You don't have to look far to see rapidly growing or already mature "disruptors" in retail and manufacturing, but the long tail of market disruption goes way beyond that in today's age of consumer disloyalty.
For example, there are more 1,900 “invisible brands” competing globally in the carbonated soft drink category that account for 13% of total sales. What’s more, sales within the long tail of brands ranking outside of the top 200 in the U.S. are up 3% in dollars compared to one year ago — and growing at more than twice the rate of the 20 largest brands.
Lower barriers to entry have made it easier than ever for small brands to effectively compete with category mainstays. Individually, a single small brand won’t likely generate sales that affect the overall category. But they can make a great impact collectively. Often hiding in plain sight, small brands hold the potential to emerge as even bigger threats in the future as they tap into an unmet need, or ride the wave of a rising trend, to which leading brands struggle to respond.
But no matter their size, it’s crucial for companies to have visibility into which of the thousands of small brands in their competitive set are primed for growth.
But how do you strategize against this pivotal group, especially knowing that it comprises such a disparate array of brands? How do you define small brands consistently in all the markets and categories in which you operate? What factors make certain small brands successful despite their limited scale and resources?
Data Science at the Core To unlock the science behind successful small brands, the big players should leverage certain types of technologies, particularly machine learning and data science.
Machine learning is often mischaracterized, and it’s easy to understand why. In reality, it represents a continuum of solutions that work together and process complex data inputs. At the early end of this spectrum, you have longitudinal and regression analyses that establish criteria to monitor and predict potential threats in the future. More advanced machine learning often centers on neural networks and has the potential to determine key drivers responsible for the success of small brands by category and classify them based on threat level. As machine learning technology advances, new ways to business value will assuredly emerge.
When it comes to understanding consumers, multivariate clustering involves grouping them based on behaviors, needs and other factors. Machine learning can help rectify the inherent impurity issues of big data (whether census retail data or crowdsourcing). It’s like a 1,000-piece puzzle where you only have 750 pieces: You get a general picture but still need to find those missing pieces. Robust data science initiatives can help a big brand fill in the gaps.
Also, whether via regression or neural networks, brands need to optimize their marketing investments using a mathematical, data science-driven approach. This should inherently provide an advantage for larger brands over smaller ones that have fewer resources at their disposal — like intellectual capital or a larger pool of trained data scientists and experienced coders.
The message to big brands is this: Leverage predictive modeling to seamlessly, and scientifically, analyze all products, geographies and channels; early trend identification and response not only protects against the long tail, but builds competitive advantage and, ultimately, growth.
Emulating the Disruptors with Personalization To stay successful, large brands should marry their robust architecture and resources with the best practices of the disruptors themselves. Gone are the days when they can take for granted that consumers will keep coming back. Large brands must be more personal. Technology has opened up avenues of interaction between consumers and brands, particularly via e-commerce and social media.
Consumer see themselves as individuals, and tech platforms enable them to make that a reality by letting them curate their lives. Brands that don’t become part of that lifestyle risk being easily substituted by something that provides a better fit.
Personal engagement is essential, but also must be conducted at scale. Machine learning and data science can help facilitate this level of engagement. A legacy brand can use data science to conduct micro-targeting that gets it closer to personalized experiences.
Large brands must now think about how to disrupt themselves and better engage with consumers rather than executing their own brand strategy. Data science enables large brands to get ahead of the curve, remain relevant and identify and work together with the right combination of small brands.
With up to half of organizations expecting to lack the AI and data literacy skills needed to achieve business value in 2020, there has never been a better time to put science at the forefront of maintaining market share. Companies that exploit this gap now will gain the insights needed to compete more effectively and prevent smaller brands from encroaching on their turf.
About the Author Richard Cook is senior vice president of data science for Global Connect at Nielsen. In his tenure of more than 25 years at the company, he has served in various roles across marketing, product, business process improvement, and now data science.