“I have had nine fares in 12 hours. My total profit for the day is $128 – about $10 per hour. After GST and income tax, it is much less.”

…

“Two girls tottering on high heels approach the cab. One pours her semi-conscious friend into the front seat then smiles at me: ”Don’t worry, she’s not going to be sick in your cab. You’re not going to be sick in his cab, are you? Nah, she’s not going to be sick in your cab.” The friend hands me $20 and tells me an address in Richmond. I take the girl home and make sure she reaches her door.”

…

“Only after I drive past a pregnant lady and see her middle finger raised in my rear-vision mirror do I realise I’ve left my dome light on.  I am not the driver I set out to be. This taxi deserves better.”

…

“In my three days and two nights on the job, I worked for 59 hours. I drove 1006 kilometres (477 of which were paid). I did 50 jobs in total, about 10 a day. Two-thirds of my customers sat in the back seat, and most said little more than ”Hello” and ”How much?”

After splitting the metered fare 50/50 with the operator, my take-home pay for the week was $618. Roughly $124 in my pocket every day. After paying income tax and GST, I was clearing roughly $8 an hour. This is a little below average, but not much.”

Konrad Marshall. Photo Credit: Wayne Taylor

You can read the whole article here.

A powerful example of Genchi Gembutsu, going to the place of a problem to learn the truth for yourself; Or as Nick Sayer put it at MDX13 on Tuesday – getting dirty.  Many thanks to Nick Sayer from Refletca for pointing me to this insight earlier this week.

Updates:

Is this the most expensive music video ever ??? Of course Randal (IKCD – What If ?) has the answer for us.

Also – Commander Hadfield’s son tweeted this photo of him packed to head home… The Soyuz which Nigel and I love dearly is certainly not a spacious space craft, but it’s been getting the job done for over 50 years.  I do believe those bags strapped between their heads are 6 months worth of wet waste and garbage.

Hadfield and his crew have now touched down and are enjoying the perks of Earth like long hot showers and spring air.

I love the internet so much.

One of the fine young developers at REA-Group (Luke Chadwick or @vertis to the twitterati) is coincidentally launching a new iPhone app for personal training. Time for an unashamed blog plug!

We’re delighted to have sparked a conversation about the state of IT training and the possibility that the profession of personal trainer had become more prominent in Australian tertiary institutions than much-needed technologists. I wrote about it after hearing numbers put forward by a fellow industry CIO in Melbourne.

Whilst not the most humorous contribution (unpublishable as it happens), doubtless the most erudite thus far has come The Register, who appear to have cleared a few things up with a bit of research.

They don’t fancy the 3% statistic (of total Australian students that are enrolled in IT courses) uttered by The Australian Computer Society as being reputable, nor the Minister for Tertiary Education’s numbers on how many students are at Uni (I guess both parties have particular agendas to push…); and offer some good alternate sources and calculations; as well as filling in the gaps in the data for students enrolled in TAFE type learning.

If you followed our article, you’ll enjoy Simon’s -  I did, even if I have had to endure James calling me ‘Dalton’ all this week in deference to my status as the Luna Tractor dodgy economist of the week.

You can head over there for a full explanation of black swans and the white lies that result from their discovery. You can also learn why I have put a 300 year old painting of Perth on our blog!

In a nutshell, the uncertain and sometimes brutal business conditions we face today demand a new approach, and we are lucky enough to have Japan’s vast lean industry experiment from the 1960s onward to learn from, as well as the pioneering work done by agile IT professionals since the 1990s.

However <sigh>, as James often quotes from Deming – ‘It is not necessary to change – survival is not mandatory’.

“We’re bringing home enough major components to fashion displays of two flown F-1 engines. The upcoming restoration will stabilize the hardware and prevent further corrosion. We want the hardware to tell its true story, including its 5,000 mile per hour re-entry and subsequent impact with the ocean surface. We’re excited to get this hardware on display where just maybe it will inspire something amazing.”

