A recent article got our data team’s wheels turning. In particular, it dealt with the idea that CMOs are frustrated with analytics. With all the growth, all the budget increasingly being put towards data and analytics resources, CMOs are experiencing frustration because, despite all the additional data, it’s unusable when it comes to making big, company-changing decisions.
Consider this hypothetical example: You’ve just come in as a new CMO of a company and you want to make some changes. You want to use data to not only inform your decision, but also support it. Are data analytics supposed to make decisions for you as the CMO?
Our answer? A resounding no. Data analytics inform decisions, they bring confidence to the decision-making process to some degree, but human input and perspective is still a critical component.
Sometimes you can make decisions solely based on data, when the path forward is clear and straightforward. But in most cases, these are smaller, tactical decisions that deal with a narrowly defined scenario like year-over-year sales patterns in specific markets.
The big, important, complex questions companies face are never binary equations that have clearly defined variables. That’s why the insights that analytics generate can sometimes seem disconnected from what a CMO might expect.
The risk of overpromising and overdelivering
In this situation, the CMO might want to use data to inform and understand such a large decision, but it’s unlikely that McDonald’s has an existing data model that would give clear-cut direction to the decision. Because it’s a long-term objective that pays off positively or negatively over a period of time, there are all kinds of variables, some impossible to define, related to costs and whether or not the McDonald’s customer base would even buy in to the change.
So, what to do?
An analytics specialist might feel some pressure to find answers to these tough questions for the CMO. If you’re part of an analytics company, analytics are your product and you’re contracted to complete work for other businesses and brands. It’s also a different kind of pressure than when you’re inside the organization, depending on where you are in that organizational tree.
In some cases, the data analytics role is located within the finance group, and they work directly with marketing. This sometimes creates a situation where the data analyst is asked to prove something doesn’t work in order to ask for more money to do something else.
If you’re on the other side of an analytics or marketing agency, the pressure is flipped. It becomes a question of proving that marketing is, in fact, doing what it’s supposed to. Prove that something is happening as a result of marketing dollars spent. And if you can’t, the marketing dollars may go away. Both of these scenarios can lead to overpromising results and reporting, which can lead to a negative reputation for analytics over time. Analytics don’t always provide the answers, and perhaps shouldn’t be positioned as being able to deliver them.
Run marathons, not sprints
The perception that analytics provides answers can easily permeate through an organization, especially when an analytics specialist isn’t in the room when the big decision is made. It’s the analyst who churned through the data who can help explain the insight, making it crystal clear.
The CMO, unfortunately, usually has one of the shortest tenures in the C-suite. If you come into the organization as a CMO, you have to deliver, and sometimes the cards are stacked against you. You can only do so much on a product or business. Then, if there are problems inside the business, no amount of marketing is ever going to fix it. It’s in many ways like a sales position – if the CMO comes in and doesn’t deliver, bad things are going to happen. CMOs are the lucky recipients of all this pressure to deliver on the marketing plan, make changes to messaging and tactics and then produce measurable data-based ROI on the back-end, all within one to two years.
So, it’s a convergence of forces. And, within that convergence, the CMO needs to be empowered to perform more as a marathon runner, and less a sprinter. Doing so has the advantage of relieving any unrealistic expectations or associated pressure from your analytics group to create new models or data sets out of previously nonexistent metrics.
And this is the key to the frustration. A CMO’s pressure, the pressure on this person to build results quickly, does not line up with a good analytics strategy, which is often designed to influence gigantic business-altering decisions effectively. You can achieve short-term stuff and increase confidence a little bit, but to really drive the level of confidence most people want and are sold on when they purchase these analytics services, you need time, sometimes a lot of time.
Frustration from CMOs and analytics teams
For example, from a media planning perspective, the current trend is to move money out of TV and into digital. But, for a company that has executed in TV for decades, that’s a tall order. It’s not something you can just switch in a year or even a couple of years. To make a decision like that—if someone is really on board and wants to build confidence in data to do it—they need to take a multiyear approach to building the confidence. Typically, what that means is testing at the outset, and growing test sizes and the impact of them over time. And – as is the case with most marketing cycles and business cycles, you can’t just do it all in one year.
Instead, it may take three or four years to go from, say, 80% TV and 20% digital to maybe 40% TV and 60% digital. But now you’re past the average tenure of a CMO, and people are wondering where the results are. So you get a new CMO, and the process starts all over again.
From an analytics perspective, it can be just as frustrating. An inexperienced analytics specialist is going to tell you they need time to make a decision. And that’s an issue with overpromising. A less experienced analytics company or specialist may try to force the decision, result or insight to be ready to report as quickly as possible, because they know what people want.
In essence the frustration stems from several things. First, the expectation that complex decisions are, or can be, binary and can be answered rather than well-informed by analytics. This can create expectation from the CMO, which pressures analytics teams to over promise and over deliver, to blur the line between assumptions and facts. Second, and more importantly, it’s essential for CMOs to remember to adopt the mindset of a marathon runner – not a sprinter. It takes time to gather the pertinent information and insight to inform decisions. Predictive analytics can provide the confidence to present ideas, make changes and produce results, but it’s a marathon, not a sprint.