With all the technology needed to bring the promise of the Naytra paradigm to life, there’s ultimately one factor that will shape whether it lives or dies. And that is the human consumer in front of it.

As I talked about earlier, the consumer needs to train her own model through implicit actions and explicit feedback. This is so that their Naytra can constantly learn, grow and get better over time. If this feedback loop doesn’t happen, a consumer’s Naytra will be ineffective and lose the trust of the consumer as well as that of brands who interact with it.

The value Naytra brings will be a function of the strength of the interaction between the consumer and her Naytra, through it’s user interface. The Naytra user interface has to engineered be easy, enjoyable and most importantly habit forming. It has to be made to stick! And machine learning can help with that.

My research took me to the works of Nir Eyal, the author of ‘Hooked: How to build habit-forming products’ and Dr. BJ Fogg who created the Fogg behavior model. They have developed frameworks to design products that drive human behavior. I’m proposing the application of machine learning models to the Hook Framework. These models will propel the UI of Naytra interface, and potentially make the Naytra consumer interface a habit-forming and indispensable part of a consumer’s life.

The Hook Framework framework is a 4-phase cycle that starts with cuing the consumer to take action in implicit anticipation of a variable reward. A variable reward could be:

• A social reward fueled by connectedness with others
• A reward of the hunt, which is the search for material resources
• A reward of the self, which is the intrinsic reward of mastery, competence and completion

The cycle ends with increasing the likelihood of the consumer returning by improving the value Naytra provides the more it’s used. It enables the accrual of stored value in the form of better relevance for the consumer the next time around.

By applying learning models to each of these phases, the UI is personalized to maximize the engagement from the user and deliver value to the consumer in a manner that makes it habit forming for the user. The feature vectors used in these models are a combination of the content and context and activities of the user within this interface. These learning models could be based on content filtering, collaborative filtering and hybrid filtering techniques that I talked about earlier.

Conclusion

We saw how the Naytra UI could be designed to become habit forming through applying machine learning to each step of the Hook framework. Design appraches along these lines need to be explored in order to successfully make Naytra a part of the consumer’s life.