Management Strategy

No doubt, 60% of the top management of Semco, a small Brazilian engineering company, thought none too highly of the integrity of Ricardo Semler. They were promptly fired when the new boss, aged only 24, took over from his father. The firings, though, stemmed from the fact that the business was in deep trouble. Nobody could have guessed that the dismissals were the first stage in a progress that made Semco world-famous.

It isn't famous for its sales or its profits, but for the inventive, anarchic management which Semler pioneered in the effort 'to make people look forward to coming to work in the morning'. In doing so, Semco has created an organisation which, more than any other in our experience, strives towards the ideal of integrity - 'entireness, wholeness', says the dictionary, adding for good measure 'uprightness' and 'honesty.' A company with Semco's levels of integrity and credibility is amazingly different from a typical organisation - and there, no doubt, lies a lesson in itself.

Semler's father, an autocratic engineer, was middle-aged, when his son was born, which explains Ricardo's youthful take-over. The management style which created his fame wasn't shown at the start - Semler went through a fairly typical period of professionalising management and forcing a specialised manufacturer (marine pumps and centrifuges) into diversification: over-diversification, you might say, after looking at the present range. It still includes pumps along with commercial dishwashers, cooling units, mixers and whole biscuit plants.

What's more, that's after several other lines vanished as Semler adjusted to the hard, variable weather of the Brazilian economy. He'd thrown himself into acquisitions and other activities with frenzy and stress (personal and organisational). The growth was sensational, doubling every year, but it 'took us almost a decade to learn that our stress was internally generated, the result of an immature organisation and infantile goals'.

The 'infantile goals' were those still pursued by adult managers in every company with any ambition: what Semler calls the 'adolescent urge' for 'more people, more plants, more products, more revenue' - an urge which often explains (but doesn't excuse) gross lapses in integrity. At the price of 'money, time, and gastritis', Semler cured himself of the urge, and is now adamant: 'To want to grow just to be big is an idea that comes from the sandbox'.

The adult manager still seeks growth: to create new opportunities and wealth for people, boost motivation and productivity, and generate change. But not to please the sandbox minds of financiers. Sandbox management, moreover, is a prolific source of wasteful activity. In Semco's acquisitive phase, 100 targets were studied: there were negotiations with 15 of these; four were actually purchased.

Since 'buying small family firms is a certain way to skip the ulcers and go straight to bypass surgery', the buys were discarded subsidiaries of multi-nationals: 'the closets are typically full of skeletons', but at least the books are accurate. These subsidiaries had suffered from the same kind of matured but 'immature' organisation that Semler was to abandon. The mature corpocratic system achieves splendid results - it goes through all the right motions: what it may not achieve is right outcomes.

After professional management had wrought its wonders, Semco could 'track with great precision virtually every aspect of our business, from sales quotations to the maintenance records of each of our innumerable welding machines. We could generate all sorts of new reports almost instantly, with dazzling charts and graphs...We thought we were more organised, more professional, more disciplined, more efficient. So, we asked ourselves with a shudder, how come we were constantly late on delivery?'

The poor outcomes, Semler came to realise, were the consequence of processes that led to other objectives which, while inner-directed, caused as much internal dissatisfaction as external. The quest for 'law, order, stability and predictability' becomes an end in itself. Because nobody trusts, or is trusted, the corporation makes 'rules for every conceivable contingency. Policy manuals are created with the idea that, if a company puts everything in writing, it will be more rational and objective'.

The consequent standardisation of methods and conduct, Semler writes, 'works fine for an army or prison system'. Actually, even that can be challenged. Armies and prisons have contemporary problems, too. But you can't challenge Semler's contention that in business corpocracy doesn't work - 'certainly not for a business that wants people to think, innovate, and act as human beings'.

That realisation drove him to devise a new concept of the company. 'Looking back on it, I can't remember a single decision that I made in that period. Which was just as well, for I am at my best when I am doing nothing'. Ricardo Semler's new dispensation at Semco was built on a new dispensation for himself. While the company continued to revolve around him in an enabling sense, its processes stopped doing so. He lived by a new philosophy of management, not only for himself, but for other managers.

In a dozen years, an orthodox pyramidal hierarchy was transformed into a bazaar of unorthodox trust: of operatives trusted to set their own production quotas and to have a say in product redesign and marketing: of managers trusted to fix their own business strategies and their own salaries: of open, trusting revelation of all financial facts (including those salaries): of major decisions, including acquisitions and plant relocations, submitted to the vote of all employees: of factories moved to multi-disciplinary working in groups that are trusted to reorganise and innovate, doing it their way.

Semler calls this last unorthodoxy the 'amoeba' treatment. At one amoeba factory, set up to make electronic scales (which had been money-losers), 'The Kids' took only a few months to create such excellence that their plant became Semco's 'flagship operation'. Its productivity doubled that of the food service equipment plant, its inventories fell by 40%, its defects dropped to under 1%. What caused the turnround? 'The Kids innovated all over the place'.

Their actions recognised an important and much-neglected management truth: that internal, non-technical innovation is fully as important as technological invention (and much cheaper). Examples from The Kids are a short start-the-day meeting for all employees and, instead of a time-clock, a peg on which arriving employees placed a mood tag - green (Good Mood), yellow (Careful) or red (Not Today).

As Semler notes, the mood peg may be somewhat 'cute', but it exemplified the new culture, in which people made their own rules, while top management abolished central ones (a particularly easy and satisfying form of innovation). After concluding that 'some departments were better not created and some rules were better not written', the management wisely decided to proceed by stealth. Rule manuals were simply collected up over three or four months. People gradually found that no new manuals were forthcoming, and the rules (including control over travel expenses) came tumbling down.

Perks were abolished, too, and so was power - much to the distress of many managers: 'To be the boss is what counts to most bosses. They confuse authority with authoritarianism. They don't trust their subordinates'. Semler points out that in the pyramid structure, 'there is always a group of supervisors, department heads, and other professionals in the middle...It isn't unusual for these middle managers to be more zealous with authority...than those at the top'.

Semler doesn't gloss over the resistance he encountered, mostly from the middle, and not least because he insisted on doing business with the unions. Nor does he minimise the problems that Semco met as it grew and became 'too big for our own good' - and too complex. Attempts to combat complexity by computerisation made confusion worse (Semler believes this to be a general rule). A clinching discovery was that only 120 invoices were being issued monthly against 150: 'Two days later the unit was off the computer and back on the primitive, manual system'.

To Semler, everything now seemed clear: 'either you can adopt sophisticated, complex systems, or you can simplify everything'. Semler simplified. For example, the mainframe was disconnected. 'Our worries about making one computer compatible with another are over. It's every microprocessor for itself and to hell with the economies of scale'. Semco may well regret this as networking becomes more important. But it will never regret the introduction of manufacturing cells or its simpler 'amoeba'-like division into smaller, identifiable units.

With similar simplicity, Semler cut through the Gordian knot of profit-sharing. Instead of complicated schemes, each unit got a democratically agreed 23% of any profit and was free to allocate the money as the members decided (they mostly opted for equal amounts for everybody). This principle - letting people decide on issues that most top managements keep to their troubled bosoms - cuts out forests of complication and bureaucracy.

Semler spills out other useful hints by the score. For instance, insist on one-page memos with newspaper-style headlines ('New Toaster Will Sell 20,000 Units for $2 Million Profit'); or set up a 'Woman's Programme' to sweep away discrimination; or rotate managers between functions. Or subject managers to six-monthly appraisal by their subordinates with 'multiple choice questions designed to measure technical ability, competence, leadership and other aspects of being a boss'.

There are three dozen of these questions, each with four possible answers, with a passing grade of 70%: below that, a manager's job is in danger. Here, doubt begins to creep in. The procedure looks somewhat cumbersome. The suspicion of overkill arises at several points in Semler's account of his progress. Originally, Semco seems to have had a dozen hierarchical stages for 100-odd employees: the transformed business never had more than 850 people, but the number of managers and managerial posts mentioned by name in his book, Maverick!, makes the company sound strangely top-heavy.

Today Semco has only 300 employees, averaging $92,000 of sales each - nearly nine times the 1980 figure. That high number, however, has been achieved largely by 'hollowing out' the corporation, subcontracting everything possible, often to small workshops established by ex-Semco workers with company assistance. In this respect, as in most others, Semco does little that's unique: it's the relentless combination of many advanced management ideas that's remarkable.

Can Semco's combination transfer to other, larger firms in other countries (whose big companies, especially in America, have shown flattering interest in Semco)? Can you emulate Semler in a pyramidal organisation, or do you have to demolish the pyramid and, like Semco, go 'circular', with directors becoming 'counsellors', unit heads 'partners', other managers 'coordinators', and everybody else 'associates'?

That's all less important than the absolute integrity of management, the introduction of trust and delegation, the abolition of futile restraints and complications, and the whole-hearted use of modern manufacturing methods and people policies. Those principles should work anywhere. But very few other companies have followed Semler's practice and preaching. Nor does he actually want followers. He says there's a 'fundamental difference' between himself and gurus who are wedded to particular management theories.

'What we are doing is not the solution or the model - it's one more laboratory experience of what can happen when you remove some of the strictures'. For Semco, it's been a 20 to 25-year process, which is no use for people 'looking for a 90-day solution'. He's not even sure that 'what we're doing is a finished product, or ever will be'. While he's 'done a lot of talking and sharing experience', he's done 'little to help others follow our path' - and won't.

