We are excited to announce a whole new way for marketers to communicate in real-time with their customers. Real-time predictive segmentation changes the way you interact with your website customers. With predictions being made in milliseconds, our engine will know what your customers are going to do, even before they think about it. Magic.
Arresting churn by presenting discounts even before visitors exit your website is one of many things real-time predictive segmentation does. A leap over the existing standard, real-time predictive segments know what your customer will do next before anyone else.
How does predictive segmentation work?
Before we dive into onsite or real-time predictive segmentation, let’s understand how most predictive segments have been working in the industry till today.
When millions of website visitors take different kinds of actions, it’s impractical to create segments based on geological, demographic, and behavioral factors. Especially behavioral factors. The unique set of actions one user takes varies widely depending on their choices and what they saw before.
Identifying sets of users who are likely to perform a specific action and setting up journeys for them is so tedious that it can be called impossible. Well, manually. This is where Netcore Cloud’s AI engine Raman and the predictive segments come in.
What if triggers could automatically execute messages based on historical data and a step further, with real-time data?
That is the essence of predictive segmentation.
What is predictive segmentation?
Let’s understand this better with an example.
Jane is a marketing manager at an ecommerce company and has a monthly conversion rate target of 10% for the appliances category.
Other predictive segments
If she were using offsite predictive segments, when users visit the ecommerce store and perform actions like add to cart, checks would be done in an hour or day, and communications would be triggered during this time. This is a reactive way of interacting with customers and brings a large gap. If it’s the 20th of the month and the current conversion rate is at 3%, but she has already exhausted her Ad spend budget. She’s in trouble. The only way to reach her goal is to retarget visitors with organic emails and SMS. Naturally, she’d send out discounts on these channels. But there’s a problem. People have already left the website, and getting them to open a message and complete the checkout process is not easy. An average case scenario would be:
- 1M views → 100,000 clicks → 10,000 add to carts → 300 purchases
- Even if emails and messages are sent to 9,700 people with products in their cart, average cases suggest a 5% conversion rate from abandoned cart emails = 485 purchases bringing the total to 785, a 7.85% conversion rate.
Raman’s new predictive segments
The new generation of predictive segments is more preventive than reactive, it runs continuously on every page. By doing this, it tries to predict the next steps a user is likely to take and if the patterns match, web messages or communications can be set to trigger to arrest churn by showing a discount on the spot. Considering a very modest scenario for real-time predictive segments, the conversion rates for real-time push notifications can easily be 10%. Now, the purchases become 970 + 300 = 1,270, a 12.7% conversion rate, and hundreds of thousands in increased revenue.
This was a modest case. With real time predictive segments, we have seen up to 450% increase in CTRs.
Is real-time predictive segmentation better?
Both have different applications. While most of the solutions in the market use offsite segmentation and use historical data, we’ve added the ability to make predictions in real-time even for anonymous visitors. This gives you predictions that evolve with each user interaction and changing behavior.
Let’s draw a clear distinction between the two:
|Offsite segmentation||Real-time predictive segmentation|
|Prediction is made and communication is sent to users before or after their interaction with website||Prediction is made in real time and communication is sent in real time when users are on the website|
|Prediction is at a group level||Prediction is at a 1:1 level|
|Trigger executes messages after visitor takes action||Trigger executes messages even before user takes action|
|Has a fixed set of outcomes to trigger predictions||Predictions can be set to trigger communications for custom defined outcomes|
Here’s a diagram showing further distinction between the two:
Without a doubt, communicating proactively with unique messages brings distinct advantages that lead to better conversions.
A closer look at Raman’s predictive segments
In our latest offering, you get the following options:
- Likely to Purchase
- Likely to Abandon cart
- Likely to Purchase with Coupon
- Not Likely to Purchase Product
- Not Likely to Purchase Category
- Custom segments
Along with the Custom option, you can account for almost any use case the visitor will likely perform on your website and set up appropriate triggers for it.
Our intelligent AI engine, Raman, accounts for various factors like product click, add to cart, checkout, and more to make calculated predictions that can trigger messages set by marketers.
How does Raman and predictive segmentation benefit you?
Let’s look at some benefits of predictive segmentation and the AI engine Raman.
Provides direction to your marketing strategy
With Raman’s aid, you can easily identify the customers they should be targeting for each marketing campaign or customer journey. You can now easily craft a high-conversion marketing strategy that wins with these insights.
Helps you understand your customers better
As a result of Raman’s Predictive Segments, you can now better understand the dynamics of the three micro-segments and the larger consumer market. Targeted and personalized marketing campaigns avoid unwanted communications while achieving maximum ROI.
Develop a long-term competitive edge
Raman helps you anticipate what your customers need. This will help you plan your product or service roadmap effectively as well. You can then reach your customers before any other brand does, fulfill their evolving needs, and attain a competitive advantage over others.
Know your customers even better
AI predicting what customers will do based on historical data was one thing. It’s a whole new paradigm where AI can predict what your customers are likely to do, even before they think of it. To say it’s like reading minds would underplay its potential.
If you want to know how you can optimize conversions and get the most ROI out of your visitors with predictive segments, reach out to your CSM or contact us.