Multivariable Marketing Q & A
Now is this interesting...
The real question: Is this new or just new to the online-internet-marketing marketplace? I see a mad scramble to create products that are "new" tools for the online marketer.
In the last 5 weeks, I have seen at least 4 other "taguchi based" products enter the market. I had my call with Perry Marshall on 8/27/04. And the words "success factors", "taguchi" and "multivariate" are now part of everyday online internet marketing language.
This is great because that tells me that there is a real need in the marketplace for better tools to increase MROI (marketing return on investment).
Many of the products claim to be better than taguchi, automatic, self-tweaking, work while you sleep, not a black box, easy to use and set it and forget it solutions.
I have been extremely busy working with clients and projects. So, I really didn't have the time to sit and write anything to address all the questions that I am getting.
So, I will attempt to answer the major questions here.
Here we go:
Q1) Are these other solutions Taguchi?
A1) They may be. But, I haven't checked the code to see what they are doing. So, I can't say for sure. Any text book has the standard formulas in the back. Any programmer, if he takes his time can copy the formulas and create a program. But, if you don't understand what those formulas are and how they work and the implications of the data, it is just copying formulas. There is more behind taguchi than the formulas.
Q2) For these other products that say they are "better-than-taguchi". Is that a true statement?
A2) Better is a relative statement. If the program achieves the results that you want within your budget, then those solutions may be better for you. Some companies are more than happy with an increase of 25% or 30%. Now that there is the possibility of higher increases - everyone wants 1x, 2x, 4x increases.
"Better than the taguchi methodology", I truly hope that no one is promoting this as a fact. The methodology has been around for too long and working too well to all of the sudden just be wrong.
I would not ever say that I am even close to inventing something "better-than-taguchi" This method is over 40 years old and is too well proven.
I don't think that people are inventing something "better" per se. Maybe a better platform for use in a specific area; is a better way to say it.
Q3) Why is your software so much more expensive as compared to the other solutions out there that sell between $200 and $500.
A3) I can only answer this one way. I did not just go get a textbook and open it up to the back and claim to be a taguchi expert. I am an engineer and salesperson with real world in the field experience. And you are not only getting the software you are getting my help and support throughout your experiments, plus a users forum, plus coaching calls, plus access to any of the tools that I am using to run my experiments.
But the key here is that they know what is going on with their market, message and media. They know how to interpret the results that they are getting and not just going on blind faith that the software is all powerful.
Further, they are building on that learning to produce increase in other areas of their market strategy.
Remember, any of these solutions are just tools. In the hands of someone who truly understands his tools, a masterpiece can be created. In the hands, of another the tool is not nearly as effective.
Q4) Some of these solutions offer that they can test an infinite amount of variables. Is that true?
A4) Again...I haven't checked the algorithms to see what they are using "under the hood". But, I will say this. I was in conversation with Dr. Rayjit Roy @ http://www.rkroy.com. Many of you have purchased his books about taguchi methodology.
Anyway, I was in the process of designed some very specialized arrays. And I called him to see if his software could handle my design which was more than 4 options per factor. He told me that does not exist within the taguchi methodology. So when I hear that someone is promoting that they can test infinite options.
I speculate that they may be doing DOE (Design of experiment) but you are not doing taguchi.
Remember, there are very few people in the world who are qualified to modify the taguchi methodology. Dr. Taguchi is one of them. Dr. Kowalick is another. All the others are interpreting the methodology in ways that may not be fully grounded with case studies and real world results.
Dr. K told me once. "David, garbage in, better garbage out using taguchi."
So if these solutions are taguchi based some incremental increase is expected.
But you may not get the optimum results for your efforts. There is a whole method of test design and input generation that is being missed with these "automatic" solutions.
At this writing, I spend two hours a week answering questions. Not about the software. It is a piece of cake to use. MR2 is well documented with online demo and complete instructions. These questions are about what to test and how to best setup experiments so that real results are obtained as quickly as possible.
Further, the MR2 software was created not only for testing online but also for offline applications.
So there is a difference in what the products are able to optimize. And the scope of their design.
Remember, any of these solutions are just tools. In the hands of someone who truly understands his tools, a masterpiece can be created. In the hands, of another the same tool is not nearly as effective.