Even though these missions were finished in the early 70s we’re still going to be able to look back and learn new things.  As a new wave of modern day space explorers start building new rockets, and we figure out difficult problems like how to fly to Mars it’s cool to see a project to learn from the past.  And so Deming’s cycle continues – Plan – Do – Consider and Act.  The Bezos team is providing great fodder for teams to consider from the last time we did this, act and plan then do.

(images via http://www.bezosexpeditions.com)

In true Luna Tractor fashion, and as the economist in residence, I felt the need to dig a little deeper.

The truth appears more like ‘Australia is training almost as many personal trainers as IT professionals‘ – around 14,000 ICT undergraduates in 2012; versus 10,000 studying personal training at private and public tertiary institutions in 2012.

The scarier statistic is that perhaps only 1/3 of those ICT undergraduates are coming into the workforce each year (3 year degrees on average); whereas MOST of those personal trainers (who do 1 year of study) are hitting the local gyms of Australia.

That means Australian employers get access to maybe 4500 ICT grads every year, whereas the Aussie fitness industry is over-run by 8,163 fresh grads every year.

There are twice as many personal trainers as ICT graduates entering the Australian workforce every year!

I won’t pass judgement on the state of Australia’s graduate developers versus passionate industry-trained coders (apparently we have about half-half, and I love them both equally); nor on the benefits that personal trainers bring our nation’s health. But something is NQR.

And for the record, IT is a way better job. Just ask this guy. What’s more, he’ll hack an award winning app with his team, and then school you on personal fitness at lunchtime for free.

My workings (for the inevitable statistical pseuds to comb through) are as follows. Note, I did not obtain the original ACS report on graduate numbers from 2012, as I am not a member. I relied on their press release.

a. How many Personal Fitness Students were there in Australia in 2012?

It’s a big and complicated industry – but there is this useful 2012 industry report by Deloitte Economics to consider, and in the executive summary (p3):

‘On the supply side, in 2011 the headcount for registered exercise professionals was 29,875 (24,875 registered with Fitness Australia and 5,000 registered with Physical Activity Australia), with average annual growth rate between 2005 and 2010 around 7.2% (Job Outlook, 2011). Approximately 56% of fitness professionals are female (16,749), and 44% male (13,126).’

This amounts to about 11,000 FTEs working in the industry (the full-time number diluted by so many part-timers). But how many are studying?

 A. Personal Fitness graduates from public RTOs in 2012: 2,768 graduates (with 65% completion) see p36 of the Deloitte report.

Suggests 4,258 total students studying – assuming it is a one year course, as the report concludes from research that there is an average time to completion of 1 year (page 35).

 B. Graduates in personal training from private RTOs in 2012: 5,395, with 90% completion (see p37)

Suggests a total of 5,995 students studying personal training in private RTOs every year.

The report was obtained here via this reference on Wikipedia.

Thus, in 2012, about 10,253 students were enrolled in Australian tertiary educational institutions studying personal fitness.

How many ITC students studying in Australia in 2012?

The Australian Computer Society suggest less than 3% of Australian tertiary undergraduates are enrolled in ICT courses in 2012. Half the number a decade ago!

The interweb handily reports on total undergraduate numbers in Australia, thanks to this emission from the government in 2011 = 480,000, so…

480,000 x 3% = 14,400 undergraduates studying ICT in Australia in 2012, across all the years of their courses. This of course ignores graduate study, which I am taking the economist’s stance on (ceteris paribus), and calling them relatively irrelevant for both professions (Deloitte certainly confirm that for Personal Trainers).

By the time this is published I will have presented the results of many hours research, debate and reflection on the design of modern IT organisations. Sadly, without actions and interpretive dance, the Powerpoint slides on their own don’t add up to much more than pictures. Invite me (or James!) for lunch sometime, and we’ll happily proffer an opinion on the subject.

