Open Enrollment | Subscribe to Target Marketing HERE
Follow us on

Data Driven : 'Know Thy Customer'

3 tips for making the most of your research data

January 2013 By Geoff Wolf
Get the Flash Player to see this rotator.

We are in the midst of a true paradigm shift. Consumer behavior suggests consumers want to be more in control of their interaction with the marketing that brands are pushing out. As marketers collect more data and gain access to more powerful analytic technologies, there also is a new dimension we need to access to understand consumer behavior.

Database practices focused on what customers are feeling and sharing about their favorite brands must now be added to our ongoing data routines. Research, ironically an old marketing tactic, provides data to address this new change in behavior.

Research data does not typically have a customer record identifier and, therefore, exists as an external tool. Even so, industry standard data rules and practices apply here. There must be enough data to act on, integrity in the manner it is collected, and segmentation included. Here are three tips to make the most of research data:

1. Use Surveys
If you're going to execute a survey program in-house, be sure to offer an entry into a drawing or some other incentive to encourage participation. Avoid having too many open-ended questions by using a numerical matrix for answering most questions. A ranking grid also enables visibility of how the combined answers change over time. The average may be 2.3 one time and then 2.9 the next time, indicating a directional pattern.

Also, include a question or two to provide segmentation visibility. For instance, by asking if survey participants place their orders by phone or using the Web, you can have media channel segmentation in place to learn how differently media channel users answer other questions.

There are three easy ways to execute in-house surveys:

• Package stuffers in your outgoing product shipments. These tend to be focused on customer service questions, but you can add marketing content, as well. These surveys reach all your customers and the data tends to be actionable across a more comprehensive portion of your customer base. Therefore, segmentation questions are very appropriate here.

• A survey at the end of your shopping cart using a service like Survey Monkey. Keep in mind this only addresses online shoppers, so both the questions asked and the resulting actions must be targeted to this segment.

• Surveys sent via email. Email deployment of surveys can be a quarterly staple of your email program. Like the online survey, this will only reach online customers and specifically those who open your emails. Keep this niche target in mind as you develop and act on the surveys.

2. Go With the Pros
Professional research offers a more in-depth analysis. While there is a more significant cost involved here, the results are much more reliable for a couple of reasons.

First of all, there is a very specific way to script survey questions. Part of what you are paying for is the training and experience that is necessary to use the proper questions. Questions that lead responders in any direction are risky. The second reason is that professional services offer a "third party" perception that can provide more honest data. A third upside to the expense is that it can go beyond your customer base to learn about consumers who are not your customers yet.

The larger the marketing investment you intend to make using research data, the wiser it is to make a different investment to make sure the data is right.

In addition to traditional research resources, there are new opportunities available now that leverage social media, like Facebook, to execute surveys. Consumers are spending so much time on these sites that it makes a lot of sense to include surveys here. These are very scalable, provide reliable data and are actionable immediately. Web Publicity has an exciting program targeting exactly this opportunity (

3. Open Up to Feedback
There are many opportunities to capture what is "open" feedback from customers and consumers in your markets. Analyzing these types of data sets is usually more of an art than a database science. An informal process will offer a glimpse of real consumers speaking to each other about your products and brand. If repetitive content starts showing up, we make notes and address these issues in more scientific surveys.

A more advanced opportunity is to use some custom programming and query tools to analyze large sets of "open" data, looking for words and phrases that are repeated over and over in a specific context. Some examples of this communication are gift card messaging, social media ongoing conversations in relevant community forums and Internet chat arenas, and customer service comments made while on the phone to the extent you collect these.

Perhaps certain conversations generate distinct phrases that indicate consumer versus business gift needs. Perhaps moms, dads or grandchildren show up more often. A specific adjective or verb repeating itself may indicate a value or benefit you have not thought about. It might turn out that your products are often used for get well thoughts, or moving occasions, or support of some type. Try a little research into conversations you have access to and see what shows up. You may uncover some learning about your products and brand you have not thought about before.

Basic database rules and practices apply when using research in your marketing programs. Be sure to apply the same standards and rigor you have in place for your customer database here as well. I would venture a guess that consumer behavior will only continue to demand that research becomes a larger part of industry data practices in the future. I recommend staying ahead of the curve on this marketing reality by getting a comprehensive research data program in place now.

Geoff Wolf is executive vice president of client strategy at the Mission, Kan. direct marketing agency J. Schmid & Associates. Reach him at



Click here to leave a comment...
Comment *
Most Recent Comments: