Month: October 2014

On a very crowded train the other day I watched a short story unfold. A woman with two small children, each with their own seat took one of the children onto her knee to free up a seat, ignoring his protests of 'that's my seat'. Instantly a young suited man sat himself down in it. Almost at the moment the 'closing train doors' announcement was made a much older man with a small girl leapt in the closing gap and onto the train both breathing hard having run to get it. Seeing no seat he and the child sat on the train floor. The woman saw this and asked the young man to give up the seat (the one she'd freed up) to the older man and the girl. He said he wouldn't as it was 'his seat'. (So two seat 'owners' within a few minutes). To her credit, the woman persisted and the young man finally gave in – but not with good grace -and gave up the seat to the older man with – as it turned out – his grand-daughter.

I'm often struck by people's attachment to 'their' seating. Watch people at a conference after a break – they'll usually return to the seat they took when they first walked in rather than seek out a different seat. In an office if hot-desking is suggested the first outcry (of the many that follow) is about 'my seat'. Look around most offices and you'll see seating with dire warnings stuck on the back. 'This is xxx's chair. Do not adjust it.' I haven't seen the same on monitors though they too are adjustable.

I know there are workstation layout recommendations to avoid ergonomic issues, but they are not a concern in short duration training or conference seating. Even so, a form of possessiveness about 'my' seat seems to take over regardless of the event. I don't know why this is. I've tried some experiments in sessions that I've run asking people to change seats after the coffee break. For the most part a handful are happy to do so, the bulk are not too happy about it and some point-blank refuse. It seems to me to be a great replication of people's reactions to any change process. We usually have a fun discussion about the bell curve range of responses on the lines of – how can we change a whole organisation if we can't even change our seats (and what's stopping us from doing so?)

I was on train where I watched the seating story, coming back from a couple of meetings I'd been to. The first was a workshop I ran on organisation design with about ten people. Because I'd forgotten to specify my preferred room layout, which is cabaret style, I found it was pre-set up with chairs arranged in an open circle. They were not writing chairs (the ones with an integral flap up surface). I know that 'open circles are fashioned in such a way that interconnectedness, interdependence, and equality within the community are highlighted. Participants are encouraged to share a sense of mutual responsibility for the well-being of the community and the individuals within it, and an understanding that what happens to one person affects all.' But when I'm a participant I don't like it because a) how can I easily write anything b) how can I look at my BB on my knee under the table c) why do I feel exposed and vulnerable d) why do I have to address the whole circle and not just a few people?

When I'm facilitating such an arrangement it's marginally easier but I still don't feel it works very well. Although it's supposed (I think) to be informal and 'restorative' to me people just don't look comfortable balancing handouts on their knees with no-where to put their coffee cups. My experience is that the circle often feels stilted and formal, as if inclusiveness is being required before the inconsequential niceties of getting to know each other a bit, and when people do know each other the conventions of 'inclusiveness' seem to gloss over the tensions and politics. (After the coffee break I suggested that this group's members change seats with the same result as I mentioned above).

The next workshop – same building, different group – had the same chair arrangement when I walked in but also randomly scattered bean bags. What started in an open circle quickly got rearranged via various activities into smaller clumps of mixed chair, bean-bag, floor sitting which gave a very different feel to the time spent together. And people moved around depending on their interests without any suggestion that they do so. What made this a more collaborative meeting I wonder? There are lots of variables – the facilitation, the group dynamics, the topic, the activities, the seating available, start time, etc. I wondered if the addition of the bean bags was the variable with the most impact – they kind of invited a relaxed feeling different from the 'therapeutic' feeling of the formal circle.

I've been conscious of seating in various settings this week because we are about to try out hot-desking (non-allocated workstations) with a group of volunteers and we're also thinking about introducing informal 'soft seating' arrangements. I've tried the hot-desking experiment before (see my blog on it) and I've learned that there is something about the desire to possess one's own seat that almost over-rides the desire to have a pedestal to put one's stuff in. To work with this own seat need – where is it on Maslow's hierarchy? – we had the idea that seat owners park their chairs in a 'parking lot' each evening so the next morning they could take 'their' chairs to the workstation they happened to be working in.

