million at risk.

  3. alastair Says:

    Why pick on spreadsheets? You get faulty design and c**p data in “proper applications” as well! At least with a spreadsheet the discerning user stands a chance of spotting and accomodating errors.

  4. Dennis Wallentin Says:

    Garbish in Garbish out no matter what tools the data is presented with.

    Personally I’m quite —- with the general approach to always blame Excel.

    Data quality seems to be a sensitive case. I have a client, xxxxx, that refuse to discuss this subject and instead requires that I must solve it in Excel.

    Somehow this reflect how people think, Excel is a magic tool and if it cannot measure up then it’s something wrong with Excel.

    Kind regards,
    Dennis

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: