Date: Mon 22 Sep 2014

The New Model Organisation

A Visit to Jimdo

In February 2007, three young German entrepreneurs, Fridtjof Detzner, Christian Springub and Matthias Henze, founded Jimdo, a web hosting service. They did not start out in a garage in California but a farmyard in Cuxhaven, on the flat and windy shore of the North Sea near the mouth of the river Elbe. Jimdo is now based in Hamburg, with offices in San Francisco and Tokyo, and employs 180 people from 21 different nationalities. It produces software which allows users to create their own websites. The company motto is: ‘Pages to the People’. In 2013, the 10,000,000th Jimdo website went online.

In the last five years, the company has grown from 30 to 180 employees. Growth like this always poses serious challenges, the more so because in Jimdo’s case it was self-funded, so it could not afford any inefficiency. The first challenge is simply finding the right people. The skills required are in short supply and demand is very high, so as well as competing for customers, Jimdo has to compete for employees. The number of people they can hire sets the limit to growth. The second challenge is creating organisational mechanisms which enable the company to run its business effectively whilst growing at this rate. The two challenges are related. In fact, Jimdo has found a single solution to both which it describes in terms of processes and culture. The processes develop over time; the culture is stable.

Jimdo’s offices look and feel different. Based in a former industrial building on the edge of Hamburg’s Altona district, it is unprepossessing from the outside. A wine outlet along the side road leading to it - I was told it was heavily patronised - gives a hint that the location has more to offer than meets the eye. Coming out of the lift on the top floor, the first thing that struck me was the smell of food. The kitchen is always active. I had the bad luck to be there on a Friday, which is Sam the chef’s day off. A former sous-chef in a top London restaurant, he is a friend of Fridtjof’s and has made himself a legend in Jimdo. Four days a week he does a lunch for over 100 people which I was told would get a Michelin star. Even on a Friday, despite his absence, his spirit wafts through the office.

The décor, on the other hand, is almost starkly functional. The only unusual aesthetic features are the specially designed lamps which hang, large and pendulous, from the high ceilings. The space is mainly open-plan with some partitions and side-rooms which can serve as offices. It is steel, glass and wood with no carpet, initially giving an industrial feel. But then you spot a sofa with a teddy bear on it. Then you realise that the reason there is no carpet is so that Fridtjof and others can get around the office on a skate-board and mothers can move their prams and pushchairs about easily. And when you take the spiral staircase to the floor below there is a fish tank. Aquariums help contemplation. It is a touch Buddhist. On the next floor down there is a ‘napping room’ containing a few bunk beds. On the ground floor a new kindergarten and gym are nearing completion. All this is designed by Jimdo’s in-house architect, whom they call ‘Captain Chaos’.
The office sends messages. The minimalism tells you: ‘No waste. We are here to work. We do not care about appearances. We are here to get things done’. The sofa, the prams and the smells from the kitchen tell you: ‘We work as a family. Work is part of life. We appreciate the good things in life and enjoy them together as we work together’.

None of the three founders has an office. This is MBWA 2014-style. There is no middle management, just teams. That is it: the roaming founders – and teams. In the office you recognise teams because each has a whiteboard covered in post-it notes which visualise workflow. Team members, dressed in whatever they like, sit or lounge around desks and peer at computer screens. Their activities are regulated by processes – cool ones.
Processes were allowed in only as they became necessary. When there were 20 people, everything was based on pragmatism – if it works, do it. What proved to work consistently within teams was a pattern of methods combining Kanban and kaizen. Both evolved through trial and error. The kanban methods are inspired by David Anderson’s approach to managing work in progress. In Jimdo they like to sum it up by quoting David’s imperative: ‘Stop starting, start finishing’. The kaizen process of continuous learning is manifested in daily ‘standup meetings’ and ‘retrospectives’, both of which function as feedback loops. Over time these processes have been refined. There is now a Community of Practice for Kaizen and they have a pool of 21 moderators. Everyone uses them – even Sam the chef. Following one of David Anderson’s principles, they build in slack time to allow learning.

When they reached about 50 people, they needed more processes to link the teams together. There are three key formats: the Teamverløtung; the Open-Prio Meeting; and Goal #1.
The purpose of the Teamverløtung is to create a shared context across the company and make sure everyone knows what it going on. The German word ‘verlöten’ means ‘to solder’, so they made it a noun and gave it a cooler feel by linking it to an English word and replacing the German ‘ö’ with a Scandinavian ‘ø’. After all, Denmark is only a couple of hours away. Doing things like that is very Jimdo-esque.
Soldering the team together takes 30 minutes every Monday morning and consists of a general gathering in the Hamburg office. The founders present the numbers about signups and sales, new joiners are introduced, and four teams give updates about their progress, problems and where they need help. The meeting concludes with announcements about any company events and people are encouraged to meet anyone they don’t know. A video of the Teamverløtung is made available in San Francisco and Tokyo via internet streaming.
Open Prioritisation Meetings or OPM’s are held by teams every fortnight or so. They used to have an Idea Wall on which everyone posted their ideas for things to do. Ideas were plentiful, but resources were scarce, and nothing got done. The Idea Wall turned into a black hole, so it was replaced by the OPM. At an OPM, everyone makes a case for what they would like from the team, and the team decides what to do, given their capacity. The rule is: ‘Never make promises you can’t keep!’ The OPM’s make capacity transparent, reduce queues and stress, ensure more care is taken over ideas in the first place and can lead to spontaneous solutions to problems. Being a direct form of communication, they avoid overhead. Good ideas are shared. Some good ideas are rejected but logged for a later meeting. It forces prioritisation and that in turn forces a clarification of strategy. It raises the question of main effort. This in turn led to Goal #1.

