SAS for DIFM & DIY Customer Recommendation Systems
If you’ve ever used Amazon, Netflix, or YouTube, you’ve experienced the value of recommendation systems firsthand. These sophisticated systems identify recommendations autonomously for individual users based on past purchases and searches, as well as other behaviors. Customers get algorithmic recommendations on additional offerings that are intended to be relevant, valued, and helpful. Consumers can use recommendations to:
- Find things that are interesting or useful.
- Narrow a set of choices.
- Explore options.
- Discover new things.
Marketers can enhance offers that proactively build better customer relationships, retention and sales. For example, organizations typically realize:
- Stronger customer relationships by providing personalization.
- Higher engagement, click-through and conversion rates.
- New opportunities for promotion, persuasion, and profitability.
- Deeper knowledge about customers.
SAS’s vision is to help marketers be effective through analytic techniques. Consumer preferences are hard to predict. By using SAS’s deep library of algorithms, recommendations can automatically shapeshift to meet the demands of the consumer, and create brand relevancy through data-driven personalization. The content shared in this article for the application area of recommendation systems represents an exciting opportunity to showcase technology and approaches across SAS Customer intelligence 360 and SAS Viya for different profiles of users (marketers, analysts and data scientists) in the context of customer analytics. While we attempt to maximize diversity, it should be noted a reasonable, non-exhaustive number of examples will be shared.
To begin, no matter what type of marketing or customer experience team you are a part of, there is likely a business leader wrestling with a challenge. Typically, this translates into a question (or set of questions) that inspire research, projects or new assignments for others. Ultimately, the finish line is a decision. Because if a decision is never made, no value can be derived as a result.
SAS provides functionality without constraint through a no/low/high code software experience. Across the user spectrum, SAS enables DataOps, ModelOps & Customer Experiences. This includes, but is not limited to, data access, preparation, exploration, reporting, machine learning, AI, model management, decisioning and multi-channel journey orchestration. Our promise is to help users overcome business problems by gaining deep customer knowledge that extends to action by seamlessly enhancing the activation of customer data.
Now, recommendation analysis leverages product, content, and digital data to uncover hidden patterns in order to identify related products or content to surface to customers for personalization. Recommendation analysis has long been used for personalized shopping, as well as streaming providers in personalizing viewer content. This analysis easily extends to other types of experiences and data to build more customer relevance, especially with the infusion of AI/ML techniques. Brands can deliver relevant and accurate predictions of what customers will buy or view next based on prior behavior.
SAS delivers decision-oriented solutions that accelerate the timetable to actionability, as well as customizable modeling recipes and patented procedures that optimize the in-house AI talent your brand employs. Let’s gently walk through SAS recommendation analysis & orchestration capabilities through a few demo examples.
Chapter 1: Do-It-For-Me (DIFM) Recommender Targeting
In SAS Customer Intelligence 360, users can create tasks (Web, Mobile, etc.) that display different creatives based either on a product being viewed or a user’s behavior. There are two methods for delivering recommendations to users. User-centric recommendations take a user’s behavior into account. Product-centric recommendations are based solely on a product.
The Chapter 1 demo video below will be in the context of the financial services industry. After a brief introduction, a customer visit to a brand's website will be exemplified, and an identity event will occur when logging in. Subsequently, the customer will receive analytically-recommended content. We will pivot and show how the DIFM recommender task was configured in SAS Customer Intelligence 360.
Need a step-by-step tutorial to configure a recommendation task in SAS Customer Intelligence 360? Check this out.
Chapter 2: Do-It-Yourself (DIY) + Do-It-For-Me (DIFM) Recommendation Analysis
Martech industry solutions frequently offer "easy-button" or automated analytical solutions that over-promise the potential of machine learning and AI. In the end, for readers who have used DIFM features, they automate analytical model templates with limited abilities to accommodate customization. Our viewpoint at SAS is the availability of DIFM features in software is important, especially in the absence of any analytical enhancements to a brand's present-day use cases. However, the desire to incrementally improve on DIFM technology features allows us to pivot to DIY approaches in recommendation analysis.
SAS Visual Data Mining and Machine Learning on SAS Viya enables no/low-code users to explore, investigate, and visualize data sources to uncover relevant patterns, as well as extend these capabilities by creating, testing, and comparing models based on patterns discovered. Users can export the score code or analytic store, before or after performing model comparison, for use with other SAS (or 3rd party) software products to put models into production.
SAS enables users to rapidly create recommendation models in a web-based interface. One example involves factorization machines, which is a predictive model that creates a factorization model. By modeling all variable interactions with factorized parameters, factorization machines are able to handle large, very sparse data and can be trained in linear time. A common application of factorization machines is for recommendation engines. A factorization machine can consider all items that a user has rated, and then predict ratings for other items.
The Chapter 2 demo video below will be in the context of a business or marketing analyst, with a preference for no- or low-code software interfaces. The intent will be to show how a DIY approach to performing a recommendation analysis can be accelerated and improved by DIFM features to support the analysis workflow.
Chapter 3 Preview: Do-It-Yourself (DIY) Champion-Challenger Recommendation Analysis
A sampling of native recommendation analysis algorithms available in SAS include regularized & non-negative matrix factorization, k-nearest neighbor, bayesian personalized ranking, factorization machines, data translation w/ optimal step-size, slope one, market basket, link analysis, and the list goes on. Additionally, SAS supports usage of open source (Python/R) recommender packages. Given the high volume of algorithms to select from as an analyst, SAS enables champion-challenger recommender modeling prior to deployment.
In a forthcoming SAS Communities article, the Chapter 3 demo will exemplify this concept using three algorithmic approaches leveraging factorization machines (FMs), bayesian personalized ranking (BpR) & data translation w/ optimal step-size (DTOS).
The use cases for recommendation systems are expanding every day, across the entire martech industry. We look forward to what the future brings in our development process – as we enable technology users to access all of the most recent SAS analytical developments. Learn more about how SAS can be applied for customer analytics, journey personalization and integrated marketing here.