This, we felt, would solve the ownership issue and be perfect for those who a) didn't want to turn up to a workstation and find that the chair that went with it was a fit ball one b) didn't want to waste time adjusting a chair to their own spec c) felt queasy about sitting in someone else's chair. (Though the same people seemed to be ok commuting, eating in restaurants, going to the cinema, and so on). We didn't get this to work though because there wasn't enough space for the 'parking lot'.

Of course we could circumvent the seating issues by going to standing: standing desks, walking meetings, standing-up meetings. However, we have diaries packed with 'standing' meetings by which we mean regularly occurring ones and we also have 'stand-ups' which are used in agile software development teams. Paradoxically, people can sit in both of these types. Thus we would need to be careful about not confusing people about what we mean by 'standing' as opposed to 'sitting'.

I mean standing (on two feet) and since we now know about the health benefits of non-sedentary work we could make the case for removing chairs altogether. In fact it has already been made by Professors Andrew Knight and Markus Baer, of Olin Business School. They 'suggest removing chairs could be a low-cost way to redesign office space and tackle the health effects of prolonged sitting.' (An alternate to this is taking out all the desks and converting the chairs to standing desks using the new-to-market Storkstand – an ingenious invention).

The nearby park is an ideal location for the walking meetings that are said to increase productivity, health and fitness, and lead to better quality conversations. Some of the effects might be off-set if you were walking with someone smoking a cigarette as it's allowed in the parks, at least for the moment. You'd have to do an organisational cost-benefit analysis of the loss of productive time spent on solitary smoking breaks v the time spent on productive walking meetings alongside a smoker but with possible passive smoking consequences.

So you see, I'm still working on the 'I own my seat' conundrum which could be problematic as we head towards smartworking. How do you handle it in your organisation? Let me know.

Job families hit my agenda last week. If you're not familiar with the term it means categorising jobs into like types. So you might have a job family of 'Digital' that has a range of roles in it: software developer, user experience designer, web ops, web analyst, content designer, and so on.

In a large organisation knowing what categories of jobs and how many people hold each type and at what level is immensely useful, for various reasons (in no particular order):

• Enabling skills and experience development to a higher level with the role, for example from 'practitioner' to 'expert'.
• Facilitating the match of skills to available roles within the organisations
• Giving a good platform for focused recruitment
• Aiding with feedback and performance discussions
• Supporting succession planning
• Grounding capacity forecasts in some evidence (is what we have likely to meet forecast skills demand requirements?)
• Matching reward to market rates
• Help with comparing/evaluating jobs that are in the same family but not in the same business area

I'm quite a fan of the possibilities job families offer both to the individual and the organisation but as with all discussion of families, job or 'real', there are multiple competing views on them and multiple expressions of what a family is or isn't. Just look at the number of books, films, plays, and poems there are about families. I just today read a review of a new book about playwright Tennessee Williams who 'was raised in a supremely dysfunctional family anchored by two parents who loathed each other' and as I continued to mull over families I discovered a list of 25 films about them.

When we started discussing the notion of moving towards 'job families' in the organization I'm working with we didn't have an easy discussion resulting in 'great idea, go ahead'. What unrolled was the spectrum of views on what was supposed to be job families but seemed to come across as perspectives on families in general.

At one end were people who clearly saw families as nurturing units interested in doing the best for each other in a supportive encouraging way. At the other end were those who felt families were systems of abuse and disharmony and not good for anything much, expressed in statements like, 'I never got where I am today by being in a family'. Then there were those at various points between those extremes.

I wondered if there was a bit of role confusion going on – were people talking about the job families or their own families or their roles in a family? Unpicking all this and casting personalities and perspective aside seemed to reduce the discussion to three useful questions:

1. Will categorising all the jobs into families be a huge, time consuming, expensive exercise with little benefit?
2. What's wrong with the system of generalists we've got now (and that's served us well in the past)?
3. Will putting a role in a job family box the job holder into that family so we limit their opportunities rather than extend them?

Question 1 about being a huge expensive exercise is worth exploring. Each of the major HR consulting firms offers a service around job evaluation which seems to involve some kind of categorisation. Mercer, for example, offers the International Position Evaluation System which is a 'battle tested' methodology. And Hay now offers 'Hay Group Spectrum'. Looking at this it is difficult to see if it involves job families as it used to (see one of their presentations) because it doesn't mention the phrase. But it does offer all sorts of rigorous analysis around four aspects of work. Unfortunately they have the word 'rigorous' mis-spelled several times which gives the whole spiel a slightly dubious flavour in my reading. Towers Watson offers job levelling which does include the words 'job family'. Aon Hewitt offers 'Career Link' providing 'clear descriptors for 11 levels in 5 separate career families' and I see from other providers references to career families.

Hmm – now I'm confused. What is the difference between a 'job family' and a 'career family'? Googling the question tells me that in any family with two potential earners and some dependants, one is the main bread winner who has a 'career', and the other has a pin-money 'job' which leaves time for the dependants (and, implied, the washing-up). So now it looks as if we may have two classification systems – one for the career people in organisations and one for the job people in organisations? Can this be fair (or likely)?

To get back to the serious question about categorising jobs into families: yes, it can be costly and time consuming but it doesn't necessarily need to be. As with anything it depends on the level of detail required, the purpose for doing the classification, the use it will be put to, the type of data that already exists that could be pulled from to inform the classification work, and the system that it is going to be pulled into.

In some organisations there is already a platform of job family – many organisations employ 'professionals' who have certifications relevant to their profession – accountants, for example. The UK Civil Service 'is made up of a wide range of professions and includes every kind of professional – from beekeepers and veterinary surgeons, to bomb disposal experts and accountants. There are currently 22 recognised professions, each led by a government head of profession.' Job families by another name, perhaps?

Question 2 'What's wrong with the system of generalists we've got now (and that's served us well in the past)?' is a question that seems to relate to the culture of the organisation in part and the way jobs and positions are perceived. Oracle has a daunting 580 page manual on the topic but you can read a slimmed down few paragraphs which talk about jobs and positions where 'jobs' are the generic instance something e.g. 'Accountant', (in other words a job family) and 'Positions' are specific instances of a position within that job e.g. Payroll Accountant. They go on to describe the need to decide whether to use jobs, positions, or a combination of both, and state that 'Enterprises fall into one of three general categories:

• Rule based: in this organisation you regulate employment, roles, and compensation according to strict policies and procedures. Fixed roles tend to endure over time, surviving multiple incumbents. You manage roles rather than individuals.
• Project based: a project-based enterprise, such as a construction or software company, requires the flexibility to assign people to new projects or organizations on a regular basis. You manage people and their skill sets, rather than fixed roles. This requires the flexibility to match competencies to tasks quickly and easily. Project-based organizations, where roles end when individuals complete a project, typically model the enterprise using jobs.
• Hybrid: If your organization is a hybrid enterprise, you assign some individuals to fixed roles, and others to multiple projects. Hybrid enterprises model the enterprise using both jobs and positions.

What I've noticed in organisations that don't make conscious decisions around jobs, roles and positions is that they can end up with literally thousands of job descriptions with very similar characteristics but with very different grades, pay scales, career possibilities, and so on. Determining and governing the choices made avoids this kind of organisational anarchy that can more limiting than the job family decision. For example, job family classification enables technical progression 'up' a family e.g. to Chief Economist, as well as managerial progression. Many hierarchical organisations lack the parity of technical/managerial progression putting good technical people into managerial routes because it is the only promotion option.

