A recent Forbes.com article shared that CMOs placed “data scientists” at the top of their 2015 holiday wish list. Guest contributor Scott Davis stated, “This year in particular, CMOs are struggling with tons of data that our new digital world is creating. They want to know how to leverage the data in cool (not creepy) ways that drive commercial value.”
The popularity of data scientist as a profession has grown in leaps and bounds. According to job search site Indeed.com, searches on “data scientist” has grown a startling 20,000% in the last 5 years. With the amount of data increasing exponentially and effective usage being paramount to business success– one figure states that a typical Fortune 1000 company increasing data accessibility even 10% can see more than $65 million in additional net income– it’s no surprise that CMOs are eager to get their hands on data science talent.
But like the pandemonium we have come to expect at a Black Friday sale, how much of this interest is just a feverish reaction to the current state versus an objective appraisal of actual need? How much are marketers adopting a “me too” mentality without a fair assessment of actual business benefits and organizational proficiency?
To be clear: this is not to suggest that marketers do not need to leverage data. The benefits of integrating data into the marketing function are established and manifold, including increased speed in engagement, more precise customer targeting, and better campaign outcomes. But the question becomes: do CMOs really need data scientists within their organizations or should they consider other data science solutions that might yield similar or better results?
Here are three reasons why CMOs might want to consider data science solutions instead of seeking in-house data scientists:
1) Data scientists are hard to hire, so CMOs simply may not be able to hire them for their teams
In 2012, Gartner made an often-quoted statement about the expected dearth in data science talent. Research VP Doug Laney asserted that the need for data scientists was “growing at about 3x those for statisticians and BI analysts” and that there is an “anticipated 100,000+ person analytic talent shortage through 2020.”
In addition being rare and in high demand, data scientists are the classic “purple squirrel” — a rare combination of academics, skill and experience that recruiters typically struggle to locate. Additionally, the true value of a good data scientist is best summarized in the title of a 2015 Harvard Business Review piece: “The Best Data Scientists Know How to Tell Stories.” Author Michael Li stresses the dualities that should be assessed when adding data science talent– that individuals possess not only computational, technical and statistical skills but also the ability to align data with business metrics and “are good at communicating the story behind the data.”
Bottom line: your organization may struggle to secure the data science talent that you believe you need. If you are facing a long recruiting timeline, you may want to start thinking about alternatives, such as arming your existing workforce with the knowledge and resources to make sense of billowing marketing data– either in the interim or as a suitable longer term option.
2) CMOs don’t know what they want from the data or don’t know how to articulate business needs to data scientists
We recently co-hosted a series of dinners with the CMO Council in cities across the US, speaking directly with marketing leaders from top business products and services companies about their greatest challenges. A marketing leader from a Seattle-based business applications company admitted that while they have a dedicated team of data scientists, the team provides almost too much information and data. The marketing team must now focus on simplifying process and focusing on the key measures of success, including what works and what does not work so that they know what they can eliminate.
This complements feedback that we collected in a survey commissioned through Forrester Consulting. Sixty-seven percent (67%) of marketing leaders agreed that the problem was not lack of data but rather being able to draw insight from the data. The key idea appears to be “actionable”– as in “actionable insights.” Marketers want clear and immediate guidance on what they should do and the responsibility falls on both marketers and data scientists to determine the metrics and activities most worth addressing to drive business benefits.
Bottom line: marketers must ensure that there is clear alignment between marketing and data science resources to ensure that efforts and activities are generating usable, actionable and thus highly valuable business insights. Marketers must assess the process maturity and bandwidth of their team in selecting from data science options. While additions should improve efficiencies and outcomes, wrong-fit options can have the opposite result– increasing friction and detracting from existing resources and process efficiencies.
3) CMOs are handing data scientists “dirty data”
Dirty data can happen for a number of reasons, including data source, duplication, and decay. In the Forrester Consulting survey, B2B marketing leaders reported that the top sources of business data were manual inputs from sales reps (73%), list purchases (68%) and manual inputs from contact forms (55%)– data sources which are susceptible to errors.
Handing dirty data over to data scientists is tantamount to passing rotten ingredients to a chef and expecting that he/she transform the inputs into a gastronomical masterpiece. In both instances, the quality of the inputs impacts not only the quality of the outcome, it also impacts the experience and efficiency of the professional– how much time can be spent experimenting and applying the artistry for which the professional was hired versus overcoming hurdles to get to a sufficient baseline.
Bottom line: the quality and state of your internal data can impact– and even worse, impede– the ability of even the most talented data scientist to generate breakthrough ideas. Many turnkey data solutions can help you maintain data, even enhancing accuracy and comprehensiveness, in addition to extracting insights. It’s not simply a means of “killing two birds with one stone”; accurate and complete data is a critical first step. In other words– and without being too macabre– good data is the essential and necessary “first kill.”
For the three reasons that have been detailed in this piece– challenges in hiring data scientists, marketing process maturity conflict, and internal data quality issues– CMOs need to take a step back from the chaos and make an honest assessment of needs, internal capabilities and expected outcomes. While data science is a vital addition to any marketing strategy, it’s important to note that there is a wide spectrum of options that slot within this category. Whether it’s bringing data scientists in-house or integrating a turnkey data science-based solution from the wealth of options available in the market– from campaign analytics to lead scoring to predictive analytics solutions– CMOs need to calmly assess needs and options to make calm and level-headed decision. Falling into the “big data” quagmire, like with impulsive holiday shopping, may cause a CMO to wake up the next day, scratch his or her head, and wonder, “What was I thinking?”
