online marketing research

I work from home most days, but this weekend I was able to get out of the house and into the Chicago social scene. The main thing I noticed was how much more I can learn from other people just by listening and putting in some face time. As much as I love reading my blogs, there’s nothing that sparks ideas like interaction with people offline.

For example, I spoke at SocialDev Camp this weekend and my co-panelist Marty was poking fun at traditional marketing segmentation. His point was that it seems like every B2C company starts off with the same segmentation criteria:

• Female head-of-household
• Age 25-54
• HHI (household income) $75,000+
• [insert more ridiculously broad data here]

This worked so well in traditional marketing, yet when you bring it online to a site like Twitter, the system crashes because much of the information companies want to segment on just isn’t available. It got me thinking beyond traditional targeting about the different ways you can group people and organizations of people:

• Demographically – age, gender, income, education
• Geographically – address, zip code, area code, neighborhood
• Psychographically – activities, interests, opinions, attitudes, values
• Firmographically – industry, size of company, revenue
• Behaviorally – site usage, click-through rates, purchase history

It seems to me that traditional market segmentation looks at just a few of these ways to segment and those don’t translate well online. Companies focus on the demographics and geography of a person, when most of social media makes it easier to classify people by behavior and psychographics.

It makes sense – in the past, it’s been much easier to classify people by known information, and a few psychographic variables. Look at the published advertising kits for a popular magazine like Seventeen and you will find mostly demographic information. The psychographic information is lacking, unless you are selling clothes, shoes, or beauty products.

But what if you want to sell other items of interest to teen girls, such as yoga mats, guitars, bedspreads, laptops, college essay services, video games, or whole grain cereal? Are American teenage girls really as shallow as we paint them?

It made me wonder – is there a need to better understand our customers in order to find them online? The beauty of social media seems to be in the rich psychographic and behavioral data that’s become available to us in recent years, and my intuition is these are far better predictors of a target segment than demographics and geography alone.

This is illustrated easily by Facebook ads targeting. Sure, it lets you target with the same old demographics and geographics (minus HHI), but it also gives you some unique psychographic options, like keywords, relationship targeting, workplace targeting, sexual orientation targeting, and page, event, group, or application targeting. These modest differences show that we are more than just data points and there is a need for a more detailed level of segmentation than traditional media wants to provide.

I want companies to think beyond the norm:

• What company does someone work for? All companies have culture. If someone has worked at a company for awhile, what does that say about them?
• If someone plays Mob Wars on Facebook, what does that say about the type of products they might buy? There is the obvious (video games, The Godfather collector’s edition), and then there are the non-obvious that we still don’t understand much about.
• How can we get beyond demographic segmentation to really understand generations? There is a huge need amongst marketers to understand how different segments of Gen Y (high school, college, post-college, parent) is using social media to interact with brands – why isn’t anyone studying this in-depth through the APIs of Facebook, Twitter, and Google? Imagine what that report would be worth.
• How can we use non-intrusive means to understand individuals and make connections between online behavior and purchase decision? Amazon’s recommendation engine is amazing at finding insight that humans can’t.

It also seems like a great time for small companies to double or triple their business, because so many of them target their audience based on psychographic and behavioral data, rather than broad demographic data. Like the bicycle repair shop down the street – that guy could be doing awesome on social media. Why isn’t he?