Attack of the Killer Data Scientists: How to Prepare for the Invasion

This blog is inspired from a topic I presented this year at the ObservePoint User Groups in New York and Boston. The audience thought it was pretty cool and I hope you do to.

The last five years of my career have been built on a shift in the way we “do” digital analytics.

I entered this space back when most companies had two groups of analytics people: one group of technical people who installed analytics tools (usually a web dev team), and the marketers who used said tools (usually me).

I was leading the digital analytics team at the American Cancer Society in 2012 when we became early adopters of a tag management tool called Satellite (developed by Search Discovery, now Adobe DTM).  It was love at first sight, and I quickly became the world’s biggest evangelist of this tool. This is not a shameless plug, I really was passionate about it and told anyone who would listen about how it would revolutionize the way we do digital analytics. Eventually, I left my job at ACS to join Search Discovery, where I’ve spent the last four years helping other digital analytics teams harness the power of tag management.

Tag management changed a lot about how we do our jobs, and more importantly the types of skills we need to be successful. Analytics practitioners provide the same value today that we did before tag management, but the way we do our jobs has shifted dramatically.  I believe that data science is bringing the next shift.

The Status Quo

What Makes A Successful Digital Analytics Team

If you read enough job descriptions online for a Digital Analyst, you’ll notice a theme: analysts are expected to solve business problems.  Here are a few examples I pulled this week:

  • “Guide the use of data measurement to drive the strategic roadmap”
  • “Business-savvy and action-oriented strategic leader”
  • “Self-starter with outstanding analytic skills, strong business acumen and the ability to synthesize data into meaningful insights”

And I don’t disagree. In fact, I summarize the sentiment like this:

“Digital analytics teams provide leadership on interpreting data to advance the business.”

I’ve consulted a lot of analytics teams over the years, and although every company is different, they all do four things on a daily basis to achieve this goal:

  1. Collect data – Own the tag manager, or work closely with the dev team.
  2. Operationalize data – Distribute baseline information through dashboards and reports.
  3. Reactively interpret data – React to business questions, such as “Why was revenue down last month?”
  4. Proactively interpret data – Identify opportunities to drive new business, such as “A blogger drove a ton of traffic last week, we should reach out to them about an affiliate opportunity.”  Or, “This list of products have a poor conversion rate from users who view the product detail page, so we should update that content to make it more appealing.”

Just about every task that we consider to be digital analytics falls into one of these four categories (I would even stick A/B testing into #3 or #4 since it involves making a hypothesis, interpreting data, etc).

Getting Good Data

If our goal is to provide leadership on interpreting data to advance the business, then #4 is where we really show our value.  Unfortunately, most of my clients struggle to find time for proactive interpretation because they’re consumed by trying to make sure they’re collecting the right data or are just churning out KPIs to the business.

Items #1 and #2 are just about keeping the lights on. Tagging websites and mobile applications are thankless tasks that are more complex than most people realize. No one is impressed when they are done well, but they get very upset when there is a problem. After all, no job description says, “seeking someone to work through the weekend to figure out why revenue in Adobe Analytics differs from revenue in our accounting tool”!

Change is Coming

The world we operate in today requires us to disproportionately invest our time in data collection and other tasks that offer little business value. But I have a prediction:

The problem of collecting data is going to be resolved

It won’t be tomorrow and it won’t happen overnight, but technology will eventually reduce the amount of effort required to collect and maintain good data. We’ve come a long way from collecting server logs to the way we work tag managers today, and we are moving quickly in the direction of standardizing and automating data collection. When the problem of collecting data from websites and mobile applications is resolved, the demand for highly skilled implementation specialists will inevitably decline.  

So, those of us who have built our careers on this demand must ask ourselves: how will our priorities shift as the challenge of getting good data is resolved, and how do we prepare ourselves for this change? The answer is through data science.

How data science skills will make future analytics teams successful

Imagine a world where analysts have excellent data at their fingertips, taking data collection and data operationalization off their plates. and they still have the same responsibility: to provide leadership on interpreting data to advance the business.  What would their priorities be, and what skills would they need to do them well?

To answer that question, I look to the work being done today by people who focus exclusively on interpreting data, whom I am broadly defining here as data scientists.  The definition of data science is under debate, and not every aspect of data science applies to digital analytics, but I find the following from the Data Science Initiative at NYU to be helpful:

“One way to consider data science is as an evolutionary step in interdisciplinary fields like business analysis that incorporate computer science, modeling, statistics, analytics, and mathematics.” (http://datascience.nyu.edu/what-is-data-science/)

Concurrently, while we have been making progress to solve the problem of collecting good data, the historic barriers to entry into data science have been quickly dismantled by the increase in free training materials and the decrease in cost for cloud computing/storage.  If we consider that we will be spending less of our energy on data collection in the next 5 or 10 years, and think of data science as the next evolutionary step in digital analytics (which is becoming increasingly attainable), it is helpful to explore how computer science, modeling, statistics, and mathematics are already being applied to digital analytics problems today.  

Here are three examples that are occurring on the fringes of our industry today, but which I predict will be commonplace in the next 5-10 years:

  1. Anomaly detection: Alert the business to unusual activity.  This is being used today to monitor user behavior, as well as to identify potential errors within your tracking code.
  2. Calculating lift: Understand the return on investment we generate with the projects we tackle by measuring how a metric performs relative to its forecast.  This is being used today to determine if an advertising campaign was able to increase revenue above our forecast or to measure if a site redesign effectively improved user experience.
  3. User Clustering: Assign users into groups that share similar characteristics and/or behaviors.  This is being used today to detect the users who are about to make a purchase, or users who are interested in consuming certain types of content.

And these are only a few of the ways that data science capabilities are being applied to common digital analytics problems.

This Is How We Prepare

Today, our team at Search Discovery is making significant investments to expand our data science capabilities and solutions.  As part of this process, we’ve outlined three things that anyone working in our industry can do to start preparing for the future we’ve outlined above:

  1. Take advantage of free online training tools:  For SDI, this meant asking a few team members to identify the best resources for developing particular skills, and sharing that list with the rest of the team.
  2. Join the community: Pay attention to those members of the digital analytics community who are currently applying data science techniques in our field.  We recommend joining the Measure Slack community of analysts (join.measure.chat) and taking a look at dartistics.com.
  3. Experiment with ambitious projects: Seek out opportunities to apply what you’re learning.  This might mean tackling a business problem or simply competing in a Kaggle competition.

We have chosen a young, dynamic, and constantly evolving field to work in with digital analytics.  I hope you’re excited about that because it means that those of us who are flexible, paying attention, and eager to learn have the opportunity to drive all of us forward as a community.

If you’re interested in seeing the full deck from my presentation, or if you’d like to learn more about Search Discovery’s approach to digital analytics, contact me here.

Search the Blog

Scroll to Top