Margaret, this is Kevin. I'll take that. The annual revenue profile per patient, we think, is going to vary based on condition and use case. We believe internally, or at least what we discuss internally, is for a type 2 non-intensive patient. For example, even if a full-time use, the revenue per year per patient is going to be a lower number because we're solving a different problem. We're not managing the lives and the health and safety of these patients, but we are making them healthier and we are giving a better experience than they have. And ultimately, we will take costs from the system. As far as what that number is going to be, I don't have a number completely. We have numerous models, and many of those depend upon the final use cases. I've also said numerous times, and I'll stick to this as well, I think the ultimate use case and the way we'll ultimately price it will be for full-time use for these patients. And if they choose to use it less often, that will be their choice. So on the health and wellness side, again, many, many more patients, a lower annual revenue number, but more interactions with more different people in a completely different distribution channel. With respect to the products, certainly, the core technology of sensing glucose, from our perspective, on the glucose side will remain the same. As those of you who listened to our Investor Day heard when I had my chat with Dr. Peter Atia, we talked about accuracy. And he said it's actually more important for his patients because if they only go between 70 and 120, 20 points off is a big deal. And they'd be making a wrong decision as far as their health and what they eat nutrition-wise, et cetera, within accuracy. So we know that fundamental core technology needs to remain accurate. Where we get into different products is the experience that we create, possibly even different wearables based on the category, different places to receive the data. Then as you look at the new ventures unit that Steve will be working in, then you look at software experiences for patients, data platforms to get data to these platforms better, possibly other things we could sense or measure and add to glucose. So we believe these markets are going to require something different. At the end of the day, one size fits all is not going to work for us as we try and expand and hit the goals we've laid out probably being in the future. We're going to have numerous experiences for these patient groups.