So, it seems that the next round is to carefully consider the topic and make conscious decisions about it. What's your view of job families and what factors inform your thinking? Let me know.

I was in a phone discussion last week with someone who wanted to know how many layers their organisation should have and what the right span of control should be for each manager. (Layers in an organisation refer to the number of levels of staff there are from the most junior to the most senior. A span of management is the number of employees that a single manager is responsible for, usually in terms of allocating work and monitoring performance.)

It's a question I get asked frequently and people want to know the 'right' answer. I'm usually a bit flummoxed by this need for a formula when there are so many variables. Some organisations e.g. the US Government has 21 layers while others organisations e.g. Valve, a software gaming company, have one or two – they cheerfully say that 'nobody "reports to" anybody else'. This is likely to be the difference between being an established 'analogue' organisation and being a newish 'digital' one (See the white paper 'Building a Digital Culture ' from PWC) although that's a bit glib. Digital organisations are more recently established than most analogue ones and thus haven't had the time to accrete the trappings of bureaucratic infrastructures.

However, since almost all organisations are aiming to move from analogue to digital (including government organisations) surely organisation designers and developers should be looking again at the layers and spans accepted wisdom? This established approach is typified in the information I found when I scanned my 'layers and spans' folder. It comes from the Arizona National Guard Human Resources Office, and unfortunately is undated. It firmly states:

Basic requirements:
(1) First level supervisors over a General Schedule (GS) position should supervise 6 to 8 positions.
(2) First level supervisors over trades and crafts, wage grade (WG) work should supervise 8 to 15 positions.
(3) Second level and higher supervisors (WS or GS) should have no less than three (3) subordinate supervisors reporting to them.
(4) Deputy and full assistant positions should be used only in large, complex organizations.
(5) When two (2) or more employees occupy identical additional (IA) or almost identical positions, each employee must work at least 50 percent of his or her time at the grade level of the position. Where many employees in the same organization are classified to the same type and level of work, the percentage of time spent working at the classified grade level should substantially exceed 50 percent.

This prescription was probably ok in an era (or organisation) of command and control, but this has been eroded by all kinds of factors: easier access to information, employment patterns, technology, societal changes, and so on. Looking for a prescription on layers and spans is not going to give a 'right' remedy.
But because we seem wedded to layers and spans (see the article 'Equality may be cool but staff still like the security of hierarchy') I have a section on them in the forthcoming second edition of my book on organisation design. Here's an extract:

'Layers and spans are structured to help managers get work done, so the first part of an organisational decision on the number of management layers and the span of a manager's control requires discussions and agreement on what managers are there to do. In general, managers plan, allocate, co-ordinate and control in order to achieve what the late Peter Drucker described as their three tasks:

1 To contribute to the specific purpose and mission of the enterprise.
2 To make work productive and the worker achieving.
3 To manage the social impacts and social responsibilities of the organisation.

Clearly, determining what configuration of layers and spans is likely to work in a given organisation depends on the situation, organisational purpose and a host of other factors related to the interpretation of what the three tasks entail and the weighting given to each of them.

Multinational conglomerate General Electric (GE) has learned that as situations change so does the need to review layers and spans. Its annual report 2013 (p.8) states that

'To achieve our goals we must simplify GE … Having lived through multiple crisis events in the past decade we attempted to manage volatility through layers and reviewers. Like many companies we were guilty of countering complexity with complexity. … We are transforming GE around the 'culture of simplification' … We are driving a leaner structure … We have learned that fewer layers, simpler rules and more field empowerment improve execution and accountability. … At GE, we think simpler is better. Simplification means quicker execution and closer collaboration with customers. It's a focus on efficiency, speed and market impact. '

To help get a good enough answer to the "how many layers " question, there are four rules of thumb (related to the four management activities of planning, co-ordinating, controlling and allocating). Each layer should:

