The problem I have seen with most market research is they are not very well done. This is the case for both primary as well as secondary market research.

Biggest reason for this is the people who conduct the market research do not have in-depth understanding of the product and the market. The market researchers in my last company is an example. They work hard and they are very sincere. But they do not have the in-depth understanding of products and market.

That is why I think PM should do their own customer research to understand or at least verify customer needs. Do you agree?

Hi Yuen,
You make very good points in your comment. I agree that PMs should be involved in market research to avoid such shortcomings. For PMs to do own research may or may not be possible - due to other work load. But certainly, they should be involved closely.

- Michael

I agree. Researching what the customer needs forms the foundation of the product. Hence, the data derived from that research better be accurate.

Thanks for the interesting story and the link.

I think the quality of articles in this blog, which was good to start with, is going up. Keep 'em coming, mate.

This post is just plain wrong. We should never ignore market research because without it we will be designing in vacuum. Razr is a lucky break for Moto, but other companies should not read too much into it. It is the exception case, not the normal case. Market research should drive everything. Without it how do you know who your designing for and what their needs are.

I think market research can provide a guide to what "could occur" because it is usually based on a lot of data collected. But, many times the data collected represents a unique group of people in a unique situation and should not be broadly applied to any other groups but the one, or repeat potential misinterpretations.

For example, if I collect market research on the opinions of hand size and the relation to phone size likeability, and the group I collect data happens to be a group of people in a remote area of the world with unusually small hands, I would say I would not find similar opinions in another part of the world where the people have unusually large hands.

I think these days, people are very surprising in what products they find helpful to their everyday lives. It is a new ball game out there when conumers can get online and instantly let everyone know what they think about your product. I don't think traditional market research can keep up with that. But, I hear there are some new players that will take market research into hyper drive - TagWorld.

Anyone who thinks market research should drive everything is a moron, or minimally, have trained themselves not to view things objectively or to think.

What assumptions do you have to make to believe that market research should drive everything?:

a) That it is possible to create a research instrument which is error free in construction (no word nuance errors, the right questions, statistically valid, a good sample, etc.)

b) That there are no mitigating factors which could change research conclusions. For example, as in the case of the RAZR, that personal use, quality design, and building the thinnest phone ever could not trump a couple of trivial design rules that applied to previous phones.

c) That the data are not biased by the simple act of observation (The Hawthorne Effect - if participants know they are being studied, it biases the data http://www.accel-team.com/motivation/hawthorne_02.html). This commonly invalidates focus group research, as does group dynamics.

d) That research subjects are not unduly influenced by the opinions, actions or body language of the researchers ( http://writing.colostate.edu/guides/research/rhet-res/com2e3.cfm ).

e) That the researchers in drawing conclusions select the correct theoretical model to describe behaviors (see Evolutionary Epistemology, http://en.wikipedia.org/wiki/Evolutionary_epistemology specifically the notion that social and methodological processes select the "best fit" to describe empirical evidence, but that they may have either missed a new model, or mis-fit the data.

f) That participants in market research can reliably imagine a future condition that is different from their current understanding of the {world, product, competitive options, etc.}, and therefore offer valid opinions about utility, design, form or function.

g) That there are no external factors which change the validity of a research study (pricing, market shift, politics, legal, environmental, socio-cultural, technological).

I could go on, but needless to say, there are many assumptions made about research data and the studies that produce them which are not necessarily true.

Of course, market research is a valuable input, but its best use is as a guide, and if one consciously goes against the guide, there should be good reasons to do so. In other words, don't apply rules blindly, and use common sense and healthy skepticism and intuition to decide. Remember that most of the biggest companies today exist because their founders were squelched in their efforts to create new products at their then current employeror that went against the flow -- most often their inventions rejected the status quo and conclusions of market research.

Need convincing? -- here's a couple of examples. In 1947, it was estimated that the worldwide need for computers was 12. Hardly a need that would justify a single company investing in it, let alone provide the basis for an entire post-industrial economy. Xerox was formed because IBM, Kodak, GE, and RCA and others all passed on the concept of a copying machine because market research said no one needed that many copies. Or this old saw -- there is no market for small cars in America. Toyota is still laughing about that one. And, oh yes, there's the RAZR. The stories aren't hard to find.

The remark that Sandy made is something I expect accountants, or salespeople or housewives or secretaries to make, but certainly not anyone who is a credible marketer or product developer.

Paul makes several valid points about the notion of the "market research should drive everything." Perhaps it would be better said that "market research should inform the attempt to make better decisions whenever possible."

True innovations that really make a difference are almost by definition things that come from beyond the existing horizon and thus are things that current customers or users can't readily conceive of, much less describe or request in the form of market research.

Apple did the iPod against all odds. The music labels certainly weren't chomping at the bit to encourage more demand for downloadable music; and music downloaders didn't seem to be actively asking for a fabulous new iPod-like device that would free them from their PC's. Yet Apple's innovative form factor, user interface and ease of functionality (coupled with iTunes) single-handedly rejuvenated the entire space for digital downloads and devices. I would think that the other major consumer electronics players (Sony etc.) were doing plenty of research to better tap this market, yet none of them arrived at the iPod.

"market research" and customer input are vital when making incremental improvements to existing product offerings, but this impact is less clear when in comes to pure innovation.

Another great example of ignoring market research is the Aeron chair. Malcolm Gladwell talks about it at length in Blink as well as in this talk from Poptech 2004. Bottom line - market research is good as part of your knowledge (knowing your customer, Commandment #1), but especially when determing whether innovations are really any good, it's at least as likely to be wrong as to be right.