I was reading an article about how measuring Average Handle Time (AHT) in a Financial Services company, did not allow a very eager Financial Services Agent to provide the best service she could have provided a customer.

The customer wanted to do something on the Financial Services company online website, but the Average Handle Time (AHT) metric did not allow her to spend more time on the phone with the customer, and give him more information that would have made his online interaction with the company’s website easier. Seems like that would brought the AHT metric for the whole center, as well as that particular agent down. So she had to bite her tongue and not tell the customer about something that would have prevented frustration on his part in the first phone call even though she knew about it! When he tried doing what he wanted to do, he could not and has to call them again on the phone!

The center explained that they were trying to increase the Customer Satisfaction KPI and if the AHT value goes up, the Customer Satisfaction KPI might suffer because that meant someone else was waiting to be serviced. However the above customer who had to call back again, would definitely would have brought that measurement down anyway!

So they may not have achieved anything more than frustrating an important customer!

Measurement and Reporting of KPIs should also identify these kinds of outliers, or exceptional cases, and allow them to be included in the analysis of the KPI performance. Metrics drive behavior and your interpretation of metrics should not enable the driving of undesirable behavior, eventually!

Many time, once we put technologies and metrics together, we think we have put behavior on auto-pilot! We may have automated the driving of undesired outcomes, rather than the desired ones instead! They are our tools and not ends in themselves. The ends, we will have to be very clear about and communicate them well!

Exceptions and outliers happen all the time and as long as we interpret metrics and measurements along with them, we should be fine!

The young man knows the rules but the old man knows the exceptions - Oliver Wendell Holmes

The funny thing about Process Improvement, whether Six Sigma or Lean, is that not all Process Improvement efforts are equal to others.

Some are "More Equal" than others!

Efforts needed for different Process Improvement ideas may not be linearly proportional to their payoffs. Before jumping into Kaizen activities, it may be worthwhile to prioritize efforts with respect to effort needed vs payoffs.

Sometimes a simple, almost "effortless" improvement effort may lead to a disproportionate payoff while a very expensive improvement may result in a not so impressive payoff.

In processes, especially when it involves the public, delays in service cause more dissatisfaction than the actual services themselves. Everyone hates waiting. In as much as waiting can be eliminated in business processes, customer satisfaction metric registers increases.

Analysis and eliminating delays in business processes, invariably leads to better service and disproportionate payoffs. Funny thing is that some of these payoffs show up as better word-of-mouth recommendations and additional business; something not immediately recognizable and acknowledgeable.

However beautiful the strategy, you should occasionally look at the results - Winston Churchill

We seldom think of Unused Capacity as a Wasted Resource! In fact, it can be a wasted resource that can be fatal to the long term survival of your company and even your industry!

The Wallstreet Journal ran this story last Saturday - More Car Plants at Risk. It talks about Overcapacity in Automobile Manufacturing for Light Vehicles. We all know the trouble the car makers are in because they ignored the small car segment in favor or gas guzzlers and the market for them suddenly collapsed in 2008.

They used the following graphic from IHS Global Insight in the above article showing the difference between current and Projected Capacity and Projected Utlization!

What is remarkable is the very strict management of Capacity and Production in the past and the future of Toyota Vs. the Big 3 American Automakers! Toyota is also making losses currently but they might recover sooner than the other ones, just looking at these projections.

If you think about it, Over Capacity has a lot of costs associated with it - Idling Plants, Idling Huge Investments in these Plants for which some of these companies may be paying interest, Idling workers that are paid not to produce, Idling workers that are maintaining these plants even when they are on Ice with no actual workers around (Security, Preventive Maintenance people), etc.

Over Capacity may prove to be a huge huge waste and could be sucking a lot of the profits of the company even when some of your plants at producing at full capacity and making enormous profits for you!

Something to pay attention to, not just in Manufacturing but also in Services! Keeping the Capacity very close to Production in services, can be done easily with Multi-Skilling and good workforce optimization! There are lots of algorithms and software based on those, to account for seasonality of demand by hour of the day (Evenings and Nighttime for cusomer service on the phone, for example) and , day of the week (Mid week is peak for many business services), month of the year (Summer Travel Season or Thanksgiving for AAA services, for example) or season of the year (like Christmas!).

Training people to perform multiple tasks at work could go a long way in balancing demand and supply for business processes and services, smoothing out the overcapacity problem at any time.

Over Capacity could be one of the biggest wastes whether in manufacturing or in services or in business processes. Keeping capacity very close to demand adaptively with multi-skilling and good workforce optimization.

