Oh, you haven’t heard? Siri decided to get her MBA from Wharton, Echo is attending marketing classes at the local community college, and Einstein dumped relativity to start dabbling in email deliverability for Salesforce. According to The Boston Consulting Group’s report, “Man and Machine in Industry 4.0,” up to 25% of current jobs will be replaced by adaptive software or robotics. While advances in AI and machine learning tend to provoke one of two reactions—a general dystopian anxiety or vision of a gleaming connected futurist society—the fundamental truth is that we are entering a major disruptive cycle in many verticals, but particularly within the marketing landscape.
In marketing, we’ve reached an inflection point where the volume of data and processes required to run personalization initiatives at scale have outstripped human capability (even highly skilled ones). A 2016 CMO Council study cited that only 3% of marketers say their current data sources are aligned with delivering a comprehensive view of their customer. The micro-targeted customer experience is simply unsustainable with a traditional technology stack leveraging manual data modeling and human-curated, rules-based decisioning.
However, unlike many industries that must consider and respond to the prospect of jobs loss, marketing leaders need to recalibrate their thinking on how these technologies will disrupt, transform, and ultimately, enrich their work life.
First, let’s take a look at how a modern marketer spends their time.
Moving Advanced Marketers Back Up the Intellectual Pyramid
As marketing automation technology has advanced, the role of the CMO has evolved from strategist to technologist and statistician—leaving little, if any time for the creativity that sparked interest for a career in the field. Likewise, other highly paid and highly skilled analytics and data science team members are spending much of their time doing the menial tasks of pulling, organizing, and packaging target audiences and manually designing and managing marketing experiments.
In fact, get up and go ask a senior digital analyst for their largest complaint.
Go ahead. I’ll wait.
Besides parking, the primary complaint is nearly always related to data accessibility, data quality, and the manual pain of manipulating it. It’s not without reason. From Fortune 500 enterprises to smaller marketing teams, senior analysts, marketers, and even data scientists are spending much of their time pulling reports versus the advanced analysis, marketing campaign strategy, or deep-dive insights that they’re paid to do. Data scientists, according to New York Times estimates, spend from 50% to 80% of their time mired in the mundane labor of collecting and preparing unruly digital data. Likewise, a recent Forbes article put the number at 80% based on survey results.
It’s repetitive, uncreative work done by highly paid, highly skilled people.
And, it stands to reason that the same will happen in marketing that has happened in many other major industries before us—manufacturing, retail, etc.—repetitive jobs have largely been automated by technology. Currently, we have made massive strides in ingesting, processing, and activating the sheer volume of customer data across the modern enterprise. However, it’s not enough just to take action. In the age of the connected consumer, brands have to take intelligent action at a personal, 1:1 level to stay competitive (Think Uber, Netflix, and Google).
While the prevailing thinking around automation focuses on job replacement, we’ve seen early adopting brands experience a massive scale-up and creativity surge on their teams, while neither contracting nor expanding the human team.
It’s a multiplier, not a replacement, of about 1,000x.
When the focus shifts away from labor-intensive segmentation and modeling practices, consumer engagement has a chance to become much more timely, targeted, and impactful.
Some trends we’re seeing from leading machine-learning technologies:
- Simultaneous management of thousands of experiments and tests, and exploration of two to the hundredth power of distinct interaction contexts (that’s longer than the age of the Universe in seconds!) allows for optimization that effects true personalization of the customer experience and increase in ARPU
- The capability to produce automatically derived insights into what is working and what is not, and determination of the true causation of outcomes, and quantification of those insights communicated intuitively to the marketer
- Dynamic adjustment of models to ensure ongoing optimization of personalized marketing interactions to drive sustainable results
- Increased freedom for marketing teams on strategy development and new creative ideas to test for impact, rather than operational tasks
Ultimately, scalable machine learning technologies are able to move the senior marketers and data scientists back up the intellectual pyramid—where they belong.
How to Recruit and Think About Your New Machine Hire
To truly realize transformative value of machine-learning and digital intelligence platforms throughout your organization, you need to understand how it intersects with both your organizational chart and your legacy tech. How would true automation evolve the current roles on my team? Do we have the right skill set to build new strategies to take advantage of the multiplying effect of thousands of experiments and tests?
With automation ultimately taking over a portion of the menial tasks that occupied the time of your marketers, analysts, and data scientists, these teams will need to evolve their focus to realize true organizational ROI from machine learning adoption. It might be time to set the bar higher on campaign strategy and deep-dive insights from your teams.
Similar to your people, it’s critical to understand how your new tool interacts with your existing data environments and campaign management tools. Within your data environments, do you have access to your raw customer behavior and attribute data? What organizational or technology barriers will you need to account for?
Likewise, on the campaign execution side, will your new machine hire work with your existing campaign management tools? In Forrester’s Q1 2016 Digital Experience Survey, technology integrations were cited as one of the largest hurdles to success for marketers. No matter what technology you’re using, this is a key question to ask as you evaluate machine learning for your 2017 technology roadmap.
Lastly, if you aren’t assessing some form of machine learning and adaptive experimentation implementation for 2017, the only marketing leadership job that’s at risk might be your own.