What Data Buying Isn't

Data-Driven Thinking"Data-Driven Thinking" is a column written by members of the media community and containing fresh ideas on the digital revolution in media.

Today's column is written by Michael Katz, CEO of interclick, a Yahoo! company.

With all the great data that data providers are making accessible today, it's possible to interact with consumers in ways that were never before possible. The tremendous breadth and depth of available data moves consumer views from one-dimensional to multi-dimensional, helping to paint a much more complete picture. The implications positively impact the entire value chain from the marketer all the way to the consumer. For all the progress however, we’re still very early on and there are still several misconceptions about the successful application of data.

One of the biggest misconceptions is that optimal data consumption is on an "as needed" basis since data is expensive and more data may not add incremental value. Utilizing data for targeting is only one of many applications however. One of the most important and innovative applications of data isn’t for targeting at all but rather enabling marketers to implement more effective customer segmentation strategies.

Many transaction-centric B2C companies rely on effective segmentation to align messaging with business objectives in order to maximize LTV (lifetime value). Successful implementation of segmentation helps these companies define business models, build customer loyalty programs, and further value proposition discussions. The exercise of creating an effective segmentation strategy should result in a comprehensive understanding of the various types of customers as well a coherent plan for achieving business results.

When helping marketers address their segmentation strategies, the first step is to understand what the objective of the marketer is. Typically objectives are either tactical or strategic. Tactical segmentation strategies typically encompass cross-selling and upselling opportunities. For a financial services client for example, it may involve messaging certain “gold” card members in order to move them to “platinum” card status which may yield much higher fees. Strategic initiatives are usually much broader and align to business objectives. For example, determining that users within certain segments may require less support than others would allow a company to deploy capital more effectively.

The next step is to construct the segments. This is where the tremendous amount of data available makes all the difference. In years past, marketers relied on single data types or sources for segment construction which typically yielded no additional insight or lift in performance. Today’s data landscape allows us to combine diverse data sets such as transactional data, intent data, attitudinal data, social data, occupational data, and even survey response data to gain a much more complete understanding of the users that make up marketer segments. As the knowledge of these consumer segments becomes deeper, we see greater accuracy within the models, not to mention greater flexibility. This results in a smaller error rate and allows for much greater performance and scalability.   When you can begin to close the loop with incorporating revenue and profit data, the conversations between the CMO and CFO become a lot easier.

As new methods of data capture from innovative companies are deployed, we’ll continue to see more and more great data come to market. With an expected influx of massive amounts of rich data, it will be important to continually refine models and segmentation strategies. Developing operational discipline around building and maintaining well-informed segmentation will become more and more critical to marketers as fragmentation continues. I envision a tremendous amount of innovation in and around this area from both the buy- and sell-side in the coming years.

Follow Michael Katz (@mkatz_ic), interclick (@interclick) and AdExchanger.com (@adexchanger.com) on Twitter.


  1. Anyone who is not thinking like this about the many uses of audience-related data is already behind the curve. Everyone in the space needs be trying to figure out how to distill the data available into the data that's necessary and looking for multiple value stories. Right on point MK.

  2. From all the data sets mentioned, I would assume that the transactional data and intent data are the most valuable for targeting purposes. Such data is focused, short term and can provide the highest impact as compared to generic profile level data like attitudes, occupations, etc which many times are derived and highly error prone. Is this what the others are observing as well?

  3. Joseph Fahr

    Concise description that lends itself perfectly to the rise of the "data scientist". Ironically, the rapid increases in data availability has not been matched with an influx of people with the skills to properly handle and analyze so much information. The dynamic has left us with a large gap in need vs. demand of big data handlers. Great for those of us with the skillset, but a detriment to the evolution of our industry as a whole.

  4. Nice post MK. Gourav, transactional data is the end game for many, if it can be afforded and properly modeled.

    In light of, the industry has also directly paired third party data (pre-existing audiences & intent data)with user provided data (answers to specific questions via online survey) to further build upon customer segments and package them in new ways for brands.

    e.g. Customer Segment = moms with children ages 1-3; expressed interest in purchasing a new SUV; would purchase based upon safety first followed by price; are willing to spend no more than $45k

    Any combination of the above variables with intent data becomes increasingly valuable in a number of diverse configurations.


Add a comment

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>