Running Open Organisations – Lessons from the Past No. 1: Ricardo Semler

(Wirearchy Cartoon above by Hugh McLeod of GapingVoid)

There has been a lot of discussion in recent months about ways of organising work and structuring organisations in future enterprises – Podularity, Holacracy, Wirearchy, Post-Shifting , Smart Working, at a conceptual level Umair Haque and what I call “Stowearchy” ( 🙂 ) to name but a few recent thinkers, and that is not to mention all the academic work.  We have been around longer than we’d like to admit, and some of these are using concepts that have been tried before, and we wondered if there was anything to learn from these. The first is to look at the organisational revolution Ricardo Semler started off in SEMCO. To recap, the Semler story in brief is that (Wikipedia):

Ricardo Semler went to work for his father’s company, originally called Semler & Company, then a Mixer & Agitator supplier in São Paulo. Ricardo favoured diversification away from the struggling shipbuilding industry, which his father opposed. After heated clashes,  Antonio Semler resigned as CEO and vested majority ownership in his son in 1980 when Ricardo was 21 years old. On his first day as CEO, Ricardo Semler fired sixty percent of all top managers. He began work on a diversification program to rescue the company.

That was the start of a  fairly revolutionary set of steps

He fired most of the top managers and got rid of most management layers; there are now three. He eliminated nearly all job titles. There was still a CEO, but a half-dozen senior managers traded the title every six months, in March and September. Executives set their own pay, and everyone in the company knew what everyone else made. All workers set their own hours. Every employee received the company’s financial statements, and the labor union held classes on how to read them. Workers choose their managers by vote and evaluate them regularly, with the results posted publicly.

Not all was sun and light, there were problems. Attempts to introduce a WL Gore style matrix organisational structure in 1986 failed to achieve desired improvements.  It takes experimentation to make things work in any one business. There are some specifics worth noting – firstly, on cells (or Pods, as they seem to be called now):

In 1985 one of his managers suggested to Semler that he should create self-managed teams of six to eight production workers who would be entirely in charge of all aspects of production. They set their own budgets and production goals. Compensation was then tied to budget and production performance. Costs went down. Productivity and profits went up. Semler liked that. Many production workers liked that. Others were leery of taking on what they saw as “management” responsibility. It was the middle managers that didn’t like the new concepts. They felt they were losing their power and prerogatives. In a little over a year, one third of them quit.

Then came Brazil’s financial crisis and the Government severely limited access to liquid capital. The company tried everything to cut costs, but eventually it came to cutting staff, which with Brazil’s labour laws would have broken the company. At that point:

…a worker’s committee approached Semler with a proposal. They’d take a pay cut, but with three conditions. First, the profit-sharing percentage would be increased until salaries could be restored. Second, management would take a forty percent cut in salary. And, third, the workers would get the right to approve every expenditure. Semler agreed.

In the plants, workers started handling multiple job duties and using their knowledge of how the factory worked to come up with new procedures that saved time and money. At one factory they divided themselves into three manufacturing units of about 150 people each. Each unit had complete responsibility for manufacturing, sales, and financial management.

The autonomous team idea was adopted throughout the company. As it evolved the teams began hiring and firing both workers and supervisors by democratic vote. Policy manuals disappeared to be replaced by a policy of common sense. There is an actual manual, though. It runs about twenty pages and is filled with cartoons and brief statements of principle.

150 seems interestingly close to the main Dunbar Number. The next step was how these units may be set free to change focus or expand:

One more change had to be completed in order to create the Semco we see today. In the late 80s a group of engineers had received permission to become what was called the Nucleus of Technological Innovation (NTI). The idea was that they, and a group of workers, would become fully autonomous. In effect they were seeking to extend the autonomous team concept to a larger group.

Effectively the group would operate entirely on its own, though with the same culture as Semco. Their performance would be reviewed every six months. They took a percentage of sales as compensation. That model, essentially extending the autonomous teams, eventually became the model for all of Semco.

At the end of the first six months, NTI had identified 18 such opportunities. Following the success of this initiative, satellite units were encouraged throughout Semco. By the late 1980s, these satellite units accounted for two-thirds of its new products and two-thirds of its employees. As Fortune wrote in 2001, “Obviously it’s all insane, except that it seems to work.”  There is much more, described in Semler’s book “Maverick”.

After the first decade, partly due to a fainting spell when he was 25 due to the huge workload inspired him to want a greater work-life balance for himself and his employees,  Semmler wrote a second book “The 7 Day Weekend” describing their ongoing experience, focusing on the human issues. This structural freedom requires real disciple to operate properly:

The corollary of democracy and treating people as adults – the only real rules at Semco – is huge peer pressure and self-discipline. ‘It’s as free market as we can make it. People bring their talents and we rely on their self-interest to use the company to develop themselves in any way they see fit,’ declares Semler. ‘In return, they must have the self-discipline to perform.’

There’s no hiding place for those that don’t, even if performance is judged in non-standard ways. ‘To survive here you have to get on someone’s list of people they need for the next six months, and you can’t do that by playing political games.’

They also had to continually work on reducing heirarchy:

Even now [2002], laments Semler, ‘we’re only 50 or 60 per cent where we’d like to be’. Hence the constant attempts to unsettle even Semco’s unusual order – the latest of which is the disbandment of the firm’s headquarters in favour of satellite ‘airport lounge’ offices dotted around Sao Paulo. Not only do people not have fixed desks, they don’t even have fixed offices.

‘They thought it was about location. In fact, it’s about eliminating control,’ says Semler happily. ‘If you don’t even know where people are, you can’t possibly keep an eye on them. All that’s left to judge on is performance.’

Operating a business with a very low level of heirarchy and structure requires extraordinarily discipline.  In fact,  Semmler noted one of the major tasks was to screen all the New Age Work enthusiasts and those who didn’t want to get fully involved with the business out of the applicant pile, as neither type has worked out (which has provoked many to question what happens to those who cannot become fully engaged with business life). But to finish, he left a series of principles which are interesting to think about in a Social Business context:

• Forget about the top line. (i.e. profit, not revenue)
• Never stop being a start-up.
• Don’t be a nanny.
• Let talent find its place.
• Make decisions quickly and openly.
• Partner promiscuously.