Thursday, November 20, 2008

Kaizen for Product Development

Kaizen (改善, Japanese for "continuous improvement") is a very interesting concept. I attempt to apply Kaizen to product development at Maysoft. For me, this means going over each product feature and looking for areas for improvement.

For example, the most obvious feature that needs attention in the world of SpamSentinel and spam blocking in general is the need to continuously improve the block rate. When we released SpamSentinel in 2003, it blocked 70% of spam, and we were ecstatic. 2004 brought us to 90%. 2005 saw 95%. In 2006, we passed 98% block rate. In 2007, we hit 99%. Now, with SpamSentinel v7.6 we added some more blocking logic.

We are at 99.44% block rate, and we are working towards 100%. An external proof of success in the new version block rates comes from one customer comment that we recently received:

We are seeing significantly less mail & spam volume now that we are running 7.6. I have received many comments on how small spam reports are now. Very effective update.

Lee Keener, Knoxville Utilities Board

It may not be possible to achieve 100% spam blocking, but Kaizen does not say "do not try until you are sure you will succeed". In fact, just the opposite. Try little things, every day to build up to a success. Some other sayings come to mind, which I believe support our approach: "If at first you don't succeed, try, try again". And Edison's famous "Genius is 10 percent inspiration and 90 percent perspiration" If you substitute "Kaizen" for "Genius" you get:

Kaizen is 10 percent inspiration and 90 percent perspiration

which to me means that we need to think of good ideas for improvement 10% of the time and actually work on implementing these good ideas 90% of the time.

Which brings me to urgency, another concept that I believe is intimately linked with Kaizen. Improvements that are conceived need to be implemented right away, now. There needs to be a sense of urgency. If not, Kaizen loses its focus on results and moves into the category of discussion, which produces nothing except words. Tom Peters, in his
Search for Excellence book talks about "A bias for action, active decision making - 'getting on with it'. I believe you cannot have improvement without action. The risk and costs of "not thinking it through" are smaller, in my experience, than the risk of "doing nothing" or "delaying until the perfect solution is conceived".

Kaizen is a great way to guide one's thinking. You can apply it to every aspect of a product design, not just product features. For example, my Lotus Notes Spam blog is part of my personal Kaizen to improve how we communicate with resellers and customers about SpamSentinel and Maysoft. It was right after Lotusphere 2008 was finished that I decided to start writing about all the things that we were doing to help stop Lotus Notes email spam. I did not really discuss blogging, I just did it, writing the first posting, titled
A Successful Lotusphere. Now this blog is an important part of the product, announcing features and explaining new options, and telling a little about how we work here at Maysoft.

For me, Kaizen is fun, as every time I attempt to improve something, I learn something new. Blogging was a whole new world for me, but I learned a lot and now I really enjoy writing these blogs.

So, my overall thought on how to apply Kaizen to work is to never stop asking the question:

How can I start making this better right now?

Thursday, November 13, 2008

Three Laws of Customer Support

Someone asked me why I don't hire cheaper help for our support staff for my company, You see, we only hire experienced Lotus Notes administrators or excellent Lotus Notes developers in our customer support.

So I did a little bit of math on the subject of Customer Support. Having majored in Chemistry, I like equations and formulas. These equations are beautiful, in that they are simple and precise. Today, I attempt to translate some business principles related to customer support into mathematical equations to explain the three laws of customer support.

First Law of Customer Support
(Time to Completion) = (Problem Difficulty) / (Skill Level)
Explanation: We know that (Time to Completion) is proportional to (Problem Difficulty) and inversely proportional to (Skill Level) of the problem solver.
So, the greater the skill level, the faster the problem will be resolved. The denominator, as it increases, reduces the Time to Completion element. Which brings me back to cost. Our top notch support people cost a lot more per person but they perform most jobs in a fraction of the time, so they actually cost less than mediocre talent.

Here is an area everyone likes to discuss, customer satisfaction.

Second Law of Customer Support
(Customer Satisfaction) = ( Skill Level ) / (Time to Completion)
Explanation: We know that (Customer Satisfaction ) is proportional to (Skill Level) and inversely proportional to (Time to Completion). If the time to completion approaches infinity, the (Customer Satisfaction) approaches 0. As (Time to Completion) approaches 0, then (Customer Satisfaction) approaches ∞ , meaning the customer is infinitely happy.

The greater the skill level, the greater the customer satisfaction. Otherwise stated:

Faster service (done well) = Happier Customers

Now for some advanced mathematics, with substitutions. We already saw in Law #2 that:
(Customer Satisfaction) = ( Skill Level ) / (Time to Completion)

and we know from Law #1 that

(Time to Completion) = (Problem Difficulty) / (Skill Level)

So, substituting #1 into #2 gives us:
(Customer Satisfaction) = ( Skill Level ) / ((Problem Difficulty) / (Skill Level))

Simplifying terms gives us our third law:

Third Law of Customer Support
(Customer Satisfaction)= ( Skill Level ) * ( Skill Level ) / (Problem Difficulty)

So, Customer Satisfaction is improved with the square of the Skill Level! This means that a person who is twice as good will deliver 4 times the Customer Satisfaction!

This Third Law amplifies the importance of (Skill Level) as the single most critical factor in (Customer Satisfaction).

These Three Laws are quite simple and are the reason why all of the people who handle customer support are experienced Lotus Notes administrators or excellent Lotus Notes developers. I know these items cannot be measured in units like weight or volume, but the Three Laws are still true, and they help explain results that many of us know empirically, and helps me justify paying a lot more for talented people to talk to our customers, who, by the way, ultimately pay all of the bills!

You can read other postings for Frank Paolino at his Lotus Notes Spam blog: