The lumpy carpet syndrome

When Alan Mulally became boss of an ailing Ford Motor Company in 2006 one of the first things he did was demand that his executives own up to their failures. He asked managers to colour-code their progress reports—ranging from green for good to red for trouble. At one early meeting he expressed astonishment at being confronted by a sea of green, even though the company had lost several billion dollars in the previous year. Ford’s recovery began only when he got his managers to admit that things weren’t entirely green. (Fail often, fail well)

The sea of green is the lumpy carpet syndrome. It is common and particularly true when donors are the source of revenue.

Over $750 million has been spent on elections in Afghanistan, over the past 12 years. Not one has been credible. Almost two thirds of the money has been spent on building the voters register. There still is no voters register in the country.

This figure barely registers in the waste the Office of the Inspector General has exposed.

And yet the reports from the offices responsible for these projects read of glowing success. The difference is the lumpy carpet. Anything that looks bad is hidden. After all, no-one wants to spoil their prospects for promotion.

If someone tries to expose the waste, then it’s time to shoot the messenger.

More at:
Fail often, fail well
SIGAR Quarterly Reports to Congress
If it’s broken, IT can’t fix it

Multi-tasking is a myth

The cell-phone was the start of the slide. Now with the ubiquitous iPhone and Blackberry, it’s become worse. 

Unless they have multiple processors, computers that do multi-tasking, do each task at a proportionately reduced speed. But, at least usually, each task is done properly. With humans, because they believe that they are able to multi-task, it’s not only the speed that’s affected, it’s the quality of the output.
Multi tasking
In meetings and conversations, people devote half (or less) of their attention, and with that comprehension disappears. So they agree to things they would not otherwise have done, and mistakes happen.

The e-mail responses one receives from people who write them while also otherwise engaged often reflect the same inattention. They reveal that the respondent, instead of reading the original missive, has often jumped to a conclusion of what they think the content is, and send a reply that is way off track.

Now as social media has entered the fray, attention has become a rare commodity, and otherwise intelligent people start to look more than a little stupid.

At least the phones are smart!

More at:
Cell-Phone–Induced Driver Distraction
Drivers on Cell Phones Are as Bad as Drunks
Hands-Free Talking, Texting are Unsafe
M is for Meetings and Multi-tasking
Stop Multitasking in My Meeting!

If it’s broken, IT can’t fix it

Schumpeter, the Economist’s business contributor, appears not to understand government. His recent article, Fixing common affairs, suggests why. The article implies that by harnessing the power of IT, government can become more efficient.

Government tries to harness IT’s benefits, they just do it badly.

The introduction of IT only mechanizes the processes, when there isn’t a simultaneous redesign of the existing controls. With mechanization the quality control sample is reduced to two (n = 2). By testing the first item in a batch, and the last item in the batch, when both samples meet the production standard, one is sure that the complete batch meets the production standard. But if the production standard is poor, mechanization won’t improve it.

When systems are not integrated, it requires human intervention to get controls to work. If a payment system requires that the person doing the authorization has had a specific training, integration with the training system should confirm that automatically. If the systems are not integrated, it requires that a second person checks the authority levels, and in the event that the person does not have the necessary authority, sends an email that the authorization is invalid until the course has been taken. Then, when the course has been taken, a second chain of emails needs to be sent, so that the payment can be made.

This goes to the heart of the difference between procedures and systems. Procedures simply provide evidence that something has been done. Systems ensure that what should have been done has. In a properly designed system the end event taking place proves that all the intermediate steps happened.

Sometimes reconciliation to external data is necessary to ensure that all the transactions have been correctly recorded. For example, in an accounting system the bank reconciliation checks that all transactions in the cash book appear in the bank statement, and all the transactions appearing on the bank statement have been recorded in the accounting system, and that any differences can be explained. Without the reconciliation there’s no guarantee that all transactions have been recorded.

Without these systems and complementary controls IT can’t change anything.

