HiPPOs, Presidential Election and Google…Oh My!

I participated in a panel at the recent Techonomy 2012 conference in Tucson, AZ on a subject titled “Data Is The New Science.”  Andy McAfee, who had written an article titled “The Big Data Revolution” for the October 2012 issue of the Harvard Business Review, moderated the panel.  One of the key panel discussion items was the organizational challenge that companies face in adopting big data.  Here’s my perspective on what’s hindering organizational adoption of big data.

Leading In The World of HiPPOs

Andy has coined a term “HiPPO,” which stands for the “Highest-paid Person’s Opinion.[1]”  In a section of his Harvard Business Review article, titled “Muting the HiPPOs,” Andy writes:

One of the most critical aspects of big data is its impact on how decisions are made and who gets to make them. When data are scarce, expensive to obtain, or not available in digital form, it makes sense to let well-placed people make decisions, which they do on the basis of experience they’ve built up and patterns and relationships they’ve observed and internalized. “Intuition” is the label given to this style of inference and decision-making. People state their opinions about what the future holds—what’s going to happen, how well something will work, and so on—and then plan accordingly.

One of the biggest challenges to big data adoption in most organizations starts with the HiPPOs unwillingness to give up control and decision making power to the “numbers.”  HiPPOs are typically senior executives, with many decades of experience, who prefer to rely on their own personal experiences and “intuition” to guide their decisions.  They are the classic old school baseball scouts in the book “Moneyball” who rely on intuition, the “right look,” and a few traditional but outdated metrics to guide their evaluation of baseball talent (note:  I actually saw a couple of older scouts using a stop watch to determine the speed of a pitcher’s fast ball!).

In the world of big data, HiPPOs are the anti-data science and the biggest obstacle to creating a data-driven culture.  Likely the only way these HiPPOs change their mindset is when they start losing to more data-driven competitors.  Think Walmart versus K-mart in the adoption of point-of-sale scanner data and early data warehousing capabilities in the late 1980’s.  To further emphasize the importance of this point, let’s see what we can learn from the recent Presidential election that can help organizations understand how to address the cultural change of becoming a data-driven organization.

Lessons Learned from the Presidential Election

A recent article titled “Inside the Secret World of Quants and Data Crunchers Who Helped Obama Win,” Michael Scherer outlined many of the big data and data science techniques that the Democratic Party used to give Obama a competitive advantage over Mitt Romney and the Republican Party in the recent Presidential election.  As I read the article, the following points about the Obama Campaign’s approach to leveraging big data techniques to create a data-driven culture jumped out at me:

1. Data Acquisition (and Instrumentation).  The Obama campaign instrumented as many customer engagement points as possible in order to gather more data, and ultimately more insights, on their targeted audiences (voters).  As an example, the Obama campaign was very aggressive in their use of social media and mobile apps to create more customer engagement opportunities, especially vis-à-vis the Romney campaign, with the following results[2]:

• Facebook likes:  Obama 33.3M versus Romney 12.0M
• Twitter followers:  Obama 23.2M versus Romney 1.8M (Michelle Obama, at 2.3M followers, had more Twitter followers than Romney)
• Website monthly visitors:  Obama 10.4M versus Romney 3.6M[1]
• Smartphone app: Obama yes, Romney no

Each of these web sites, social media sites, and mobile apps allowed the Obama campaign to capture more data and insights about key voters – their interests, passions, affiliations, and associations – absolutely “must have’s” if you want to improve your customer profiling, segmentation, acquisition, and influencing processes.

To quote Obama campaign manager Jim Messina, “We were going to measure every single thing in this campaign.”

Figure 1: Obama Smartphone App

2. Master Data Management.  A concept that gets over-looked in the big data discussion, yet is critically important, is deploying Master Data Management (MDM) to ensure that the data about your most critical asset – voters in this case – is as complete and accurate as possible.  The Obama campaign “created a single massive system that could merge [voter] information collected from pollsters, fundraisers, field workers, and consumer databases as well as social-media and mobile contacts with the main Democratic voter files in the swing states.”

3. Voter Profiling and Micro-segmentation.  Obama used advanced customer profiling and segmentation techniques to segment their target voters into granular enough groups that specific “treatments” could be applied and their effectiveness measured.  For example, the campaign determined that women between the ages 40 – 49 on the West Coast identified most closely with George Clooney, but identified most closely with Sarah Jessica Parker on the East Coast.  Consequently, fund-raisers targeting women on the West Coast leveraged George Clooney, and on the East Coast leveraged Sarah Jessica Parker.

4. Target Marketing Effectiveness.  The Obama campaign leveraged some of the best direct marketing techniques to “create detailed models of swing-state voters that could be used to increase the effectiveness of everything from phone calls and door knocks to direct mailings and social media.”  They focused their analytic modeling on identifying, monitoring, and optimizing the “critical success factors” in order to prioritize their time and resource investments.

5. Experimentation.  One of the key characteristics of a data-driven culture is their ability to integrate experimentation into all activities in order to gather new data and gain new insights.  The Obama campaign “allowed the number crunchers to run tests predicting which types of people would be persuaded by certain kinds of appeals” and “strategists fashioned tests for specific demographic groups, trying out message scripts that they could then apply.” Each of these experiments yielded new data on their target voters, and new learnings on what worked most effectively in reaching those target voters.

6. Predictive Modeling.  The Obama campaign used all of this data and learnings form experimentation to build more effective and granular predictive models.  “We could [predict] people who were going to give online. We could model people who were going to give through mail. We could model volunteers.” They developed scores for all types of voter behaviors including persuadability, advocacy, volunteering, and giving.

7. Actionable Insights.  In the end, all of this only matters if you are able to operationalize the results and put them into action.  The Obama campaign was successful at identifying and acting on those material and actionable “golden nuggets” of insights.  For example, the Obama campaign was able to identify that their “top performers raised 10 times as much money for the campaign as the underperformers.”