• be flexible and adaptable enough to enable managers to forward plan in a context of constantly changing operating environments;
• facilitate co-ordination between business units (Michael Goold and Andrew Campbell suggest there are six forms of business unit to unit co-ordinating activity: leveraging know-how; sharing tangible resources; delivering economies of scale; aligning strategies; facilitating the flow of products or services; creating new business);
• have appropriate control and accountability mechanisms (note that any task, activity, or process should have only one person accountable for it and accountability and decision-making should be at the lowest possible level in the organisation; overlap and duplication, fuzzy decision-making and conflict resolution processes are all symptoms of lack of adequate controls);
• enable its managers to allocate effectively the range of resources (human, time, equipment, money, and so on) they need to deliver their business objectives.

If these four attributes are working well, it is likely that the layer is adding value to the organisation, in that it is facilitating speed of operation, innovation, integration, flexibility and control. If it is not evident that the layer is doing this, it may be redundant and the reason for its existence should be questioned.

Determining a sensible span of control is possible (though infrequently done) both for an individual manager and for the type of work carried out in a business unit or organisation. The method involves considering the following:

• The diversity and complexity of the work performed by the organisation
• The experience and quality level of the workforce
• The extent to which co-ordination or interdependence is important between employees and groups
• The amount of change taking place in the work environment
• The extent to which co-ordinating mechanisms exist and are effective, geographic dispersion
• The extent to which job design and tools allow direct performance feedback to the employee
• The administrative burdens on each level of management
• The expectations of employees regarding development and career counselling.

Robert Simons suggests that any job comprises four different spans: control (including people, working capital, facilities, infrastructure and intangibles), accountability, influence and support. Each of the spans can be adjusted to reflect the business strategy and meet current organisational requirements, but to ensure high performance the spans related to the supply of resources (control and support) must be in equilibrium with the spans related to demand for resources (accountability and influence).

There is no right number of people that one person can manage (though a commonly held view is that five is the optimum – but that could well be a myth) as various factors affect a manageable span. The relationship between spans and layers is not straightforward either, although wide spans of management are typical of organisations that have few layers.'

What's your view on layers and spans? Let me know.

PS When I scanned my 'layers and spans' folder I found a guide from Booz, the consulting company, Managing Spans and Layers (Booz is now part of PWC). Bain, another consulting company, has one too Streamlining Layers and Spans.

I've been keeping half an eye on the Tesco news in the last couple of weeks. It's A UK based supermarket chain which has been found to have overstated its profits and is now being investigated, because. "Breaking accounting rules to exaggerate profits is a cardinal sin as far as investors are concerned and Tesco has been punished severely with shares falling more than 10% at one stage …" Four executives have been suspended. It is not clear yet whether they colluded in the profits overstatement, but that crossed my mind as a possibility.

Then, in the way of things, I then started to notice other stuff going in which conformed to my label of 'collusion'. I looked up the definition just to check whether I had the right term in mind.

Collusion is an agreement between two or more parties, sometimes illegal and therefore secretive, to limit open competition by deceiving, misleading, or defrauding others of their legal rights, or to obtain an objective forbidden by law typically by defrauding or gaining an unfair market advantage.

It's a firmly worded legal definition involving unfairness, malpractice and intent and in this context often applied to cartels or price-fixing. However, this didn't quite address the complexity of things I'd observed which were less about specific intent to deceive or mislead or defraud and more about people colluding in 'just obeying orders' or going along with something that they felt uneasy about for any number of reasons that include not wanting to 'career limit', fear of the person with more positional power, not wanting to 'rock the boat', fear of penalty for non-compliance and so on. These 'fear factor' things – individually or combined – have the consequences of eliciting collusion in something that isn't quite right.

When I worked for British Airways I worked on a project on collusion amongst airline crew. Many aircraft accidents happen because there is an unhealthy relationship between the pilots and the other crew members. (See some examples from the 1970 – 1990s here). First officers, for example, frequently wouldn't speak up when they saw something going wrong – effectively colluding in a belief that things were fine.