Production is not the application of tools to materials, but logic to work - Peter F.Drucker

I once did a crude study, at our local library, of the Average Financial Performance of Public companies in various Industry codes. I discovered that out of every dollar earned by companies from Manufacturing to Service Industries they were spending an average of 40 Cents to 80 Cents on Operating Expenses. Of course, the Service Industries had more of their Sales revenue dollar spent on Operating Expenses and Manufacturing was on the other end. Net profit percentages were about 0 or negative to about 6% or 8%.

My guess is also that a large percentage of Operating Expenses are spent on Internal Business Processes, especially in Service Industries. On an average Operating Expenses budget of 60%, if you were to improve your business processes and reduce your operating expenses by 10%, many of these companies stand to become Doubly Profitable if you were to save an additional 6%!

It’s no wonder that companies like Toyota that use the Toyota Production System (TPS) to continually look for waste and inefficiences and eliminate them systematically. They are able to reduce their costs constantly while increasing quality at the same time. TPS has not been applied as much outside Manufacturing but the basic idea of elimination of waste is a universally applicable!

That’s powerful motivation to look at Continuous Process Improvement! Many Continuous Process Improvement efforts are hampered by availability of good data. Some of them are natural to the problem at hand, but none of them are insurmountable! Especially when there is so much to be gained!

Typical data characteristics or problems are:

a. Disparate Data Sources - An Order to Cash Process may be using a variety of backend software systems - Order Processing, Sales Accounting, Production Planning, Manufacturing and Production, Warehousing, Shipping, Billing, and Financial Accounting Systems. These may be from the same software company like SAP or Oracle or different functions may have software from different companies. End to end collection of data becomes a task of Extracting, Transforming and Loading (ETL) data to a single repository. I know Business Process Orchestration tools can collect this kind of information but what about the other 95% of companies that don’t use those currently or plan to use them in the near future?

b. Multiple Data Cubes - For the same end to end process, analysis may have to be different in each stage of the process. While processing orders you may want to analyze by regions or zones where the orders are coming from.When it is being manufactured, KPIs may need to slice and dice process data down to the manufacturing shop or teams within. When a product or service is being supported over phone or other media, customer support center metrics may be the way process is analyzed.

c. Data Availability and Integrity - If parts of an end to end process are outsourced, then the vendors systems may be involved and you may need to get data from the vendors’ systems! Data Integrity is always brought up as a key issue.

None of the above problems are insurmountable but looking at the potential for Continuous Process Improvement to improve the bottom lines of companies significantly, it should be motivation enough for companies to hunker down and address all of these individually and collect the data needed for analysis and focused Process Improvement.

It is a capital mistake to theorize before one has data - Arthur Conan Doyle. Sr.

Value Stream Mapping, very simply put, tries to eliminate Non-Value Adding activities while speeding up Value-Adding Activities. For example, in Manufacturing a Toy, attaching two pieces of the toy together is Value Adding while filling out a Production Floor Log is Non-Value adding. This is from the point of view of the end consumer, a child. The production floor log is value adding to the company but not to the end consumer. The child could not care less whether the log was filled out or not.

This works very well in Manufacturing. However while applying Value Stream Mapping to Services, two other aspects may come into play -1. Internal Financial or Security Controls and 2. Mandatory Legal Steps required. In a services setting - let’s say it’s an Accounts Payable process and the person is approving invoices to be paid. Invariably, companies have limits for who can approve what Invoices. If it is a $100 invoice, it may need only one level of approval, while if it is for $100,000, it may need five levels of approvals. Similarly, only system administrators in a company are allowed to create new user accounts. Security needs may dictate levels of approvals for this activity also. The Fair Credit Act in the U.S may dictate distinct steps to be followed before an overdue account can be turned over to a collection agency. If the person says on the phone that they need to consult an attorney then the company may need to follow up with a legal notice of some kind. If they don’t say that they may need to follow another set of steps legally.

In Manufacturing, many of these mandated non-value adding steps may not be that much of an intrusion into Value Stream Mapping and improvement of those processes. In Services, these activities may not be eliminated completely because the law requires you to do them diligently. You may deploy technology to speed these activities up, even if you cannot eliminate them. Deploying methods such as sending a notice by Email may be legally acceptable, instead of paper snail mail.

Services are somewhat different from Manufacturing, but appropriate adaptation of Value Stream Mapping methods may produce similarly excellent resulsts!

The wise adapt themselves to circumstances, as water moulds itself to the pitcher - Chinese Proverb