Nothing can demonstrate the efficacy of Semco-style management one way or the other. But Semler knows that 'changes in mentality do affect the numbers', and such changes are the prime object of his philosophy. The Semco story is 'not about empowerment and participation', which he dismisses as merely 'topical'. The experiment is 'more generic and comprehensive', an effort 'to shift the centre of gravity towards the middle'. The shift, moreover, is accompanied by the development of 'the most exciting thing': the 'boundary-less' company.

A BBC documentary team that filmed Semco in 1993 counted about two outsiders to every insider: the factory was filled with out-workers and 'consultants'. Some of the latter were accidentally recruited when 1,430 resumés arrived for a single engineering post: Semco took on 41 of the applicants, one full-time for the actual job, 40 on a 'risk basis'. Experienced early retirees were given a day to find how they could help: one applicant, for example, had worked for the company that made Semco's gear-cutting machine. He reckoned he could raise its productivity, and his reward will be 20% of the first two years' savings.

The core preserved for Semco itself is 'applications engineering and final assembly'. In these core activities, there's a 'distinct hierarchy and leadership. On issues like what product to make, or the control of sales and distribution, the leaders have 'very clear responsibilities' - even though, as noted above, 'people who are subordinates have a say and the capacity to interview and assess their bosses'. The capacity is used. In the marine products division, whose survival was in doubt, 120 out of 135 employees turned up to quiz its potential savior.

Despite the cumbersome appearance, Semler suspects that the recruitment cycle, from meeting potential executives to bringing them on stream, is actually shorter at Semco. 'We take longer over the first decision': but the selection, by as many subordinates as care to attend the interview, is more effective - and more rigorous. The subordinates come armed with a list of tough, prepared questions. The hired managers, moreover, appear to be performing well and contendedly: in a company with a tiny 1% labour turnover, the proportion of managers who 'have left or bombed is negligible'.

Their self-set salaries and their gradings by their subordinates are published for all to see, and the average grading has steadily climbed two points every year. Some Semco managers have rated 90 to 98% for three or four years, which sounds incredible: but, says Semler, 'the trend is what matters'. An uncorrected slump means that a manager is losing - or has lost - the 'capacity to lead'. In that case the solution is harsh, but inevitable.

Whether or not Semco is over-managed is hard to tell because of the changed nomenclature. The 'coordinator' title applies to anybody with 'any kind of subordinate', which means everybody from foreman upwards. At its peak employment, Semco could have employed one manager for only nine workers: however, if the visiting cards are anything to go by, the executives are fully employed. Free to use any title they wish, 80% of them put only their names: 'they don't have time' for useless niceties.

The 'useless niceties' goes for other items (like where to park, or company car rules) that also cause 'a tremendous amount of grief' in traditional companies. In other respects, though, Semco's regime is stricter. In particular, it has a five-year plan, but only six-month budgets. The latter don't allow people to bunch 'all the ugly things, or the good things, in the second half'. If the orders aren't in the house, managers may include hoped-for purchases in an over-optimistic one-year forecast. But you can't get away with that inside a six-month period - not in Semco's markets.

Most important, Semco compares the monthly numbers with expectations: that is, each executive makes 'an educated guess about the revenues, expenses, and profits for his department at the end of each month'. Comparing that guess with actuals 'gives everyone a sense of how much each manager actually knows about his area'. That's yet another highly inventive simplification directed towards better outcomes.

Semler has no room for the 'macho culture' which worries about results rather than outcomes and sets corporate achievement on a 'big scoreboard.' He asks, rhetorically, 'What are we increasing?' and suggests that the general answer should be 'the quality of life.' That's obviously a deeply felt personal view. But when it comes to the specifics of Semco, its proprietor notes that 'I can't talk for myself on any of these things.'

Indeed, he uses the word 'we' a great deal. Who are 'we'? The answer is never the same. Semco's London lawyers were baffled to find themselves dealing with a changing cast, all equally well informed about the legal issue in hand. With 'the baton passing round every six months', a weekly update 'on everything' is essential for the counsellors. The baton won't pass to Semler, though. In 1993 he hadn't hired anybody, fired anybody or signed a cheque for eight years and was planning to reduce his involvement - for good reason.

'This myth-building thing goes on'. People think I must have the Midas touch, but it's not true. But when I'm there, I have undue influence. My opinions carry excess weight'. Semler wants to 'remove this burden from the company so that self-confidence grows'. When people are 'not interested in what I think', Semler will have achieved an ultimate ambition, that of creating a 'self-powered organisation'. Otherwise it's 'just personalistic'. That's incompatible with total integrity. And if it isn't total, it isn't integrity.