Keith Quesenberry

Does .005% Make A Difference? Ask Toyota

~ K. Quesenberry ~ Leave a comment

Manufactured products are subject to defects, they occur at random. We know that the defect rate can never be zero so a higher defect rate doesn’t cause problems if you are purchasing the product in large quantities (large sample size). If I buy a pack of 100 straws for my kids, I am ok with some having holes in them – even up to 10 (10% defect rate). I bought them at the dollar store after all! As consumers we access the true cost of the product by including the defect rate in purchase.

Problems occur when the buyer purchases the product in small quantities (small sample size). We typically purchase cars one at a time. A manufacturer may sell cars with a 1% defect rate, but we buy one car and it either works or it doesn’t. When I buy my Prius and discover a brake defect, suddenly that 10% defect rate is not acceptable!

A 1% defect rate may not be statistically significant over the roughly 2 million cars sold by Toyota in the United States every year, but ask a Toyota marketer if it is practically significant. One percent is 20,000 unsatisfied complaining consumers. An article was written about the defective brakes on Prius Hybrids based on 100 complaints – only .005%! Toyota can improve their defect rate, but never get it to zero. That is when Public Relations becomes invaluable.

Online Research: Temptations and Limitations

~ K. Quesenberry ~ 3 Comments

Internet research is quickly gaining popularity as a marketing research method. In fact, growth of research via the web has increased six times the rate of total market research. The advantages of Internet research include real-time reporting, reduced costs, personalization, higher response rates, and the ability to contact hard-to-reach targets. But what are the disadvantages?

A recent American Psychology article detailed several disadvantages of online surveys. We must be careful of sample bias that effects the generalizability of your data. Unlike random dialing of telephone numbers, nothing exists to give you a random sample of Internet users. And people who can participate in online studies tend to skew towards certain demographic profiles other than the general population. In addition, response rates for online surveys versus paper are generally lower. Online participants may drop out before completing a study and there is often difficulty in getting a hold of them at a later time. This is because email addresses change more frequently than phone numbers and mailing addresses. Online also runs the risks of multiple submissions or flooding a site and people may act differently online than they do in real-life social interactions. Also are participants who they say they are? With an in-person study, you can see that someone is who he or she says they are.

Another fact to consider is that the vast majority of online research is quantitative with companies using applications like SurveyMonkey and Zoomerang. There are limitations in conducting qualitative research online – how do you get someone to taste your product or what if you need to observe their behavior? Internet research is gaining ground and may be attractive because of its immediacy and lower cost, but it is good for us all to also keep in mind its limitations.