Take Care,
David Bullock
Sr. Advertising Optimization Specialist
www.results-squared.com
The real question: Is this new or just new to the online-internet-marketing marketplace? I see a mad scramble to create products that are "new" tools for the online marketer.
In the last 5 weeks, I have seen at least 4 other "taguchi based" products enter the market. I had my call with Perry Marshall on 8/27/04. And the words "success factors", "taguchi" and "multivariate" are now part of everyday online internet marketing language.
This is great because that tells me that there is a real need in the marketplace for better tools to increase MROI (marketing return on investment).
Many of the products claim to be better than taguchi, automatic, self-tweaking, work while you sleep, not a black box, easy to use and set it and forget it solutions.
I have been extremely busy working with clients and projects. So, I really didn't have the time to sit and write anything to address all the questions that I am getting.
So, I will attempt to answer the major questions here.
Here we go:
Q1) Are these other solutions Taguchi?
A1) They may be. But, I haven't checked the code to see what they are doing. So, I can't say for sure. Any text book has the standard formulas in the back. Any programmer, if he takes his time can copy the formulas and create a program. But, if you don't understand what those formulas are and how they work and the implications of the data, it is just copying formulas. There is more behind taguchi than the formulas.
Q2) For these other products that say they are "better-than-taguchi". Is that a true statement?
A2) Better is a relative statement. If the program achieves the results that you want within your budget, then those solutions may be better for you. Some companies are more than happy with an increase of 25% or 30%. Now that there is the possibility of higher increases - everyone wants 1x, 2x, 4x increases.
"Better than the taguchi methodology", I truly hope that no one is promoting this as a fact. The methodology has been around for too long and working too well to all of the sudden just be wrong.
I would not ever say that I am even close to inventing something "better-than-taguchi" This method is over 40 years old and is too well proven.
I don't think that people are inventing something "better" per se. Maybe a better platform for use in a specific area; is a better way to say it.
Q3) Why is your software so much more expensive as compared to the other solutions out there that sell between $200 and $500.
A3) I can only answer this one way. I did not just go get a textbook and open it up to the back and claim to be a taguchi expert. I am an engineer and salesperson with real world in the field experience. And you are not only getting the software you are getting my help and support throughout your experiments, plus a users forum, plus coaching calls, plus access to any of the tools that I am using to run my experiments.
But the key here is that they know what is going on with their market, message and media. They know how to interpret the results that they are getting and not just going on blind faith that the software is all powerful.
Further, they are building on that learning to produce increase in other areas of their market strategy.
Remember, any of these solutions are just tools. In the hands of someone who truly understands his tools, a masterpiece can be created. In the hands, of another the tool is not nearly as effective.
Q4) Some of these solutions offer that they can test an infinite amount of variables. Is that true?
A4) Again...I haven't checked the algorithms to see what they are using "under the hood". But, I will say this. I was in conversation with Dr. Rayjit Roy @ http://www.rkroy.com. Many of you have purchased his books about taguchi methodology.
Anyway, I was in the process of designed some very specialized arrays. And I called him to see if his software could handle my design which was more than 4 options per factor. He told me that does not exist within the taguchi methodology. So when I hear that someone is promoting that they can test infinite options.
I speculate that they may be doing DOE (Design of experiment) but you are not doing taguchi.
Remember, there are very few people in the world who are qualified to modify the taguchi methodology. Dr. Taguchi is one of them. Dr. Kowalick is another. All the others are interpreting the methodology in ways that may not be fully grounded with case studies and real world results.
Dr. K told me once. "David, garbage in, better garbage out using taguchi."
So if these solutions are taguchi based some incremental increase is expected.
But you may not get the optimum results for your efforts. There is a whole method of test design and input generation that is being missed with these "automatic" solutions.
At this writing, I spend two hours a week answering questions. Not about the software. It is a piece of cake to use. MR2 is well documented with online demo and complete instructions. These questions are about what to test and how to best setup experiments so that real results are obtained as quickly as possible.
Further, the MR2 software was created not only for testing online but also for offline applications.
So there is a difference in what the products are able to optimize. And the scope of their design.
Remember, any of these solutions are just tools. In the hands of someone who truly understands his tools, a masterpiece can be created. In the hands, of another the same tool is not nearly as effective.
Take Care,
David Bullock
Sr. Advertising Optimization Specialist
www.results-squared.com
0 Comments:
Post a Comment
<< Home