Customer AI overview
Customer AI, as part of Intelligent Services, provides marketers with the power to generate customer predictions at the individual level with explanations.
With the help of influential factors, Customer AI can tell you what a customer is likely to do and why. Additionally, marketers can benefit from Customer AI predictions and insights to personalize customer experiences by serving the most appropriate offers and messaging.
Understanding Customer AI
Customer AI is used to generate custom propensity scores such as churn and conversion for individual profiles at-scale. This is accomplished without having to transform the business needs to a machine learning problem, pick an algorithm, train, or deploy.
Customer AI is built to:
- Provide high accuracy customer propensity models for stronger segmentation and targeting.
- Help with understanding the influential factors and likelihood behind certain customer behaviors.
- Provide customizable options for your company’s unique use cases and data.
- Enhance Real-Time Customer Profile with customer propensity scores such as churn and conversion.
- Enhance customer profiles with influential factors for propensity scores.
- Create segments of customers based on influential factors and propensity scores.
Customer is not built to:
- Customer AI should not be used to predict dynamic pricing, or the price point at which the customer is going to make a purchase.
- Customer AI cannot determine whether giving an offer will make the customer more likely to purchase an item. While you might decide to send discount offers based on propensity scores, it’s not necessarily the best way to convert those customers.
- Customer AI is not a product recommendations tool. If you have thousands of SKUs, do not use Customer AI as a proxy for a real product recommendations solution like Adobe Target.
- Customer AI can’t predict which stage of the buying Journey the customer is in, for example, if they are in “awareness”, “consideration”, “purchase”, or “retention” stages.
- Don’t use Customer AI to determine customers who are likely buy a product launching in the future. This requires certain success events to be present in the past for Customer AI to successfully train the machine learning algorithm on your data.
The following video is designed to support your understanding of Customer AI.
Transcript
Hi, I’m Hetal Chandria, Senior Product Manager. In this video you will learn how marketers can leverage Customer AI to generate customer predictions. We will cover what Customer AI is, its use cases and benefits, the high level architecture, how it can be used with other Adobe applications. Customer AI generates customer predictions at the individual level with explanations. With Customer AI, we can tell you what a customer is likely to do and we can also tell you why with the help of influential factors. Marketers can benefit from Customer AI predictions and insights to personalize customer experience by serving the most appropriate offers and messaging, whether it’s a new prospect that you would like to convert or an existing customer you would like to upsell.
Marketers benefit from high accuracy customer propensity models for stronger segmentation and targeting. Understanding the influential factors and likelihood behind certain customer behaviors. Customizable options for your company’s unique use cases and data. Customer AI brings you AI-as-a-Service that can be easily configured to allow you to personalize your customer experiences intelligently. Grow and activate new customers. Like for example, sending promotional emails to users who have a higher chance of conversion. Retain, proactively reduce churn for example segment high risk users for personalized treatment. Engage, increase engagement with existing users to drive product usage. Enhance, personalize customer experience. The output of Customer AI can be applied to a variety of industries as long as the outcome of interest can be defined. For retail, Customer AI can predict a customer’s propensity to purchase products. For financial services, it can help predict who will open a new bank account. For media and entertainment, it can help you predict which users will churn. That is cancel an active subscription or downgrade their service. Even though we have gone over only three verticals, Customer AI can be used by any business which has measurable business outcomes. Now, let’s look at the use cases not supported by Customer AI. Customer AI cannot be used to predict dynamic pricing, or the price point at which the customer will purchase. Customer AI cannot determine whether giving an offer will make the customer more likely to purchase an item. While you might decide to send discounts, based on propensity scores, it is not necessarily the best way to convert those customers. Customer AI is not a product recommendation tool. If you have thousands of skews do not use Customer AI as a proxy for a real product recommendation solution like Adobe target. Do not use Customer AI to determine customers who will buy your product launching in future. It requires certain success events in past, so successfully train the machine learning algorithm on your data. Let’s next take a look at the high level workflow. First, with the help of professional services, the customer data is ingested, mapped and transformed into XDM and stitched in Experience Platform. With the appropriate data governance in place. A marketing analyst will now be able to easily configure the desired predictions for any specific business objective in mind. Then after training and scoring, powered by Intelligence Services, the predictive scores are written back into Experience Platform for marketing analyst to operationalize. There are three main ways of operationalizing the predictive insights. First, they can consume insights through a dashboard provided in the Intelligence Services interface. Second, they can activate Predictive Intelligence into various applications across Adobe Experience Cloud or Services on Experience Platform. For example, Real Time customer data platform and beyond across external applications like call center. And finally they can power through custom dashboards built in Business Intelligence tools. Customers can create segments leveraging the propensity scores within the segment builder and these audiences will be available for use on Adobe Advertising Cloud, Adobe Audience Manager, Adobe Campaign and Adobe Target. Customer AI scores can also be uploaded in Adobe Analytics for exploratory data analysis. All real time customer data platform customers will be able to create segments leveraging the propensity scores and activate them via destinations. That’s a quick introduction to Customer AI. -
How it works
Customer AI works by analyzing existing Consumer Experience Event data to predict churn or conversion propensity scores. Adobe realizes that the definition of churn and conversion is not uniform across all the use cases and for this reason, you have the ability to define custom target goals as a set of conditions. You can configure the predicted goal as long as the event of interest is present within the input Consumer Experience Event data.
Next steps
You can begin by following the getting started guide. This guide walks you through setting up all the required prerequisites for Customer AI. If you already have all your credentials and data ready, visit configuring a Customer AI instance. It provides steps for using Customer AI.