Sometimes IT makes things worse. Take the reply-all button. People who work in government will tell you that two days away from the office will result in an inbox with hundreds of emails. Fighting the tide becomes the dominant task on return to work. That’s hardly productive.

Many of governments inefficiencies can be also attributed to the complexity of the management structures and unbridled ambition.

There far to many layers of management and matrix structures, where people at lower levels are accountable to two or more bosses, are a norm. The inefficiencies that matrix structures create are the reason why business schools teach that they should only be used as a last resort.

The ambition that runs rampant in government creates a situation where bad news is concealed to avoid the possibility that it will taint a golden career. Schumpeter in his article “Fail often, fail well” describes how Alan Mulally, on taking up his tenure as Ford  CEO, discovered that although the company was in crises, executives were unable to admit to having any failures. Once managers learned to admit that things were going wrong, and to invite help, the company turned around, and was the one motor manufacturer that did not require rescuing in 2008 when all the others were bailed out.

Having seen some well written fiction in government reports makes this correspondent believe that the same medicine would go a long way towards making public institutions a lot more efficient.

Government must work smarter, and when that happens, IT will help.

More at:
Fixing common affairs
Fail often, fail well

Fail to succeed

South African Breweries (SAB) is the world’s second largest brewer. To many people that comes as a surprise. To me – not really.

In the 1980s SAB had already totally dominated the South African beer market. Their marketing strategy was focussed as though a significant competitor with unlimited resources was about to enter the market. The few that tried failed.

The other dominant management creed has always been: no surprises. If something went wrong, management at the most senior level had to be informed – they did not want to find out the bad news when the results had made it impossible to conceal.

As a young manager in one of the SAB subsidiaries at the time, that thinking was drummed into me.

So it was a surprise to read that Ford was the only American car company not to need bailing out in 2008 because when Alan Mullaly took control of the company in 2006 he’d required his managers to own up to their mistakes. He asked managers to colour-code their progress reports—ranging from green for good to red for trouble. At one early meeting he expressed astonishment at being confronted by a sea of green, even though the company had lost several billion dollars in the previous year. Ford’s recovery began only when he got his managers to admit that things weren’t entirely green.

It seems that not owning up to mistakes until they hit the financials is fairly common. Recent examples making the headlines are huge: Jérôme Kerviel – €4.9 billion ($7 billion) at Société Générale in 2008; Kweku Adoboli – $2.3 billion at UBS in 2011; JP Morgan – $2 billion earlier this year.

Sweeping mistakes under the carpet instead of owning up and learning has become common practice. With plenty of profits, covering the errors has been relatively easy. At JP Morgan, even with the $2 billion “mistake”, the company will still earn a $4billion profit this quarter. The headline in Forbes magazine “Less Hyperventilating on JP Morgan Chase’s $2B Loss, Please“.

Through the good times, over the past 20 years, fitting in and the EQ was all important. Managers, not challenged to overcome losses have been fitting in. Mediocrity has prevailed.

Military history provides a good example of the syndrome. Nelson, the great military strategists and considered a maverick was destined for obscurity before Napoleon became a threat to the British Empire. Once at war, Nelson took advantage of the opportunity, and understanding the strategic flaws of his superior officers would disobey orders to pursue actions that he realised would prevail. His successes made censure impossible, and led to his steady promotion.

Today, as profits come under pressure, it seems that misfits are in. Companies are looking for people that think outside the box.
Schumpeter In praise of misfits

It will be interesting to see if managers whose comfort zone is the box can work with people who don’t see a box.

More at:
Schumpeter Fail often, fail well
Less Hyperventilating on JP Morgan Chase’s $2B Loss, Please
UBS’s trading loss Swiss miss
Banking reform What’s $2 billion between friends?
Baring witness
J.P. Morgan’s trading mistakes A billion here, a billion there
J.P. Morgan’s woes cont. Damage control
SocGen’s rogue trader All his fault
Of course it’s right to ringfence rogue universals
Sarkozy v Jerome Kerviel
Strategies for Learning from Failure
UBS Ossie out