The pivotal moment in the thinking process came when reading the new book from Joyce Orsini, a deftly edited collection of Deming’s lectures, missives and thoughts from 1950 to 1992. A brilliant book, it is the closest you’ll get to Deming sitting with you and giving his opinion on a wide range of important matters. Including, organisation structures!

Deming’s simple idea (quoted in a 1992 presentation to General Motors) was to avoid traditional organisation charts in the form of hierarchical pyramids, and replace them with flow diagrams (aka value stream maps), and just put the people on the flow diagram as value was pulled by a customer. So simple!

“A flow diagram is actually an organisation chart. It shows people what their jobs are. How they should interact with one another as part of a system. Anybody can see from a flow chart what their job is. Take the chart, put the names on it. You belong here. Somebody else belongs here. Then anybody can see from the chart what their job is. And their work fits in with the work of others in the system.”

Compare that to the Hitlerian view of a flat organisation (so inexplicably popular since the 1990s, with ever-expanding numbers of direct reports), with this lightly edited paragraph from Wikipedia on the organisation of the 3rd Reich. The grey bits are the changed words. If this sounds like your IT department, run!

‘The CIO often deferred making decisions, avoided clear delegation and allowed subordinates to compete with one another, especially in the recent years. Therefore, a system of governance was formed whereby leading company officials were forced to interpret the CIO’s speeches, remarks and writings on company policies and turn them into programs and strategy.

Any manager could take one of the CIO’s comments, and turn it into a new strategy, of which the CIO would casually either approve or disapprove when he finally heard about it. This became known as “working towards the the CIO“, as the executive was not a coordinated, co-operating body, but a collection of individuals each trying to gain more power and influence over the CIO. This often made IT executive meetings very convoluted and divided, especially with the CIO’s vague policy of creating a multitude of often very similar posts.’

This is also an opportunity to put the many references given in the 30 minute talk, and used in the research, in one handy place. Enjoy.

Reading List

1. The Management Century by Walter Kiechel III, published in Harvard Business Review, November 2012.
2. The Essential Deming: Leadership Principles from the Father of Quality (2012) by Dr Joyce Nilsson Orsini. Available as a Kindle book, this is the only place I have read Deming’s theories on organisation structures and the negative impact of org charts.
3. Value Stream Mapping – understand the theory of this special variant of process map
4. Management 3.0 (2011) by Jurgen Appelo.
5. * Godwin’s Law by Mike Godwin, 1990. ‘As an online discussion grows longer, the probability of a comparison involving Nazis or Hitler approaches 1.’
   
6. Scaling Agile at Spotify (2012) by Kniberg and Ivarsson.
7. Power to the Edge (2003) by Alberts and Hayes.
8. Here Comes Everybody (2009) by Clay Shirky.
9. The Principles of Scientific Management (1911) by Frederick Winslow Taylor.
10. The Theory of Business Enterprise (1904) by the Thorsten Veblen (the witty economist who invented the concept of ‘conspicuous consumption’ among other things). Best read in Wikipedia.
11. General and Industrial Administration (1916 in French, 1946 in English) by Henri Fayol. Had Henri not been French, and writing at a tricky time in world politics, his ideas might have spread sooner. Similar to Frederick Taylor in many ways.
12. Conway’s Law by Melvin Conway.
13. Servant Leadership – best read about in this chaotic Wikipedia entry which features American Robert Greenleaf’s work.
14. Peter Drucker’s contribution to management and organisational literature in the second half of the 20th century was biblical. The HBR article above does a great job at summarising his influence, or you can buy this book on Amazon.
15. The reference to the 3rd Reich organisation structure and model can be found here in its original form (not adapted for CIOs).

The trials found that improvements in behaviour, and productivity changes observed after adjusting a workplace, were in fact largely just the results of a placebo-type effect – any change seemed to cause improvement.

There were many changes trialled – the first (and possibly most important) were changes in lighting levels (‘the illumination experiment’) in a factory environment. The objective of the experiment was to find the optimum lighting level for productivity – the results showed that more was at work than just changing lumens, every change (up or down in brightness) caused an increase in performance.