This 'going along with something' that is questionable or feels wrong/unfair, or corrodes respect and trust I think is form of collusion, which is damaging at many levels. Sadly, in most organisations it takes bravery and also recognition that it is likely to be personally damaging to challenge a situation, system or process that is questionable. Who wants to get the treatment that whistleblowers typically get or to be labelled a 'maverick' or a 'troublemaker?

One of the articles I read this week was 'Just obeying orders: rethinking obedience and atrocity' . In this, the author discusses how the now infamous Milgram experiments purport to 'provide startling evidence of our capacity for blind obedience – evidence that inhumanity springs not necessarily from deep-seated hatred or pathology, but rather from a much more mundane inclination to obey the orders of those in authority, however unreasonable or brutal these may be.'

But this researcher offers another perspective on the Milgram experiments. In his analysis 'what this means is that those who shock do so not because they are unaware of the consequences of their actions, but because they know what they are doing and believe it to be worthy. Rather than being blindly obedient, they are engaged followers'. He goes on to say 'One thing, however, is certain. Whether you agree with Milgram or not, or accept our "engaged followership" theory, there are few issues in psychology that are of greater social significance. These are not just a matter of the academic understanding of authority, obedience and genocide. "Obedience" has long served as an alibi for those involved in atrocities, and it is routinely articulated in the defence: "It wasn't my fault, I was only obeying orders."'

This is a rather startling finding when we consider how much emphasis we place on workforce engagement and 'getting people to buy in' both of which could also be construed as 'be obedient'. And this has led me to wonder if the very mechanisms we (organisation design and development people) use to either maintain the status quo or to introduce new things, is in fact done in order to create 'engaged followers' who are unwilling or unable to question the leaders, the rules, or the 'way we do things (or want to do things) round here' whether it be in the established or the new order.

Take an example of many established performance management systems that operate a stack ranking system. "Stack ranking, also referred to as forced ranking, where managers across a company are required to rank all of their employees on a bell curve, has been a controversial management technique since then GE CEO Jack Welch popularized it in the 1980s. Only a small percentage of employees, typically about 10%, can be designated as top performers. Meanwhile, a set number must be labeled as low performers and are often fired or pushed out, giving the system the popular nickname "rank and yank."

Numbers of those involved in operating stack and rank systems are convinced that it is not helpful in developing performance. Indeed, many organisations have given up on them. When i4cp asked companies in 2009 how many of them used stack ranking systems, 49% said yes. But by 2011, only 14% used them.

But this still leaves numbers of organisations who do use a forced ranking system, where people are fully aware of its numerous shortcomings and yet continue to be 'good corporate citizens' and go along with it. Are these people engaged followers who believe in the system or colluders in supporting a practice that evidence suggests is damaging?

Examples like this abound in organisations: people go along with things that don't square with adult to adult relationships, good customer service, integrity and so on. It's not just the high profile misreporting of results or overstating oil reserves, or ignoring 'massive failures in patient care' as in the Mid Staffordshire NHS Foundation Public Trust. It's also in the stuff that doesn't catch the public eye in the same way.

Examples I've met include glossing over some types of information, choosing some statistics over others and presenting stuff in a certain way because 'that's what x wants', but then I wonder whether this is just the normal give and take of organizational dynamics. Where does something shade into collusion with something that is not 'right', and what is the yardstick of 'rightness'?

These are difficult questions that I think all consultants (internal or external) need to reflect on. And Chris Rodgers has written a useful blog that does just that. Take a look at it and then try out this example for yourself. If you are in an organization that selects and recruits via a competency based assessment – as many organisations do what would you do if you felt they were not useful in selecting good candidates? A recent Economist piece on this practice of notes that, 'Not many people like them [competency based selection processes]. One complaint is that the tests favour candidates with time to practice the type of answers recruiters want-—and that they encourage dishonesty.

Would you go along with the system as an engaged follower (and if so, why) or would you challenge it (and if so, how)? Let me know?