Earlier this week in Boston, we ran a small, intimate workshop for about 20 of our 1to1 Media “insiders.” These are folks who regularly access our Web site and Webinars, subscribe to the 1to1 Magazine or to our new journal, and so forth. We hand-picked the participants from our opt-in database, in order to ensure that the room would be filled with people who had their “fingers on the trigger” of analytics at their companies. What we wanted most were those folks who were wrestling with the problem of starting, upgrading, or just managing their companies’ customer analytics functions.
It’s clear now, from our vantage point almost one tenth the way through the 21st Century, that numbers and analytics of all kinds will play a more and more important role in our existence. If you haven’t yet read the book Super Crunchers, by Ian Ayres, you owe it to yourself to get it and learn why numbers have become more dominant in nearly every arena of life. And there are a number of other books out there chronicling the same trend toward ever more useful and pervasive quantitative analysis.
Nevertheless, the typical business still can’t get its act together when it comes to using numbers for anything other than financial reporting.
Very few firms use statistical analysis to try to generate insights about the past and guess more accurately about the future, despite the fact that many companies are now sitting on extremely large quantities of data. So one of our goals at this workshop was to do a better job of understanding why companies are holding back. What is it that prevents firms from taking a more data-driven approach to their business?
As Ginger Conlon, our editor in chief, explains in her post on the Think Customers blog, our participants identified several issues they considered important, but at one level or another all of these issues tended to derive from poor organizational alignment. The issues identified by our workshop participants as those they were most concerned with were:
- Having and leveraging the right data for such high-value activities as understanding potential wallet size or determining future value
- Ability to gain credibility for the ROI measurements, especially for the softer benefits
- Data cleanliness and accuracy issues. Including the ability to integrate customer data from multiple silos
- Silos of conflicting responsibilities, and lack of accountability as a result
During the workshop, we organized ourselves into table groups to address each of these issues independently, but we came up with common-sense solutions that were remarkably similar.
One of the key mechanisms a company can use to address these obstacles is compensation – both individual and organizational compensation. At our own table, we had one participant from a large, well-known mobile phone carrier, and his problem was that even though they often had data that could be used to sell more things to individual customers, the sales force was rarely interested in these insights. Usually this was because the commission and bonus structure didn’t reward data-driven selling. If you pay bonuses based on the volume of new-customer acquisitions, for instance, then you risk having sales people who are unwavering in their attention to that goal, to the exclusion of all other objectives. Commissions and bonuses can create a marvelous engine for growth, but if you don’t align your compensation structure with your customer analytics insights, then you’re still not going to get any use from those insights.
Another participant at our table was from a mid-sized online business whose customer service center was run in Canada, and had proved particularly unresponsive to marketing’s constant requests for better customer service processes. Why? Because at this company the call center is treated as a cost center, not as a “satisfaction center.” Their budget and profit targets are set almost exclusively with an eye on cost-minimization, so every single initiative is evaluated against that yardstick. Anything that adds cost is bad. Revenue and satisfaction improvements don’t count, because they aren’t included among the evaluation criteria. The folks at the service center resisted service improvements and data-driven initiatives because the service center itself would not benefit, and would in fact be penalized by the additional costs required to execute these initiatives. So this is an example of poor “organizational” compensation.
Managing with numbers, while it sounds like the logical thing to do, is obviously more difficult than most of us realize. In the final analysis, we concluded that two important things had to occur in order for a business to embrace analytics.
- Compensation structure must further this goal. And we’re not just talking about individual salaries, bonuses, and commissions, but organizational compensation, as well – capital budgets, headcounts, and so forth.
- In addition, managers need to be engaged in the “manage by the numbers” mission, personally. They have to have personal confidence in the wisdom of data-driven decision making, rather than always trusting to instinct and judgment. Instinct and judgment are important, but in the 21st Century they must be supplemented by real numbers.