Most theorists since have concluded that the improvements are caused by the act of measuring and engaging subjects.  If you missed studying this in your management 101 course, then read this article or take the slightly incomplete wikipedia catch up lesson and then continue reading.  Lets deconstruct the impact of this theory into its parts:

Measurement

We’ve already written about the impact of measurement before.  It’s also pretty well understood that humans perform against whatever KPI you give them, the way they do it may however be surprising or unintended.

Confirmation Bias

Then there is Confirmation Bias, basically the likelihood that you will both find what you’re looking for, and ignore stuff that you’re not.  Elegantly illustrated by this group of Radiologists who can’t find a Gorilla in their scans.  If a team expects a new process to be better, then their perception will likely match their expectation.

Empowered Staff

Finally the staff in the later trials, more focused on workplace layout and reward were engaged, they were part of the process.  They were made to feel special.  We’ve talked about this a fair bit too – though Dan Pink is more fun. Smart people are motivated by Autonomy, Mastery and Purpose. Pink was not the first to the conclusion that money was directly linked to productivity.

In summary, if you measure stuff, look for a change, and then make people feel special – surprise, surprise, you’re going to see a change!  Does this mean that some of the impact of Agile, the dramatic increase in staff engagement and useful productivity are perhaps just caused by the visual changes in the offices and agile rituals, but, in fact are just a ‘Hawthorne Effect’ ? In my heart, I have to say nervously, yes, at least to some extent.

Is this bad ? No.

Ok, but is the change real and sustainable? Yes, the changes and improvement in your team’s performance and culture are real.  Like all placebos, the source of the change might not be real, but the impact is.  If we can come up with better or different ways to achieve the same improvement then we should try them… any good agilista will tell you that anyway.

In the mean time I’ll leave you thinking about a sobering equation Nigel sent me as we’ve been debating this recently.

Hawthorne > Maslow + Deming

The Dabbawal’s Service Offering

Your own lunch, delivered from home to your office and then the container (tiffin) returned all withing the 6 hour lull between commuter rush hours. The back bone of this service is the train network, dedicated workers and an ingenious system.

Each tiffin tin is marked with a simple code which indicates which train, which stop and to which building the tin is to be delivered. After lunch the system runs in reverse and all the tins are returned back to their original homes.

Some Remarkable Facts

• It’s 120 years old.
• There are 5,000 people employeed.
• They make 260,000 deliveries per day or almost 80 million transactions per year.
• The workforce is mostly only semi literate.
• They operate with a 6 sigma level of quality (Approximately 1 mistake per 6,000,000 deliveries)
• There is no IT system and no mobile phones.

Key Features and Lessons

1. The organisation is broken up into ~200 x 25 person units who operate largely autonomously. (Lesson: Scaling is hard, make your organisation a modular network.)
2. The railway timetable provies a rhythm for the system, and a natural point of review for performance and problems. If a worker is consistently late then it’s obvious quickly and both can and must be addressed immediately. (Lesson: Build introspection and a quick response to problems into your system.)
3. Each Dabbawala is an entrepreneur in their own right able to negotiate prices with their own customers (within some basic guidelines). This direct relationship with the customer means Dabbawalas own their customer and tend to work in the same area for a long time. While a group doesn’t have a monopoly over a partiular area there is also a no-poach agreement. (Lesson: Give people autonomy and have them engage your directly with your customers.)
4. New hires are trained to assist with all activities for a minimum of 6 months, after which they can buy in to a group. (Lesson: Give people mastery, pair program and invest in cross skilling.)
5. Workers with more than 10 years’ experience serve as supervisors but they also still pick up and deliver dabbas themselves. (Lesson: Leaders need to get their hands dirty and understand how sausages are made; they need to be in the system to learn.)
6. 

   Via NY Times

   The coding system on the tins contains enough information to know where it needs to go without containing a full address. The workers who run the same routes for a long time don’t need all the details and adding more would slow down sorting process and risk errors. (Lesson: Only  have as much documentation, process and governance as needed; any more is a waste.)

7. Because of the unpredictable nature of traffic and other issues in a big city, the system must have a buffer; each team has 2 or 3 extra workers to fill in wherever needed. Because all members are cross trained they are able to fix problems with transport, sorting, customer service or finance issues.   So while the system is as lean as possible, it still has to have the capaicty to work when shit happens. (Lesson: Use humans to solve your edge cases, not process.)
8. The dabbawalas vary enormously in age and tend to remain with their groups for their entire working lives. As a result team members then care for each other; an elderly worker who can’t carry heavy loads is then given other jobs but still paid the same. (Lesson: Treat people as humans, not machines, and they will be very loyal.)
9. They have a simple shared purpose – deliver food on time, every time. (Lesson: Give people a purpose they can understand and believe in (hint: not Profit or the CEO getting his or her performance bonus).).

My Conclusion?

Like the women who built ships in World War I and II and the incredible achievements of the Lockheed Martin Skunkworks the Dabbawalas provide another example that with the right system ordinary people can do extraordinary things.   As leaders, focus on building and maintaining your system. Or as Demming puts it, 95% of the improvement lies with the system, and only 5% with the people.

“In a review of major accidents from 1978 to 1990, the National Transportation Safety Board (1994) found that 73% of commercial aviation accidents occur on the first day of a crew pairing.”

“NASA researchers analyzed the causes of jet transport accidents and incidents between 1968 and 1976 and concluded that pilot error was more likely to reflect failures in team communication and coordination than deficiencies in technical proficiency. In fact, human factors issues related to interpersonal communication have been implicated in approximately 70% to 80% of all accidents over the past 20 years. Correspondingly, over 70% of the first 28,000 reports made to NASA’s Aviation Safety Reporting System (which allows pilots to confidentially report aviation incidents) were found to be related to communication problems.”

“Communication is critical in order for cockpit crewmembers to share a “mental model,” or common understanding of the nature of events relevant to the safety and efficiency of the flight. This is not to say that effective communication can overcome inadequate technical flying proficiency, but rather the contrary: that good “stick & rudder” skills can not overcome the adverse effects of poor communication. “

Considering the level of training, regulation and procedures in commercial aviation it’s somewhat counterintuitive that the quality of communication between crew has such a dramatic effect on the team’s performance, especially in a stressful situation when something is going wrong.  The same observation has been made in the medical context.

“In other safety-critical systems such as surgical operating rooms and medical intensive care units; Medical researchers have found evidence that it is not the technical or medical proficiency of healthcare providers, but rather the quality of their interactions which predicts outcomes.”

The research into cockpit communication reveals a few interesting things.

1. Individuals have a distinct pattern of language usage which is quite stable over time, and in different levels of stress.
2. The best crews were more verbose.
3. The crews communicated about two and a half times more in abnormal flight situations.
4. The number of words spoken was correlated with higher performance and lower rates of error.
5. Crews which exhibited familiarity with each other in their language (talking as a team, “we” and “us”) also had higher performance.
   
6. High performing crews make problem solving utterances 7 to 8 times more often than the poor performing crews, eg. asking questions out loud, or making statements which invite input or share crucial data.
7. The high performing crews spoke in problem solving utterances the same amount of the time whether they were in a routine situation or a high stress situation.  The poor performing crews didn’t (and thus were out of practice when it counted).

Time spent talking and building relationships in a team, learning each other’s patterns of language usage, is essential to bonding, building a shared mental model and in turn delivering better team performance.  This is one reason pair programming is such a powerful learning and quality generating tool.  Likewise, perhaps we shouldn’t worry so much when our stand-ups drag on a bit?

*All quotes are from “Using Language in the Cockpit: Relationships with Workload and Performance” - J. Bryan Sexton & Robert L. Helmreich which itself has an extensive list of supporting references.