Part 2: Process and Culture
This is the second article of a series called Lessons From 1 Year Of Experimentation.
Allow me to tell you about one of my favorite experiments (…and then I’ll tell you why we probably shouldn’t have favorite experiments, but more on that in a minute.)
About a year ago, we implemented a new experiment to add a checkbox to the patient payment app to allow users to unsubscribe from receiving paper statements in the mail. The goal was to reduce paper costs and give patients a chance to signal their preferred way of communicating with us. When we reviewed the results, it was thrilling to see that more than a quarter of patients opted to unsubscribe from paper. But a few months later, we realized that the unsubscribe metric alone wasn’t showing us the full picture. Even though we had seen good success getting patients to opt out of paper, the longer term metrics revealed that these paperless patients had overall lower engagement and lower collection rates, specifically when they received bills from providers with less brand recognition. These longer term impacts (less engagement, fewer bills resolved) were more consequential and revealed that this experiment did not move the needle in a positive direction for these types of providers. Providers with strong brand recognition and mostly recurring patients, on the other hand, fared just fine and still showed consistent levels of engagement for paperless patients. Thus, we rolled out this feature broadly for these types of clients (but not to our entire patient population).
We start all of our experiments with background data analysis to lay out the problem we want to solve, as well as prioritize solutions based on expected impact, reach and effort. But this data doesn’t always give the full story. Our patient population is constantly changing as Cedar’s client-base grows and some features, like the unsubscribe checkbox, end up not delivering the expected outcome for everyone we wanted to touch.
Sometimes, that’s not easy to accept. When I worked in finance, a former mentor taught me to “never fall in love with my trades.” The lesson has stuck with me: don’t become so attached to your ideas that you lose the ability to evaluate them objectively. In our case at Cedar, we should take care not to fall in love with our designs and experiments.
Of course, this is easier said than done. So one way we can counteract emotional decision-making is with a strong process. Sharing updates and results broadly and regularly to the entire Cedar team forces us to be accountable and make tough decisions. It’s important that we never be afraid to cut designs and features that do not move our metrics in the right direction. We are all accountable to improve the patient experience.
At Cedar, the patient is at the center of our decisions. Building products that innovate with thoughtful designs and features takes a lot of people. Everybody at the company, from our patient advocates, who help patients with their questions, to our commercial and makers teams, hold some pieces of knowledge about our patients. Using the sum of this knowledge for problem solving and idea generation is crucial.
The proof is in the process.
Here’s how the magic of an experiment actually comes together: the product team raises an issue that needs to be addressed. Our data science team will conduct data analysis to get a better understanding of what’s actually happening. At that point, we assemble a cross-functional brainstorming group with participants from the product, design, engineering, data science, commercial and user success teams. In this group, we debate our hypotheses, ideate potential solutions and then prioritize them. Once we’re aligned on our approach, we’ll get the experiments running. We’ll communicate the results to the entire Cedar team periodically throughout the experiment and once it’s wrapped via our monthly internal report and all-hands events. Based on the results, we go back to the drawing board, try different solutions or personalization.
Cultivating creative problem-solving.
Building this culture of experimentation is critical and we instill this ethos in our team with rituals both big and small. Last February, we organized Cedar’s first hackathon where all employees were encouraged to participate. The event didn’t require coding skills — it was a product and design hackathon designed to spark creativity. We created teams by mixing up people from different functional areas across engineering, design, data science, marketing, customer success, partnerships and account management. It was incredible to see people from all teams contributing — from developing a product idea to the final pitches. More importantly, for a day and half, two-thirds of Cedar worked together using their own skills and background to solve problems for patients. The whole experience was an amazing training in cross-functional work.
All of the aforementioned internal structural and cultural foundations we’ve established at Cedar enable us to most effectively build a personalized experience for patients. There is another key element that we haven’t talked about yet: ethics. Ethics are the foundation that guides our decision making about what experiments to do, not do, and how to responsibly implement changes to the patient experience. It’s a critical topic and it deserves it’s own separate post. So stay tuned.
Cedar is a healthcare financial engagement platform for hospitals, health systems and medical groups that clarifies and simplifies the financial experience for patients. If you’d like to learn more about Cedar, visit www.cedar.com or click here to join the mailing list. You can view our open roles at www.cedar.com/careers.
