Interestingly the true cost of poor process is not in doing the process badly.  Clearly there is some cost. But often it is the knock-on effect of that poor process has on other areas of the company, downstream of the problem which is the real cost multiplier.


But this is rarely calculated. If you have poorly documented and followed processes, you won’t even be able to identify the potential costs downstream – let alone measure them. (Note processes must be documented AND followed to be valuable) So, like the iceberg, you only see a small part of the problem. Assuming an iceberg is a problem is looking at an iceberg from the Titanic’s rather than the penguin’s perspective.

So to illustrate my point, I received a letter a few days ago from HM Revenue & Customs, the UK equivalent of the IRS. A largish organisation with some 100,000 employees dotted all over the UK.

The letter explained in 7 very wordy paragraphs:
–    During my online Tax Return on their website I had checked the  box “Do you want any over/under payment of tax adjusted via my tax code”.  I had said ‘Yes’.
–    As I had replied after 1 December, that check box should not have been available.  But it was.
–    As I had checked the box, could I access their website and to establish the phone number of my local tax office.
–    Could I call the tax office to confirm that I wanted any over/under tax adjusted.  BUT WASN’T THAT WHY I CHECKED THE BOX IN THE FIRST PLACE?

So, with a heavy heart anticipating a long wait in a call centre queue for a non-valued added call, followed their instructions:
–    I hit their website and needed to enter some tax information to be able to find the local tax office phone number – rather than just list the offices by county or city.
–    I phoned the tax office and followed the automated menu system and reached a human being, who was in a call centre in Scotland.
–    The operator was polite but asked me to redial the number and select a menu option which was not the obvious one, to get to the local office.
–    I phoned the number again and was eventually connected to my local tax office
–    Again a polite operator took me through slightly more security questions than my bank or credit card company ever does
–    I was asked that confirm that I wanted any over/under tax adjusted.
–    She said that she would make a note of my request and send it through to the right person.
–    Finally, I mentioned this waste of everyone’s time to the operator who was equally exasperated and embarrassed about HMRC

So how much of that wild goose chase was Value Added for me or HMRC?  Answer NONE

How much made be or the operators feel good about HMRC and the use of tax payers money? Answer NONE

So one small process error (not testing the application correctly has spawned:
–    A badly letter written, proofed, printed and sent to 10,000+ of people
–    A spike in hits on website to get phone details
–    Calls to call centres and local offices
–    Back office process to log and action my confirmation

What were the alternatives?   In order of effort spent by me and by HMRC:

–    Have a robust test process on the original web-based application
Effort – 1 hour

Accept the error and do nothing, but possibly fix a system calculation internally
Effort – 4 hours

If so some reason they need additional confirmation, create an additional page on the HMRC website and reference the link in the letter and I can go to it
Effort – 8 hours (website), 4 hours (letter), £15,000 (print and post x 10,000 letters), 1,666 hours (process  the results assuming 10 mins per entry)

–    If creating a new webpage is too difficult, create a new website with a form filling app
Effort – 16 hours (website), 4 hours (letter), £15,000 (print and post x 10,000 letters), 1,666 hours (process  the results assuming 10 mins per entry)

–    If they really want me to phone, then set up a dedicated phone number with an automated system like credit card companies have for validating new cards
Effort – 24 hours (phone number system), 4 hours (letter), £15,000 (print and post x 10,000 letters), 1,666 hours (process  the results assuming 10 mins per entry)

–    If they REALLY want to drive me to phone a local office, then create a letter with the phone number I need to call in the letter and the correct options to select from the menu system eliminating effort on my part
Effort – 24 hours (phone number), 4 hours (letter), £15,000 (print and post x 10,000 letters), 833 hours (answering calls to call centre), 1,666 hours (process  the results assuming 10 mins per entry)

– What they actually made me do
Effort – 4 hours (letter), £15,000 (print and post x 10,000 letters), 1666 hours (answering twice as many calls to call centre), 1,666 hours (process  the results assuming 10 mins per entry)

Assuming a salary of £10,000 and 200 days worked per year the loaded cost is £55 per day.

So the original process cost of £6.88 vs the option HMRC selected which is £37,935

How many little process issues cause internal problems and then have a convoluted, wasteful and ill thought-through processes to correct them.

And the people best placed to identify the problems and recommend the best solutions are those at the sharp end, on the shop floor, at the grass roots. You know.  The ones who are the lowest paid people in the company.

UPDATE:  We believe that the number of letters sent out is closer to 1.5 million.  That makes the cost to the taxpayer £5,690,250


2 thoughts on “The business case for process efficiency. Choose £6.88 or £5,690,250 Not so hard really #bpm #roi

  1. Ian, great post and absolutely spot on. My observations:

    YES, a process that is not documented can’t be measured because it doesn’t exist. But the question is how one can measure the improvement of something from not existing to existing. Especially if it is utterly abstract. I just wrote a song and it is now much better than before I wrote it? Hmm … What can be measured is how bad the first process implementation was as it certainly isn’t what people did before.

    First: CLEARLY, it is obvious that processes in government are driven by legislation that even the government people don’t understand. I am sure that is not the way any free market business wants to operate. So process governance is normal to public servants …

    Second: ABSOLUTELY correct that processes are really about content – the letter that was sent out and the form you might need to fill (web or paper does not matter). So process flows alone are irrelevant and don’t improve the process if there is no content or it is not managed as integral part of it. In fact, if you manage the content and its state – PRESTO, you have a process. No flowchart needed and people understand it without ever looking at a diagram. (Yes, a flow can be used for simple one-at-a-time content state changes … let’s not argue).

    Third: YES, I could not agree more that the people who have the knowledge are the ones at the grass roots on the shop floor. I am not sure how more often I must say that we need to empower them by allowing them to work unrestricted from flow-diagrams, but helping them to focus on process goals. Especially if letters have to be sent out to a large number of people and many of those letters are ONE-TIME letters, making sure that they are designed and sent consistently is key. Creating a flow-diagram won’t help. Enabling collaboration on the letter design and sign-off by the legal people, supervisors and experts becomes essential and ensuring that the letter will be sent exactly as designed to the right people at the right time is the software trick needed. BPM doesn’t do that.

    Fourth: YES, the process does not end when the letter is sent. As it solicits a response it could actually be an email with a hyperlink to a webpage (as you suggest). But even that response has to be handled properly. As much as governments would like they will find it hard to treat people like number and shear across them with the same flow-diagram. In the end that doesn’t work. People are people. Call centers with automated answering scripts and automaton-operators don’t work either. Empower public servants and citizens to collaborate freely (bound by some rule constraints, obviously).

    We just disagree on one thing: that these things can be improved by designing BPM process flows. Therefore the ROI that you attribute as a potential for BPM use is not even anecdotal evidence. It is an assumption and no more. I propose that the ROI will be achieved by empowered people and not by encoded processes.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

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

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s