Why Data-Dependence is Crippling Marketers

I’ve always been a data oriented marketer. This is what everyone wants now, but there can be major pitfalls that come with this mentality.

At the end of 2017, Forrester released a report saying that more companies will replace CMOs with Chief Growth Officers.

In Forrester’s report they say:

Great CMOs blend tech savviness, an insights-driven mindset, and deep emotional empathy with the customer to drive top-line and innovative growth.

The interesting thing here is fundamentally the CGO role is just what we want the CMO to ideally be.

At the end of the day, every company wants a CMO who is tech savy, data driven, and keeps the customer and her experience first.

As an industry we’re changing the title to make it clear the job of the CMO has changed.

Marketings role has changed.

Marketers focusing more on data is a good thing, but we have to be careful to not depend on it too much. For all it’s positives, data can also have negative effects for marketers.

Data gives you a false sense of security

The problem isn’t the focus on data.

The issue is the trust of data.

“In God we trust; all others bring data.”
― W. Edwards Deming

Every modern marketer has seen this quote. You’ve probably even said it to someone in a serious tone. (You can admit it…)

The truth is there are many times your data doesn’t give you the full picture and marketers who are entirely focused here will continue to make missteps.

Let’s talk about some of the issues.

Past results vs. future results

Data is 100% focused on the past.

There are many people smarter than me working on machine learning to increase the effectiveness of predictive and prescriptive data, but we’re not there yet.

Data is always looking at the past.

As marketers we use statistical significance as a way to mitigate this, but in many ways it’s inefficient.

Statistical significance tells you how likely you are to see the same result from a specific test, but that’s at a micro level. You have to look at the macro level as well.

Every major tech companies of this generation focused on this type of testing and it’s delivered bags of money onto their doorsteps, but we are starting to see these decisions blow back.

Facebook’s growth team has run millions of tests which have paved the way for Facebook to get to over 2B monthly active users. That’s the story everyone wants to hear. Running a lot of small tests will lead to huge results.

On the micro side that is true. The issue is the blow-back Facebook is seeing today. Facebook focused on driving the most usage and did a great job of that, but they never stopped to think about what this would mean in the long term.

Facebook created tests to get as many people using their product as they could and they were highly successful there. Where they missed the mark is they created an algorithm that focuses more on getting people to come back — instead of making sure it’s a site that makes people feel more connected to people they care about. But hey, they logged on, so they got that going for them.

The point is, running tests around growth is great, but make sure those results are aligned with your companies brand, values, and story. This is how you get long term results. Facebook is seeing this now and that’s why they are making the changes to the way the newsfeed works.

Facebook will rebound, but every not every company can pivot the way Facebook can.

It doesn’t take a strong leader to roll out a winning test. It takes a strong leader to chose to not roll a winner if they believe the long term results will be impacted by this kind of practice.

Data as a paralyser

We can see why moving too fast and relying and data can work against you, but the same can be said of the opposite.

Since data is always focused on the past (did I mention that yet?), waiting for data to come in can be worse than moving too fast. This is especially true in subscription and SAAS.

If you work in any business with recurring payments, you know that you can’t look at front end metrics exclusively. We want to know what creates a good member who will stick, not just a new member.

Unfortunately this data can take months.

Waiting months to take an action, shouldn’t be an option, unless that decision can make or break your company.

This is where you need leading indicators and a combination of gut.

While we want a member who will stick, waiting to prove that is generally not a great idea. Look at the leading indicators and decide if you think it will help or not. A good filter for this is trying to understand why something caused an increase. In my experience, if you can explain it, you’ll have a good idea if it’s a good idea or not in the long term.

From here you can roll it out or not, but you should note to check the long term results in a few months.

You will not be right every time, but you will be right more than you’re wrong. If you’re wrong, fixing your mistake is less of a risk than constantly waiting to be proven right before action.

How do you know if you have enough data to make a decision?

Once you have 26.5 pounds of data, then you have enough.

Wait, that’s not right.

Unfortunately this is the real challenge and there is no black & white answer. What I can say is, most people use too much or too little, thus the need for a Goldilocks rule for marketing data.

Every company needs to do continue using data and running tests and that’s why you’ll continue seeing more growth teams and CGO positions popping up. But make sure you are also gut checking your data.

We are at a point in time where these is almost infinite data. This makes it easy to “let the data make the decisions.” A lot of companies will do this and over-course correct due to the availability of this data.

The next group of companies that win will not be the companies that can grab the most data and test as quickly as possible. Google and Facebook will continue to dominate here, but they’ll continue to have brands that contain equity but little meaning.

The next cohort of winners will be the brands and storytellers.

Branding will be instrumental and you need to use data and testing to support your brand, but you want to be data-driven, not data-dependent.

Don’t forget to use your gut.

P.S. If you have any feedback, let me know on Twitter or LinkedIn or leave a comment below.

This story is published in The Startup, Medium’s largest entrepreneurship publication followed by 297,332+ people.

Subscribe to receive our top stories here.