Welcome to the BlueConic Doer Spotlight, where we highlight the hands-on doers using BlueConic to take their data strategies to new heights.
Today, we’re thrilled to hear from Ryan Nakashima, Director, Product Management, Subscriptions at Hearst Newspapers. Ryan reveals how he is harnessing the power of BlueConic to drive transformative results in both subscription management and personalization.
What's your role?
I am the Director, Product Management, Subscriptions at Hearst Newspapers.
What are the top three things you envisioned accomplishing with the BlueConic CDP?
We had three primary goals:
Develop a robust first-party data strategy
Deliver highly personalized experiences
Shape and optimize the front-end user experience
What kind of results have you seen since working with BlueConic so far?
BlueConic is a fundamental component of our technology stack for both subscriptions and personalization, and it is increasingly integral to our advertising infrastructure. Our multimillion-dollar subscription business relies on BlueConic’s content metering tools.
With the effectiveness of this successful use case starting to plateau, we are now looking for BlueConic’s help in using sophisticated machine learning models to make our paywall experience smarter. Our goal is to maximize revenue by optimizing customer journeys based on their propensity to take certain actions, such as paying for unlimited access, registering, or signing up for a newsletter.
What’s the most exciting thing you’ve done with your data using BlueConic?
Our current machine learning project is the most exciting so far and we have invested a lot of energy into its implementation. It focuses on using a next-best-action model to evaluate a user's propensity to take certain actions and then optimize their user journey to derive the most revenue for the company based on their profile.
The second most exciting is the work we've done on the personalization of recirculation modules, where we show a user recommended stories based on their past behavior, and hide stories they've already read. In testing, this has already produced a statistically significant positive lift.
In a big win from last year, we implemented an article-gifting feature that relied on BlueConic where subscribers could share gift articles with friends and family, getting the recipients past our paywall, as long as they registered an email address. Not only did this greatly increase our acquisition of known users and email addresses, it also lifted the tenure of our existing subscribers, and drew new subscribers into our ecosystem via a social growth loop on our own platform.
If you could give one piece of advice to a team just starting with BlueConic, what would it be?
I would say start with the easiest, lowest-lift, low- or no-code solution as you can and get it up on your platform as quickly as possible in an A/B test. It will not only give you something to improve upon, but also enable you to generate small wins quickly so that you can start to see a return on investment in the platform. That will spur further use cases, further development and better return. One of the key benefits of the platform is that it speeds your ability to go to market with new features, so I'd recommend leaning into that speed to improve the pace of iteration and innovation.
What’s next on your BlueConic journey?
We have several major initiatives involving BlueConic coming up. One is to enrich data on our users through email- and IP-address matching from a third-party source to benefit advertising use cases as well as better subscription offer targeting.
The second is to create dynamic user experiences using a machine-learning model known as a multi-armed bandit. We'll choose the path for the user based on their likelihood to take different actions such as pay for unlimited access, register for free, sign up for a newsletter or simply read a story with ads on the page. In creating this model, we're leaning heavily on the BlueConic data science and professional services team to create the model and help us manage the project while we evaluate success and work with our developers to be sure the right data inputs are making it into the system.
Finally, we're delving much deeper into personalization now that we've found success matching BlueConic's algorithms to our front-end systems, which will make it easier for our users to find stories that interest and matter to them.
