personalized artificial intelligence algorithms for personalized member engagement

Algorithms for Engagement: AI to Personalize Your Email Newsletter

Constructing the most engaging newsbrief

At rasa.io, we believe that newsletters containing relevant content create higher subscriber - or member - engagement. Put simply: people interact with email when it brings them the information they want. We have seen this happen over and over, and our case studies document the impact.

And when your members rely on you to deliver relevant information, you become a part of their daily habit. They depend on you.

Personalizing your emails with AI algorithms

Our role is to build newsbriefs containing the personalized recommendations that will be most likely to engage your membership. To do this, rasa.io employs several different AI algorithms, blending information to bring a relevant curation of articles to each and every individual member, each and every day.

Two of the best known models for generating recommendations are the Content Filtering and Collaborative Filtering models.  

Content filtering model

Content Filtering is a model that considers the preferences expressed by a reader, and attempts to locate new content that aligns with those preferences. A user who has presented a consistent pattern of reading articles about International Finance and English Premier League Soccer will likely be interested in a newly published BBC article about the revenue sharing arrangements of a new Television Rights deal for the league. With sophisticated, ai-generated, article-based information, rasa.io can identify new articles are most likely to be relevant for each member based on what they have read and enjoyed previously.  

Pandora music service looks at many different facets of the songs a user has liked: the tempo, the tone, the pitch. With a history of such data for a user, new songs can be identified and recommended.

This technique offers a powerful ability to match content to a user’s history, but comes with a risk of placing people in a “Walled Garden,” a world where they come to see only the articles that already fit their world view. So, we incorporate multiple models to prevent that from happening...

Collaborative filtering model

Collaborative Filtering is a model that generates recommendations for one reader based on the similarity of that person’s behavior with the behavior of other reader’s in your association. For example, many readers who read about English Premier League Soccer have also read about the World Cricket League.  

Our algorithms will use this information to suggest an article about the Scottish Saltires for a someone who has previously read articles about the Premier League. This model builds recommendations for a reader based not on their own content preferences, but on the behavior of other similar users.

As another example, Netflix movies streaming service leverages member ratings of films to help suggest other movies for you based on the movies you have watched. The more ratings people provide, the more chances there are to identify common patterns of behavior, and generate recommendations based upon this. (You can register for our upcoming rasa.io webinar on How to be the Netflix of Associations from this page!) This technique offers the ability to stretch boundaries for relevant content, by leveraging the preferences of other readers, and not evaluating just the preferences of a single reader.

Using rasa.io AI to better inform YOUR world

At rasa.io, our Core Purpose is To Better Inform The World, so we cannot rest with just a single algorithm. The best information comes from combinations of multiple algorithms. We strive to bring your membership not just individual articles that are relevant, but an entire newsbrief that they will look forward to receiving.

You can begin to engage your members in a meaningful, personalized way today. Use rasa.io artificial intelligence to automate and personalize your member newsletters. Get started within a matter of days, and increase engagement now.


computers showing association members engaged using scalable artificial intelligence

Building Scalable Artificial Intelligence Systems That Keep Your Association’s Members Engaged

Artificial Intelligence software systems demand innovative development and deployment approaches.  The techniques that have been honed over the years by companies and software engineers do not scale effectively in the presence of AI systems. We understand these challenges, so we have built our AI solutions using methods that give us the ability to improve our systems quickly and deliver better content for your association’s membership.

Where we have come from

Traditional software systems have been built following a standardized set of practices that have evolved from collective experiences, gained by thousands of large and small companies, over many years. Software developers will quibble about the differences between Waterfall and Agile, Scrum and Kanban, but these different techniques and approaches represent variants on the same general approach: a degree of planning, followed by development to meet some specs, then testing to validate the adherence to the specs, and finally deployment and monitoring of A, B, C and D.

Do you measure turnaround time in hours and points with an Agile development cycle? Do you measure time in months with detailed FDD specs? The difference is in the scope of the steps (and in perceived chances of success with the difference in scope!), but the general approach remains the same.

For years, the primary push in this cycle has been to shorten feedback loops: get features to field as fast as possible in order to get feedback fast. Early feedback - bringing an idea from whiteboard to field - enhances the likelihood of success in the form of customer satisfaction. Unit and Integration Tests, Continuous Integration, Story Boards - many techniques have evolved to help development teams move faster with increased confidence in the features they deploy.

Where we are going

Artificial Intelligence and Machine Learning systems introduce new and different challenges in the software development and deployment cycle, challenges that demand innovative solutions.

First and foremost: Artificial Intelligence systems crave data; they need data from which to learn. One programmer’s adage says, “There are only 3 numbers: 0, 1 and 2. Everything else is just a generalization of 2.”  Many systems could be tested following that adage: enumerate a small number of conditions, then test those conditions and the corresponding edge cases. AI systems throw that adage out the window. These systems do not work without mountains of data to evaluate. The need for data introduces 3 specific challenges:

  1. Data Acquisition: We must first build tools to gather enough data to be able to evaluate an AI system. Until we have that data, we are limited in our ability to build systems to consume it.
  2. Development Time: Consuming data and evaluating it via AI systems can be computationally expensive. Development cycles that used to measure in seconds from deployment through testing may now be measured in hours: modify the AI system, deploy it, then run through the learning cycles.
  3. Result Evaluation: AI systems generate tons of predications given the massive amount of data pushed in. Those outcomes are based on the machine learning, which is itself based on the data consumed. In other words, there are too many outcomes to enable testing all of them individually. And the outcomes are not known apriori in a way that a Unit Test could be written to evaluate the model output.

rasa.io is paving a new path for associations to use Artificial Intelligence

The techniques software engineers have used in the past cannot scale to support the development of machine learning systems. At rasa.io, we have adapted our processes and tools to enable us to develop and deploy enhancements to our AI engine more rapidly. Our serverless platform, built using Amazon’s Lambda, EMR and RDS services, allows us to horizontally scale our platform, reducing the time required for our AI solutions to build their predictions.

Reducing the turnaround time for processing the vast amounts of available data gives us the flexibility to develop improvements to our AI engine faster. What this ultimately results in, is a sophisticated machine that informs you on what your association is interested in and thus how you can best communicate, on a daily basis, with your members.

We recognize that the growth of our AI is critical to the delivery of the best content for the members of your association.  We are committed to staying ahead of industry trends and technologies to ensure that our AI engine remains at the forefront of personalized recommendations. This allows us to get more relevant content delivered to your members and understand what motivates them. Learn more about how rasa.io can engage your members today.