Lean On Me

Note: This is part 3 of our "Lean Service" topic, you can find part 1 here and part 2 here.

 ”The present style of management was a modern invention, a prison created by the way people interact” W. Edwards Deming, 1994

Imagine A New Pathway

It’s the new year and we all make our resolutions: lose weight, exercise more, pursue the important goals in our lives. We give up smoking and put on more weight; we exercise but injure ourselves; we follow our personal goal only to lose something else we did not realise we had. Our bodies, our relationships, and our world are all interconnected systems and how these systems influence each other often goes unnoticed.

Systems are also dynamic also can be adaptive in ways we did not appreciate. Take a recent article “The Fat Trap” in the New York Times. Many people when they lose a lot of weight can’t keep it off because their bodies ‘readjust’ their chemical settings and trends them back towards that pre-set fat level. Looking at the weight loss problem from this point of view is an example of systems thinking.

Systems Thinking and the principles of Lean Thinking make for an interesting journey. Companies can spend a fortune on software products and training but the metrics don’t move, or the metrics move but the customers are still no more satisfied than before.

In this post I am going to present some lean service principles and I will pepper the discussion with examples of systems effects. I am very much hoping that all this ties up in some great point about how cloud communications, SaaS, and ‘next generation’ platforms could learn from delivering value as per Lean Principles.

Some Principles

1. Costs are in flow, not exclusively in the activity or scale effects
:

We’ve tended to look at efficiency as being driven by “scale” and through getting subsystems and departments to the right scale through seeking the appropriate degree of focus. So we break customer service away into a call center, into a shared services center, or into an outsourcer that amalgamates the work from many other clients to achieve scale effects. Breaking work down into manageable units that can be standardised, given standardised times, that can optimised via specialisation (lower cost training). It isn’t always bad, but as a system produce suboptimal outcomes.

That’s because in service environments not all customer interaction work is “the same”. Thus applying mechanisation to what is essentially an organic (high variation) problem is bound to create failures and these failures cascade through the organisation creating unaccounted for costs (sic – I’ll return to this in a future post, this depends on an agent being able to effectively “pull the right” answer from the knowledge base. Of course, the percentage of the overall interaction work that is non-standard, or dynamic in nature, is also a factor here). But if the question and answers are standard and non-changing surely some kind of self service solution is preferred, i.e. for codified information it is best to automate.

As the nature of knowledge stocks becomes even more dynamic and rapidly changing our ability to respond to these changes needs to become more flexible. Simply put, the information we had about an issue yesterday, may not hold today. Fortunately the world is moving towards “collaborative work” and the ability to self organise and to create adaptive workflows is increasing. Many see promise in the new social enterprise layer to make a further impact here.

This is not to say that basic fundamental analytical attention to variation has ceased to add any value. Predictable failure demand is driven by “common underlying causes”. These are faults in the system that prompt us to look to the system itself and redesign. These causes can be quite simple such as having a confusing product install manual or a configuration that is prone to short-circuiting. You can address the inbound calling by removing the underlying cause of this failure; a better manual, a different product design. Of course if you are not capturing the true cost of these inbound calls you may never achieve a business case for redesigning the manual or the product.

Some root causes have been considered “small black boxes into which no one can see”. Why do people just not show up for appointments? In the hospital appointments post we saw that failure to attend an appointment was “predictable” in that X amount of people would just not attend. The unpredictable part of the equation was we did not know exactly which ones would not attend. Through proactive contact, at a low cost per contact, you insert a small feedback loop at a time closer to the event and receive confirmations. You thus remove one underlying root cause. People forgetting or people not “renewing their promise to attend”. These kinds of Nudges are now part and parcel of online commerce, service design, even policy development.

Breaking down the contexts of how, where, when, and between whom these interactions occur reveals a variety of contexts that previously “showed up” as the one “black box” in our flow. The delay in confirming an appointment, what does it indicate? can we generate more context information and re-ignite the flow of communication, co-ordination, and commitment cycle?. If we can, we introduce increased velocity to the process.

Thinking about all forms and modes of communication in terms of “jobs to be done” in a process flow reveals many opportunities to be more effective. Nudges, feedback loops, automations that can at the right time, in the right mode and context help both parties complete the job that needs to be done at the point in the process.

Indeed my friend Mitch Lieberman at Sword-Caboodle has a lovely post about the various “jobs to be done” of email. Each mode of communication can be mapped to different contexts in the customers life and in their journey “to get things done” with you.

Next post I will look at Variation in the System

• 
•