December 6, 2017 // By Russ Miller
Now that over 10 years have passed since the first iPhone was introduced--and more than nine since the first Android smartphone (remember the HTC Dream/T-Mobile G1?)--most of us have created multiple mobile apps for our customers, employees and partners.
And over time we've learned and gotten much smarter about how to approach mobile initiatives, increase our odds of success and mature mobile app portfolio management.
Today, it feels like mobile app development is a known process, we have our arms around it, and we know the recipe for success. In many ways this is true; for example, the gap between iOS and Android has never been smaller. Almost all of the mobile market is owned by two mature operating systems with little chance any viable third competitor will emerge on the near horizon. Mobile is quickly becoming the dominant channel for eCommerce. There are many effective cross-platform approaches to mobile development that save time, money and deliver engaging experiences.
However, the mobile story from the beginning has been rapidly evolving technology, innovation and disruption, and continuous transformation of customer experiences and expectations. This remains the case today!
As you think about how your mobile apps need to evolve in 2018 there are three themes you should focus on:
- Artificial Intelligence (AI)
- New Experiences
- Data-Driven Interactions; e.g., through Machine Learning (ML)
Data-Driven Interactions through Machine Learning
- Netflix uses linear regression, logistic regression, and other machine learning algorithms to perfect its personalized recommendations by means of ML
- Tinder's "Smart Photos" feature shows a random order of your profile photos to people and analyzes how often they’re swiped right or left. This knowledge allows Tinder to reorder your photos by putting most popular ones first. This system is honing itself constantly
- Google Maps' "Find Parking" feature uses anonymous aggregated information from users who decided to share their location data as input to a standard logistic regression model. Then the app--based on the dispersion of parking locations--predicts when, where, and how difficult finding an empty spot will be