What went wrong on the railways?

UK readers will know about the train crash in Cumbria. Early reports say the cause was a faulty set of points. They had not been inspected, as planned, a matter of days before the crash. Investigators will now turn to the organisation’s procedures.

I have no idea how railway maintenance is organised, but from knowledge of working with similar ‘preventative’ systems, I can say with confidence that if their inspection and maintenance work is managed in a classical command-and-control way, the system would be open to such failures.

In the classical command-and-control ‘inspection’ design managers worry about meeting inspection standards and schedule their resources accordingly. Often scheduling is done by computers following management’s rules. Work is prescribed, the inspectors have to follow procedures and make records of what they do. It is a ‘turn up and write reports’ design. Managers assume if standards and activity targets are met all is well.

But it is not. In telecommunications and oil pipeline maintenance, we have found such designs cause workers to run hither and thither, their activity being concerned with meeting the requirements as set, which is not the same as maintaining the system. There is no continuity or ownership, leaving the system open to people not being concerned about whether action has been taken. That, after all, is not designed to be their concern.

When work is re-designed using systems principles you get a massive increase in performance, instead of people turning up and recording things you get people turning up and solving problems. Responsibility is the design principle, instead of compliance. It is a truly preventative system, not an inspection system.

To take telecommunications as an example, we found the highest frequency cause of telegraph poles falling over was low-drop wires. The inspector’s job was to send off a form and following forms through the system revealed it could take as long as a year to remedy the situation, leaving the risk exposed. In much the same way with oil pipelines we found the form completed by the inspector would travel through no less that seventeen functions before anyone would think about turning up to do something. Through studying demand from the pipeline (things you need to be able to fix) we re-designed the job to be ‘turn up and take action’, and by allocating workers to their own geography (pipeline) they soon took pride in keeping the pipeline available.

I wonder if the investigators will think about these things. If anyone can tell me how the maintenance system is designed I would be interested to know. My bet is I am right and, furthermore, the ‘solution’ will be more inspection; probably inspection of the inspectors; the wrong answer.

Not the time to have a huff

The chief executive of the Local Government Association was reported as walking out of a meeting with the minister for local government because an advance copy of her speech showed she was going to say that citizen satisfaction with local authority services was still a problem.

Instead of having a political huff he should have seen it as an opportunity to do his job – lead – and engage his intellect in working out why we have this paradox: the measures (targets) authorities report up to government show improvement while citizen satisfaction paints the opposite picture.

The evidence we have, for every service about which we have knowledge, is that compliance with the targets and other specifications is what’s making the services worse. He might do something useful if he knew that.

CPA results show ‘improvement’

The Audit Commission reports that local authorities are improving on their Continuous Performance Assessments (CPA). More authorities are ‘four stars’, fewer are ‘no stars’.

As we showed in our report on adult social care (available from the web site), authorities with widely differing star ratings all had the same poor performance when studied as systems. We find the same with all services; it is not unusual to be working with a ‘four star’ authority that has lousy service delivery. The authorities comply with the things that got them their stars, but that is not the same as improving their services.

The legislation going through parliament is dropping ‘Best Value’, the predecessor to CPA, as a statutory obligation on local authorities’ reviews of services. Good thing, as it didn’t work. But think about how this is being done. A good leader would shout from the rooftop: ‘We’ve stopped it because it didn’t work!’ Politicians bury it quietly. They talk about ‘narrative’ – the story they tell. Just like managers who have wasted millions on customer-first and people programmes, they rationalise, they talk as though these things have ‘made their contribution’.

Perhaps it is too tricky to address why they didn’t work.

From wrong to wronger

The fact that the CPA results have gone up suits the narrative on inspection. The good news is the Audit Commission is cutting costs, there will be less inspection. The next ‘new thing’ is to be Comprehensive Area Assessment (CAA). The idea is to assess groups of organisations. I imagine this will cause top managers of these organisations to spend their time in meetings trying to work out how to prepare for this together and they will create a bureaucracy to respond to the requirements. I don’t imagine it will be easy to generate a consensus amongst the players. Moreover, it will take all of the leaders away from where they should be spending their time; how stupid is that?

The promise is CAA will mean less inspection. It will, like CPA, start with a self-assessment. Whether the specifications for self-assessment will lead to useful knowledge and improved performance or merely create waste of many types remains to be seen. But they don’t have a good track record on this.

Forget risk, think knowledge

One of the features of the CAA will be risk assessment. This nonsense has grown in recent years. Managers are obliged to brainstorm what could go wrong, sort these into probability of occurrence and potential impact on the organisation and then develop action plans to monitor and do things about those things they think are of highest risk. It becomes a cottage bureaucracy, creating and distributing waste. It is of little value but it gets you a tick in the box.

I always encourage managers to ask a different question: What does go wrong? It means you have to start with getting knowledge and I tell them they must get that knowledge themselves, not ask for reports. Most things that do go wrong are of little consequence beyond the organisation but the stuff of learning internally. It is quite OK for many things to go wrong. We are not all flying aeroplanes or running trains, where risk-avoidance is paramount. To adopt such an attitude to all risks leads to waste, not control.

Deming often pointed out things will always go wrong. It is how you behave as a consequence that matters. If you treat a one-off event as though it is predictable you create waste. I wonder if the people writing the CAA understand that.

Just returning your call – to the UK

Organisations are now promoting their withdrawal from Indian call centres as a ‘customer bonus’. Don’t you love’ em? They set up their first call centres arguing they would be ‘attractive to customers’, they just forgot, and they still forget, the purpose of the call centre is to give service, not take calls. When they sent the calls to India managers argued customers would be tolerant, knowing that the consequent lowering of costs would translate to cheaper services. I didn’t notice.

The reason they are coming out of India is the reason they went in – cost. They are learning that the cost of service is end-to-end from the customer’s point of view. There is no point in having a cheaper transaction if it just drives up the number of transactions. I guess they don’t tell the customers that because they think we’d find it hard to understand.

It’s the system stupid

One of our new consultants (at least ‘new’ in the sense that he is through his first year’s training) wrote to me. Before joining Vanguard he was an IiP (‘investors in people’) specialist. He wrote:

‘I landed my first (own) client and carried out scoping this week. It is an ‘IT support’ organisation and a previous IiP client of mine. They could not believe how much I found out about their business in just a few days. Their customer service questionnaires (all compiled from their perspective of course) had told them that they were truly a ‘customer-focused’ organisation – so we found out what really mattered, and identified that many of their customers were not getting ‘service’. We found plenty of failure demand (35%) and the capability data revealed some fairly horrendous end-to-end times from the customer’s perspective. The highest frequency demand was ‘my email is not working’ and the analysis revealed that the response was to send an email acknowledging the request! Progress chasing failure demand would then ensue. This company’s unbillable hours currently equate to 50% of last year’s profit – all due to failure demand and other waste within their system.’

And a few years ago he’d have told them to ‘invest in their people’. Now he knows: It’s the system stupid.