The rise of consumer-centric, personalized wellness solutions
Technology is advancing at an accelerated rate, allowing for consumers to have access to resources that before could only be found in the clinic or at a research center. Now, many companies are appearing that offer personalized solutions to improve the consumer´s health and wellbeing, taking into account their microbiome, genomics, lifestyle habits, dietary preferences, and medical conditions, among other factors. In this sense, the majority of these companies offer a personalized assessment and coaching service, although there is a rising increase in combining these services with product recommendations.
The dietary ingredient industry has always been very product-focused; however, there is an increased demand to shift towards a more consumer-centric approach. In the case of Monteloeder, we have developed a clinically-validated weight management botanical ingredient, which we have combined with an app specifically developed to manage the consumer´s experience while taking the ingredient. The app is linked to the product, and all the messaging and expected outcomes are based on the product´s effects that have been detected in the clinical trials of the ingredient. The end result is a more complete, digitalized health solution, combining behavior change techniques to promote healthy lifestyles along with product consumption management. We offer this solution to our clients as a way to provide a more personalized, up-close approach to meet their consumers’ needs. This concept has been defined as “Digital Nutraceutical”, and Monteloeder has been identified as the first in the industry to offer this solution to the industry.
Digital technology in the dietary supplement space
One of the advantages of working in the personalized nutrition space is that they are not one-shot deals; consumers are aware that improving or maintaining health and wellness is a constant, life-spanning ordeal. And their needs change throughout life, as a consequence of either normal aging process or due to health conditions that can appear without notice.
These insights make it logical to provide consumers with an ongoing service to accompany the personalized nutrition solution. This concept is called “Servitization;” which translates to offering clients the capabilities to provide the services and solutions that accompany their products. For example, the app developed by Monteloeder to go along with the weight management ingredient provides a way to manage consumer adherence, product expectations, and manage their lifestyle habits to ensure that they reach their health goal with the product. The app can be seen as a facilitator; it helps the user to adopt healthy habits while also improving their adherence to the product consumption in order to maximize the health benefits, as well as facilitate the purchasing power of the consumers by allowing online purchasing.
But the app is much more than just a habit-changing, product consumption-enhancing solution. It is also a two-way communication platform; the company provides the necessary information and tools to improve the consumer´s wellbeing, while in return the consumers “talk” with the company through their data collected by the app.
This data collected by the consumers can then be analyzed and interpreted, allowing companies to better understand their consumers. Due to the massive amount of data that is collected, traditional methods of data assessment are insufficient. Instead, we have been working on applying machine learning techniques to assess the data for several purposes; from consumer profile assessment and behavior for marketing strategy optimization, to identifying new consumer needs. In the case of the latter, it is possible to detect if an already available product can be recommended to the consumer, and therefore direct the consumer's purchasing power towards the company's portfolio of products. Or it can identify that the unmet need cannot be covered by the currently available products. This provides the Research and Development Departments new insights as to where they can lead their new developments; ones that are targeted for specific consumers, which may have a higher rate of success in the market.
Future trends that will influence the industry and new product developments
As consumer data acquisition increases, so does its complexity in interpreting the data and then making decisions based on the data. As a consequence, machine learning is proving to be an increasingly valuable asset. For example, it can be used to analyze data collected from the clinical trials to identify correlations between health parameters, as well as how and to what degree they affect the main health indicators. Also, it can be used to understand the level of impact the ingredient has on the health indicators.
Besides understanding how the ingredient affects the consumer's health, there are other applications that machine learning techniques allow to develop. For example, it can be used to design algorithms that can be implemented into digital solutions. These algorithms can be for classification, that is, allows to identify if the consumer is taking the product (and therefore manage consumer compliance and adherence), or regression, which allows to predict the consumer´s health indicator outcomes. In the case of the latter, being able to predict the outcomes of taking a product for a specific consumer can allow to design more specific interventions, taking in consideration not only the product intake but also the consumer's lifestyle.
The next challenge resides on taking this predictive model, created using data from the results observed during product development, and bringing it to the real world. Clinical trials are a controlled environment, and the “consumers” are specifically selected based on a set number of inclusion/exclusion criteria. Therefore, the algorithm generated will have a high rate of success in predicting behaviors in individuals that are similar to those used in the clinical trials. However, taking this algorithm to real end-consumers in an uncontrolled environment can be quite challenging. In this case, it is very possible to make errors in its interpretation. In order to circumvent this problem, there are methods to take the predictive model based on the clinical trial data, called the baseline model, and retrain it using the individual´s collected data, and therefore give rise to a specific, personalized predictive model. In this manner, we go from one baseline predictive model to as many as the number of consumers that are using the combined health solution. This is the real challenge that is currently at hand, and where truly personalized, n=1 solutions will begin to appear.
Jonathan Jones, PhD
