When I first started working in marketing, we talk much about the role of the CMO in analytics because there wasn’t much to say. We produced commercials, created products and put them in pretty packages, sent them through various channels, then hoped customers bought. Sometimes, we’d do some market research to understand consumer sentiment surrounding our brand, but marketers weren’t very savvy about numbers.
Well, all that’s changed.
Perspectives on the role of a CMO in analytics
Here’s a perspective on the CMO in analytics from Maggie Chan Jones at SAP:
The marketing analytics now is really about getting from data to insights, and then from insights to action. The trends that I see are, for one, an increased level of accountability, for marketing to demonstrate ROI to the business owners or business leaders. [source]
So, let’s look at how today’s CMO creates value by:
- asking the right questions
- interpreting data
- taking actions that create value.
In a recent article on Forbes, Piyanka Jain argues that insights among a CMO in analytics is pretty simple, that they don’t need big data, predictive analytics, and other fancy tools to effectively return value for the organizations. In fact, Jain, author of a bestselling Amazon book, argues that 80% of decisions rely on simple techniques and only 20% on more advanced techniques.
I call “Bullshit”.
Simple technique may work for the role of a CMO in analytics 20 years ago, but they certainly don’t work now. Correlation isn’t causation, trends are simply linear extensions, and estimates are just that. CMOs today are better than that–they have better tools, more data, and need real analytics (predictive, segmentation, time series, etc) to support decisions that increase value for firms.
Instead, the types of decisions Jain presents as involving simple analytics are the purview of managers and low-level marketers, not the CMO.
Even knowing what questions to ask means the CMO is an expert in advanced analytics. See why.
Asking the right questions
Let’s start with the basic questions underpinning marketing regardless of industry:
These are mainly segmentation analysis–on of the advanced techniques. Simple stats aren’t going to help you here.
Who is currently buying your products? And, here we need to know them in depth, not just age, gender, socioeconomic background. We need to understand their lifestyles, values, attitudes, motivations, and decision-making processes. We also need to know who influences their decisions.
Some of this we can get through our internal data, such as Google Analytics, sales receipts, etc. But, much of this we need deeper sources, such as social media to help us understand our customers. These rich data sources provide insights never available before to marketers.
Next, consider folks who look like customers, but aren’t buying your product. Why not? How do you make inroads to reach these non-customers.
Finally, who SHOULD buy from you. Why aren’t they?
Potentially, some of our customers cost more money than they’re worth — which uses the concept of CLV (customer lifetime value) — a modeling technique — to determine customer segments and strategies for each segment. This might mean getting rid of customers or charging them higher fees to compensate for their negative value.
Next, evaluate markets that you’re missing. Maybe your product would also do well with a different gender, country, region, etc.
- What should you sell?
- Do your current product match what customers need?
- Will they match customer needs in the future?
- What are unmet needs among your target market? Can you successfully meet some of them?
Again, you might use some simple forecasting models, but those are simply linear extensions. To really create value, the role of a CMO in analytics involves using AI (artificial intelligence) and NLP (natural language processing) to characterize qualitative data. For instance, a colleague at Microsoft uncovered some major failures in a recently launched product by evaluation comments on Facebook and Twitter. Before the product was widely distributed, Microsoft was working to fix the problem in next generation products.
Also, consider what price to charge, which has a dramatic impact on sales. Offering products at a lower price isn’t always the answer, especially for products where price is a surrogate for quality or when customers can’t objectively evaluate the product prior to purchase.
A number of pricing models exist (these are predictive, usually econometric models) or CMO’s can use existing data to develop predictive models that make pricing easy.
- Should you distribute online only (pure play), some combination of online and offline, or purely brick and mortar retailing?
- Should you own the stores, like Apple and Nike, or should you distribute through partners or both?
- Which specific retail outlets?
These are questions CMO’s face and need to reevaluate constantly. And, not only with respect to where to distribute physical product, but messaging about that product.
The role of CMO in analytics is complex with respect to these decisions. Again, modeling helps optimize value by building predictive models of response under each scenario.
- When to advertise?
- When to launch?
- When to discount?
- When to delete a product from your product line?
Again, all of these decisions rely on predictive analytics.
Why questions go well beyond quantitative analysis and this is where the role of the CMO in analytics is critical. He/she must ask a series of what if questions to determine the rationale behind changes or changes that might impact value.
For instance, a spike in visits or sales begs the why question. Why today? What’s different? If we can figure out what caused the spike or decline, we can work to build that into our marketing efforts.
The CMO has to be part detective, snooping through the internet and social networks looking for mentions related to their brand, their industry, or consumer markets. Collecting such data should be routine, even if the CMO isn’t using the data consistently.
For instance, one of my clients experienced a spike in website visits. Why?
I started by searching Google Analytics to see where the increased traffic came from by comparing traffic on the busy day with traffic from the same day the week before to see what channel represents the increase in traffic.
In this case, it was Twitter.
Next, I looked for mentions of the brand on Twitter to see why more traffic was coming to the website. Was is a particular message? A RT?
In this case, the spike resulted from an influencer Tweeting a link to a piece of content.
Again, you’re looking at a combination of Google Analytics and NLP to help ferret out the Whys.
Need marketing help?
We welcome the opportunity to show you how we can make your marketing SIZZLE with our data-driven, results-oriented marketing strategies. Sign up for our FREE newsletter, get the 1st chapter of our book on digital marketing analytics – FREE, or contact us for more information on hiring us.
Hausman and Associates, the publisher of MKT Maven, is a full service marketing agency operating at the intersection of marketing and digital media. Check out our full range of services.