My iPhone screen broke the other day. Fortunately I have opposing thumbs so I assembled the correct tools and sat in front of my computer closely following the steps of a YouTube video.

All was well until, just when my phone was in all its separate pieces and the dozens of microscopic screws were gently falling into the carpet, the man on the video deftly and invisibly tucked some tiny connector into some hidden slot and then continued on. But now the rest of the video was impossible to follow.

Two hours later I concluded that his advice had the equivalent information content of a large pat of the natural after product of bovine digestion. I would have phoned him and told him so except I would have needed about 14 hands to hold the pieces of my phone to my ear.

‘Management advice: Which 90% is crap?’ is the refreshingly frank title of article by Bob Sutton professor of management science at Stanford University. Sutton is also the author of ‘The No Asshole Rule’  and a strong advocate for evidence-based management. This is excellent because most management advice isn’t based on evidence at least not very good evidence or, like the aforementioned YouTube video, not presented very helpfully.

Management magazines constantly claim new breakthroughs. There are seven ways to… No eight ways, no six ways to… [insert desirable outcome here]. It looks inviting, it grabs attention, but ultimately leaves you unsatisfied. Its management porn. So Sutton lists some guidelines about how to interpret advice. Unfortunately they aren’t, in my humble opinion, very useful for the practitioner. So I have nicked the good bits and included them in my own approach.

Here are the three steps to take when you are ogling your next piece of management advice crumpet or deciding who to listen to when you’re trying to fix your iPhone.

1. Ho hum – does the advice even apply in your situation?

2. Says who? – check the quality of the evidence

3. Oh yeah? – have the right conclusions been drawn and presented helpfully?

It all boils down to this – the main underlying problem with management advice is the quality of the supporting evidence. The office situation doesn’t lend itself to the classic controlled two trial double blind experimental method. Instead lessons are drawn from anecdotes, or exceptions or examining groups of winners. For several reasons this can, and has produced, horribly flawed ‘insights’. But before we get into that lets look at the first step.

1. Ho hum? Does this advice even relate to your situation?

I spoke to someone recently who argued that the main takeaway for them from reading ‘Every Bastard Says No: The 42 Below Story’ was that they needed to add more spunk and personality to the marketing of their product.

Possibly. or possibly that advice only applies if you’re trying to break into a highly competitive well established vodka market with an essentially identical product, and edgy marketing is the only way you’re going to differentiate yourself. If you had a highly unique and original product then is that where you should be spending resources? Nope.

Let’s take another example, GE is a very successful company and it makes a point of firing a significant percentage of its management every year. Is it a policy that would really work well for you? Would that turn you into a successful company or scare the bejesus out of everyone?

Often advice is given with no little context about when it is useful. Yet the requirements of say a small software startup are completely different from say a plastics manufacturer trying to break into an export market.

2.    Says who? Check the quality of the evidence

Beware analogies and anecdotes! If anecdotal evidence was sufficient justification for correct action we would all be taking homeopathic medicines and making our millions by buying lotto tickets – because we all know of someone who has done that and it’s worked. I can understand a writer wanting to enliven his work by using a real-world example, but if close reading reveals that a single anecdote is the only evidence, then run away screaming.

Be cautious of ‘best practice’ studies. It is very common for a study to look at a bunch of winners and then see what they have in common. This inductive reasoning approach can produce interesting data, but it has some fundamental problems. Here are two.

They pick turkeys. In Tom Peters classic ‘In search of excellence’ he drew lessons from a number of highly successful well managed companies … like NCR, Wang and Atari. When did you last buy a Wang anything? Lots of management theorists loved Enron and were happy to draw lessons from it … until they didn’t.

Common practice doesn’t mean best practice or necessary practice. In the management classic ‘Good to great’ Jim Collins investigated companies that have been at the top of the game for the last 50 years, and looked what they had in common. This can be useful but it is also very dangerous as the traits you may discover may have nothing to do with their success. The very successful Winston Churchill, Ulysses Grant and Earnest Hemingway all had a problem with drink. Where is the popular best-seller ‘Drink your way to success’?

3.    Oh yeah? Drawing and presenting the right conclusions

Is it an apple-pieism? Politicians love Apple pie-isms. When they are campaigning they always say that they stand for high-quality education, free health care, a world where every child can fulfil their potential, a chicken in every garage, a car in every pot etc etc. There is nothing here that you could possibly disagree with, and it is all meaningless. We operate in a world of constraints. The pie of time and money can only be divvied up so many ways – but they never say that what they are going to take away so they can give you more of something else.

You know what you need to do in order to improve your business? Focus on cash flow, drive up sales, keep a lid on costs. Be nice to your mother. Employ the best people. Much management advice sits in this camp. Sounds good, says nothing.

Does it sound good but is impossible to execute? At the risk of upsetting the geopolitical balance here is advice about how to build an atomic bomb. Dig up some uranium ore. Enrich it in a centrifuge until you have a concentrate that is about 90% of the isotope Uranium 238. Place about 5 kilos at each end of a hollow pipe. Use explosives to bring them together rapidly. Easy peasy.

Here is advice about how to manage expenditure. Simply write down everything that you spend, recording every transaction faithfully. For six months.

Here is advice about how to lose weight. Stop eating processed carbohydrates, stop drinking alcohol. Reduced portion sizes by 30%. Exercise 30 min a day five times a week.

In all cases it sounds good but in practice it’s incredibly unhelpful because it is nearly impossible to implement. 

Can you draw different conclusions from the same evidence? Is cause shown – or just correlation? I read an article recently about a study that measured weight loss during a marathon, with the performance of the runner. It found that faster runners lost more weight than slower runners, the slowest of which actually gained weight during a marathon. The conclusion was that slower runners need to drink less. Really? Or is the takeaway that faster runners would go even faster if the drank more?

I find it useful whenever encountering management advice that sounds like an oversimplification, to ask ‘Is the opposite advice also true?’ Strangely often it is.

Does it acknowledge benefits and costs? Positive and negative impacts? Every piece of advice has benefits and costs. Seat belts actually make you drive faster (Don’t believe me? Drive along the motorway and unbuckle your seatbelt and see what you do). Suntan lotion causes cancer (you spend more time in the sun). Bicycle helmets cause head injuries. If you were to ban acts solely because of the harm that caused you would ban kissing – it spreads diseases.

It is very important to look at the net effect of an action. A great example of this was the grounding of planes because of terrorist concerns of some weeks after 9/11. It is estimated that 1200 lives were lost in additional car accidents.

My last piece of advice is to check whether it’s advice at all. Lots of advice is in fact cloaked marketing spin. If I had looked to see that the video on YouTube about fixing iPhone’s was put up by a company that happily charges for doing it for you (and taking over when you mess up) – then I might have saved myself a lot of trouble…

Comment