I must come clean here and admit that I am partly to blame for this.
In 2011 I was invited by a consultant called Arne Roock to speak at the Lean Kanban Central Europe Conference which was to be held in Munich that October. I had never spoken to an audience of software engineers and project managers before, and wondered what they would make of what I had to say. I took it on as a bit of a punt. I had the final keynote slot, and talked for an hour about friction, the three gaps and von Moltke. I called the speech ‘Back to the Future’, and hinted at emerging organisational forms that could supplant our Taylorian legacy. The talk, which they videoed, seemed to go down well and I left, thinking little more about it.

A few months later, Arne gave me a call and asked if he could interview me for an article he was writing. He also told me that he was working on some of my ideas with a client, and interrogated me in particular about the need to answer what I call ‘The Spice Girls’ Question’: ‘Tell me what you want – what you really, really want’. The client was Jimdo. Arne’s client has since added him to its payroll, a fate suffered by many good consultants.
As so often, serendipity was at work. It was the right question at the right time. In the early days, everyone at Jimdo knew what mattered most and everybody pulled together. As the company grew, this alignment dissipated. Fridtjof had been getting frustrated about this, not just because it threatened strategic focus, but because it threatened the company’s very spirit. Arne suggested that the three founders sit down and answer the Spice Girls’ Question. They quickly discovered that it was a tough one, so they went away for a few days. They came back with Goal #1.

Goal #1 was chosen from among the things they had to achieve within the next few months. The choice landed on a project to create an iOS App which enables users build their own website on their iPhone or iPad. They created a Goal #1-Team, which was given absolute priority in resources, and whose requests to other supporting teams had to be prioritised by them. It was Jimdo’s most ambitious project so far both in terms of technology and complexity, involving almost the whole company. With upwards of 20 teams to coordinate, they held Goal #1 meetings involving representatives of all the teams every two to four weeks. Surprisingly perhaps, most of them lasted only 15-20 minutes. The longest took three quarters of an hour.
The App was launched in 2013. Apple put it in the list ‘App Store Best of 2013’.
Goal #1 is what I call ‘main effort’, which is one reading among several of what Clausewitz meant by Schwerpunkt. In 2013, Arne invited me back to give another talk at the Lean Kanban Central Europe Conference in Hamburg. He introduced me to Fridtjof, who bounced up to me to say ‘thank you’. (Fridtjof tends to bounce unless he is on his skateboard.) I told him to thank his Prussian ancestors.

Their own analysis of the effects of Goal #1 is interesting.
The first effect was a new level of clarity which creates compete alignment, up, down and across the organisation.
That did not surprise me.

The second was an increase in the speed of decision-making. For example, when infrastructure support becamea bottleneck for the Goal #1 team, the decision to allocate two extra admin staff to the team was immediate because the thinking had already been done. That did not surprise me either.

The third effect was a surprise for me, and for them it was the most important. They spoke of a step-change in self-confidence, not just in the Goal #1 team, but across the company. It was summed up by one person’s remark after a Goal #1 meeting: ‘Wow - I never realised what we as a firm can do. We have so much power.’ It addressed Fridtjof’s deepest concern – the spirit of the company. I would not have expected that. Shows how much I know.

Mourning Sam’s absence once more, Arne and Fridtjof took me out to lunch and we talked about the Goal #1 experience. The conversation was serious but not earnest. Always smiling from under his shock of curly blond locks, Fridtjof is somehow both relaxed and intense. There is no hint of formality or of being ‘the boss’, but he is a man with a mission. He is someone you take seriously.

I asked whether the creation of the Goal #1 team created resentment. Did it make those not on it feel like second-class citizens? There was a bit of that at first, they said, but not as much as they had feared, and it diminished as people realised that it involved the whole organisation. If successful, Goal #1 would brighten their collective future and it is now seen as a collective achievement. Nevertheless, optimising the whole always means sub-optimising some of the parts, and a few teams felt the pinch of not being on main effort. The founders had the courage of their convictions and stuck with it. Most organisations suffer from the parts seeking to optimise themselves, which sub-optimises the whole. Taking a holistic view and actually running an organisation as an organism takes some courage.
I asked whether, conversely, the Goal #1 team suffered a lot of stress by being under the spotlight and subjected to intense pressure. Not really, they said, because the team had all the resources it needed and everybody’s support.

