We’ve all heard of the amazing ROI of big data analytics – where a few hundred thousand invested in analytics returns many millions in new revenue or renewals. That is a powerful intoxicant: If I can just repeatedly do that one, perfect thing that will really drive our business, I’d dominate our market and be a hero. Problem is, identifying that one, perfect thing is very hard. Analytics can help, and it’s more accessible today than ever.
It’s what TopRight CEO Dave Sutton calls getting to an “atomic level of marketing and business practice.” The key is to separate signal from the noise, and think about the problem from a different angle. Data patterns and a good model will help you think differently – letting the model do the work of identifying what is important, especially outside of the “sacred cows” and legacy dictates that often hold marketers back from innovative approaches.
Can business units or mid–sized firms also take advantage of this value? We think it worth exploring. In a recent conversation with Gary Samoluk, a partner with data analytics firm CxO Analytics, we mapped out a high level view of how these kinds of engagements work well.
- Consider your data sets. Using internal data in performing analytics is useful from a tactical point of view in understanding daily performance and short term trending. But it does not give insight into understanding the dynamic forces that affects sales numbers, such as customer and market behavior shifts. Incorporating external data provides greater insight into understanding the ‘whys’ of what is affecting what the internal data shows. “Just as it is important to have good data that you can rely on, it is equally important to have the expertise to ask the right questions of the data for analytical modeling,” Gary said.
- Ask hard questions. The point of data analytics – with big or small data sets – is to generate actionable insights. One caution we’ve seen: There are many questions that marketers deal with that have nothing to do with the business, but are more about how things get done at a particular company. Focus on the critical business questions that will really engage customers and help you make strategic bets for investments. These may include:
- What dynamic forces are affecting my customer and how effectively am I changing to meet these changes?
- What market forces and challenges are changing the direction of the market?
- Are their new market opportunities developing that I can take advantage of and become the industry leader?
- Would this product be interesting to our current customers? What must be true for customers to feel pain?
- Are there market segments resistant to our offerings? Are there competitive strongholds with vulnerabilities?
- Which of my promotions are working to drive real sales?
- Where is the customer headed? What compromises are they making? What needs are unmet?
- What is the product life expectancy of customer segments? (These kinds of questions are particularly important in healthcare marketing – where certain behaviors can have dramatic impact on patient health and the treatment costs.)
- Who are our most valuable customers, and over time? What outside factors impact customer loyalty and retention?
- What are the characteristics of our best prospects?
- Which marketing messages and campaigns are contributing, and when do they contribute during the lifecycle?
- Design the model. Look at market data to identify competitors and current solutions to address these challenges as a starting point to design the model to work within your unique industry segment and market sector. “It will help you better understand your product’s strengths and weaknesses based on analytics to improve product planning, market positioning and competitive positioning,” Gary said. It will also provide an outline of the marketplace and help you refine the questions to your specific business, based on your product and brand advantages, and vulnerabilities. “The model will primarily build from your own data – collected from various sources,” he said. It’s just as important to ignore some data as to include others. Most marketers do not have a data problem. They have a data prioritization problem.Consider third party data to bridge any gaps or validate any assumptions. In your quest to perfect your model, don’t discount third party data which will augment your own data and validate your market assumptions.
- Run the model and continually iterate with new data to ensure your insights do not become stale over time. This is how the model helps you maintain a grasp on those dynamic forces that can distract customers and, perhaps worse, catch marketers unawares and unprepared.