« Pepsi “Tava” fails initial web test | Main | MyStarbucksIdea.com – patronizing content… or really good idea? »

Can I really believe that research?

My recent article Web 2.0 fails the grade, according to executives, republished on www.Ragan.com (membership required) drew a string of visceral responses. Most couldn’t agree more with my assertion that “Web 2.0 does not deliver the ROI, does not live up the hype, and is not even close to being a top priority for senior management (not all, but most).”

Supporting my assertion is a recent CIO study, Top Technology Priorities for 2008, that reveals that even techies don’t consider Web 2.0 as a priority. A couple of people disagree including a colleague of mine, Shel Holtz, who points to contrary research from CIO and Forrester.

Research is a big business. And there are so many forms and so many formats that it is in fact quite easy to develop a research study that produces just about whatever results you want. Genuine researchers, and there are many, will wince at the suggestion and leap to defend their industry and practices. They are right most of th time and their defense is well-based as most studies are astutely conducted with no smoke and mirrors, but understanding the research source, type and sample or target audience is the most critical part to understanding the study and its conclusions.

Let me come back to the study sample and firstly look at a couple of types of surveys. Of the many different types of surveys or polls there are two in particular that require close attention:

• sponsored versus non-sponsored
• self-select and random sample

A sponsored survey is one that some company has sponsored or ‘paid’ for and quite often (but not always) has determined the subject matter being studied if not suggested or reviewed the actual questions and answers. These studies are typically sponsored by an organization looking to boost their PR or sales and are sometimes found to be biased with questioning and analysis that is skewed to support a conclusion that is favorable to the sponsoring company that typically is trying to sell sell you something.

A self-select survey is a purely voluntary survey versus a random sample which, if properly conducted, is the most statistically sound and representative of the studied target population. A simple random sample is a subset of (a sample) chosen from a larger set or defined population. Each individual surveyed is randomly chosen so that each individual has the same probability of being chosen.

Most web surveys are in fact self-select – the kind that anyone can respond too and often have small sample sizes and can be skewed by the strongly engaged and highly opinionated. The results of most of these studies should be taken with a grain of salt – or a bag full of it.

Here’s an example of a highly skewed study from a story I wrote in 2005:

A Guidewire Group Market Cycle Survey, “Blogging in the Enterprise” finds that 53% of respondent companies are already blogging and an additional 35% of respondents plan to begin corporate blogs within the next year.

While the survey is not a scientifically significant sample and should be taken with a great grain of salt (a very small sample of 140 respondents and the survey was self-select (voluntary) of tech magazine readers), the Guidewire study does provide some excellent insight into corporate blogging.

Does anyone in their right mind believe that 53% of companies were blogging in 2005? Hell the current level is no where near 50%. It’s no surprise to learn that the study was conducted by a company selling blogging software and the study concludes with an advertisement for the sponsoring company.

But wait, a lot of people are blogging, right? And there are plenty of case studies of corporate blogs…

• and books
• and trade shows
• and sites dedicated to blogging
• and blogs about blogs

Numbers from some of the research firms such as Forrester reveal that in fact that perhaps even more than 75% of the Fortune 500 have adopted some form of blogging (whether real, piloted or once-tried-but-now-abandoned). A lot of those blogs don’t amount to much, but they have them.

But what percentage of the total number of business are represented by the Fortune 500? Answer: Less than 0.3% (see Entrepeneur.com).

Here’s the important point about sample type: just because 75% of a group of 150 or 200 highly innovative CIOs with massive IT budgets have tested or experimented or evaluated blogging software doesn’t mean that 75% of businesses are blogging. The vast majority of for-profit employers in corporate America, Europe and the world-over are small and medium-sized companies and, believe me, they don't get and understand or embrace Web 2.0 (don't even get me started on government).

I belong to a group of CEOs (called Vistage the largest group of CEOs in the world) and believe me, a massive majority have no clue what Web 2.0 is. Oh sure, the concept is growing and catching-on by leaps and bounds, and more and more executives are catching-on every day. But these trail-blazer executives (or rather slow-to-convert luddites) represent a very small majority amongst the executive ranks. It's upwards and onwards, but there's a lot of education and promotion of Web 2.0 still needed to educate the board rooms of the World.

Finally, I have to say that I am one of the biggest converts of social media. Not only are you readnig this on a blog, but I work with just about every type of social media tool either for myself or with clients or at home in my leisure time. I'm not trying to "rain on the parade" but I am trying to provide a accurate reflection of the real world. In other words, just because your company doesn't have a blog, doesn't mean you should panic. On the contrary: take your time and do it right. But don't look to senior executives to come beating down your door demanding you launch a motherload of wikis for customers. They have other priorities and need to be educated and sold as to the value of Web 2.0.

As your mother told you: “Don’t believe everything you read.” More importantly, do read every study that pertains to your work and your business that you can get your hands on. But be sure to look at the survey type, methodology and sample and fully understand the context of the study before drinking the Kool-Aid.

Post a comment

In order to reduce spam, please enter the letter "f" in the field below:

About

This page contains a single entry from the blog posted on March 21, 2008 12:36 AM.

The previous post in this blog was Pepsi “Tava” fails initial web test.

The next post in this blog is MyStarbucksIdea.com – patronizing content… or really good idea?.

Many more can be found on the main index page or by looking through the archives.

Powered by
Movable Type 3.33