The goal itself had been cleverly chosen. The debate between the three founders had raised fundamental issues about strategy and purpose which they had needed to think about. They have now created a new Goal #1, which rests with a different team. Teams are permanent, and none is an élite. High performance is routine and extraordinary performance can come from anywhere. They let some time pass before identifying the new Goal #1 and made sure it was one which could be achieved within a couple of months. It stress-tests the whole organisation, so some time is needed to consolidate learning and they do not want the stress period to last too long.

What about financial incentives? There are no bonuses. As there is very little hierarchy, there are few opportunities for promotion and no real career ladder. They pay the market rate for new hires, and individual pay rises depend largely on how fast people develop rather than taking up a new position. For now, pay is decided upon mainly by the founders, but that will soon have to change. They used to do all the hiring. Now it is done mostly in the teams. Remuneration will probably come to be done in the teams as well. In any case, money does not seem to be an issue. The value system is orientated around different things. The bedrock of everything is the culture. They describe it as their ‘True North’.

They have some core beliefs, such as ‘Everyone can do great things’. It follows that there can be no élite teams. ‘Everyone should have fun at work’. So you can go and play table tennis after lunch if you feel like it. ‘Everyone is human, so mistakes will happen’. So no-one gets blamed for errors – they just try to sort them out. They believe in visualisation. So they have actually built a model of their current strategy, which involves some model boats in Hamburg harbour moving around each other before passing out into the open sea. Groups gather round it, discuss its meaning and watch the boats’ progress. They consistently make their beliefs real.

None of the founders has worked anywhere but Jimdo. They have not had much time to do so. Fridtjof is 31. A CEO’s job is not exactly easy, and one might ask whether the triumvirate has the necessary experience. They once had a financial backer who did indeed take the view that Jimdo needed to be given some close supervision. That relationship did not last long. In my opinion their lack of experience of corporate life is an advantage. Free from the legacies which haunt most of us, they can open themselves up to learning from the most vivid teacher of all – their own direct experience.

For that is the origin of how they do things today. It all goes back to the farmyard in Cuxhaven. Three friends worked together, had fun together and achieved a lot. The shared a vision and were all committed to it. They worked hard and played when they wanted to. Most creative activity is like that. When they formed Jimdo, they tried to keep the Cuxhaven spirit and working style going. They modified it only as they had to, solving problems as they arose and experimenting with solutions. Processes and formats evolve, but the spirit is constant. The experience with the investor was traumatic. There they encountered traditional organisational structures for the first time. They had to get away, but the experience was salutary because it made it clear what they did not want. They were not quite sure what they did want, so they invented it as necessity demanded.

Their own role as leaders has evolved and will continue to do so. They are constantly with the teams. There is of course a danger in that. People could try to second guess them, or wait till they arrive before making a decision. But given who they hire, the danger is marginal. They believe in the saying, current in the Kanban community, ‘leadership is an activity, not a role.’

Keeping the culture vibrant depends on who they let through the door in the first place. New hires must want to work there and sign up to the value system. There is now a Culture Book, based on stories about the founders and illustrating things like ‘humanity’, ‘having fun’, ‘no deadlines’, and ‘being a bit crazy’. Importantly, it is the articulation of reality rather than an aspiration. Once people are through the door, the reality has to correspond to expectations. They need a guardian of the culture. So in 2011 they hired Magdalena. Her job title is: ‘Feelgood Manager’.

I asked if I could spend a bit of time with her. She wheeled a pram into the office set aside for me, and introduced her 6-month old baby, who was kind enough not to interrupt our conversation. She is responsible for the induction of new hires, managing the sport facilities and the crèche, organising feedback sessions and having an open ear for peoples’ problems. She has a no-nonsense understanding of her purpose: it is to maximise the employees’ ability to create value. The basic belief is that the better people know each other the better they will work together. Innovative ideas emerge from talking together. You have better conversations with people you know.

Part of her role sounded like being an agony aunt. ‘What kind of problems do you deal with’, I asked, ‘purely work-related, or personal ones as well? After all, at Jimdo, the line is not so clear. Where do you draw it?’ She answered with an example. A few months earlier, someone had been having trouble with his girlfriend. It got so bad he had to move out, but he had nowhere to go, so she let him use a flat Jimdo has for visitors from the offices abroad. ‘We have no policies like the amount of time off allowed for compassionate leave,’ she said. ‘I decide on the limits. In this case I was not going to sort out his relationship or talk to his girlfriend to persuade her to have him back. But I was able to offer practical support from the company. That’s where I draw the line.’
Jimdo is hard to sum up. At its heart, though, is an article of faith called ‘The Prime Directive’, which is stuck up on a wall. This is how it reads: ‘Regardless of what we discover, we understand and truly believe that everyone did the best job they could, given what they knew at the time, their skills and abilities, and the situation at hand.’
We usually judge the success of businesses in terms of whether or not they reach their goals. Jimdo has no ultimate goal. There is no end-point. But it does have a direction: ‘Pages to the People’. When will that be attained? How would anyone ever know? There is always a beyond. What matters is not the destination, but the journey. Forget teleology. Forget business heaven. This is business life. What Jimdo is doing is something much more valuable - and much more profound - than attaining goals. It is re-inventing the nature of work. I’ll sign up to that. Would you?