The story began when I went to the clothing store to buy a pair of pants. After trying on a couple of models, I did not like anything specific and I was slowly thinking of leaving the store. In the meanwhile, one of the experienced salesmen came to me. He guided me through the racks and showed me what he thought I would like. According to my reactions to each model, he adjusted his next recommendation. So instead of quickly leaving the store, I bought two pairs of pants and a T-shirt too. Now I’m sure if I go to this store to buy bags or other things, the seller will guide me right to my favorite model fast.
This experience affected how I see the journey of mobile application users. More often than not I just can’t find anything valuable in an application for me to continue using it, so I uninstall it. Yet, there were also times that I’ve been excited by the personalized offers and promotions the app sent me just in the time I wanted them. Sadly the second experience is a fancy ability, found only in the apps made by tech giants like Google, Instagram or Pinterest. The first story is usually the case with smaller apps.
If the app owners want to create an experience similar to what happened to me in the store for their users, they should:
- Understand the users’ behavior, inside and outside the app, uniformly
- Know the users’ persona, taste and…
- Design valuable offers for each user
- Find the right time to deliver the offers
- Talk to the point and in a delightful way
- Use the lessons learned to deliver tangible improvements in the next offerings
Personalization is the key
It is a cumbersome process to monitor user behavior in the application and adjust the offers or contents according to each user. There are many users each with different tastes, behaviors, interests, and habits. Even just monitoring their behavior is not so easy.
The solution for this usually involves coding in the application and sending important events (like register, purchase, …) analytics services. Then using those tools they find out the actions each user has done in his/her lifetime and design personal offers for them.
But still, the issue of identifying user persona and characteristics and sending personalized messages remains. How old is the targeted user? How rich is s/he? Does s/he keep lots of apps or have a few ones? Does s/he look for highly-rated apps or is less sensitive to quality? Which app features are most likely to be of interest to and used by them, or maybe they will like in the future? And there are many other features that these commonly used analytics and mobile marketing tools do not cover.
So, we wish to have a smart assistant instead of sophisticated tools that need a manual design of rules by experts to work, someone like that genius seller I mentioned at the beginning of the article. But unlike the seller, this assistant should be able to handle any number of users simultaneously. It should be knowledgeable about different businesses and provide customers with the appropriate experience of communication, whatever business size and scale.
AI can be a valuable assistant
Artificial intelligence has impacted many businesses including marketing. It can act like that experienced salesman to some extent and can support a virtually unlimited number of users and requests. Some companies have exploited this tool to improve their user engagement and create personalized offers and content for their users.
According to an article on Forbes, 53 percent of marketers intend to adopt artificial intelligence over the next two years. Research from Demandbaseshows that 80 percent of all marketing executives surveyed see marketing AI as the next big thing, revolutionizing marketing by 2020.
If you’re from a big company with like $50M of annual revenue, you can stop reading and go hire a team of data scientists and engineers to build anAI-powered marketing system and pipeline for you. Hiring dedicated data scientists and machine learning experts is one of the most costly hires and probably is not an option a small business can afford.
Are you still with me? So it’s good to know that many SaaS businesses are working on the user engagement industry. A number of them have been attempting to add predictive models of user behavior using artificial intelligence and can estimate the possibility of churning application users. Using these services, you get more insight into how your users are leaving, and you can turn needed insights into actions. Here are some examples of services you can use for this purpose:
Localytics is a mobile marketing platform that provides solutions for monitoring acquisition channels and campaigns to gain valuable users, engaging current users with personalized messages and retaining users that are in a higher risk of churn. Localytics concentrates on delivering meaningful and personalized messages across email, push notification, in-app messages. It helps marketers to use deeper insights to create powerful communication with their users through intelligent solutions.
Airship (formerly Urban Airship) enables enterprise brands to use user-level data to deliver push notifications, emails, SMS, in-app messages, mobile wallet passes and more to engage the right people at the right time.it uses AI to predict churn of users and find an optimal time that customers are most likely to engage with the message.
Hengam is acting as an intelligent assistant when it comes to users’ churn and engagement. It gathers all events inside and outside the app (to the extent of the data user agreed to share) and use them in its models to analyze and predict user behavior more accurately. It automatically identifies anomalies that occur in the users’ behavior and informs you to take the appropriate reaction through the notification, email or SMS, depending on their status and persona. In your messages, you can send specific offers to users and remind them of your values, fully personalized. Hengam also offers decent prices for small and medium-sized businesses.