
Loading summary
Peter Simpson
Foreign.
Scott
Welcome back, everybody. Today I'm speaking with Jake Leach, President and CEO of Dexcom. Jake is joined today by Peter Simpson. He is the Executive Vice President and Chief Technology Officer. Jake, Peter and I are talking about the goosenecking issue with G7, different ideas around their support and replacement policies. We'll talk a bit about accuracy around G7, a little bit about 15 days, some adhesive stuff, and I even sneak in a question at the end about the G8 sensor coming up sometime in the future, but probably not as far away as you think. While you're listening, please remember that nothing you hear on the Juice Box Podcast should be considered advice, medical or otherwise. Always consult a physician before making any changes to your healthcare plan or. Or becoming bold with insulin. And if you're new to the podcast, check me out@juicebox podcast.com we have content for everybody living with type 1 diabetes, adults and caregivers alike. Juicebox podcast.com are you newly diagnosed? Check out the Bold Beginning series. And if you're working on making things better for yourself, check out the Diabetes Pro Tip series. There's that and much, much More available@juicebox podcast.com Listen on Apple Podcasts, Spotify, or wherever you get audio.
Jake Leach
Hey, Scott. Jake Leach. Pleasure to be here. I'm the president and CEO of Dexcom.
Scott
Jake, I appreciate you coming back and you brought somebody with you today. Who is this?
Jake Leach
This is Peter Simpson. I'll let him introduce himself. But one thing I'll tell you is that. So Peter and I have been working together at Dexcom since the day I joined, and he is one of the founders of our technology. An amazing person, but also an incredible scientist and engineer, and really a lot of the advancements you've seen, he has led almost all of them when it comes to sensor performance.
Scott
Oh, wow. Well, it's nice to meet you, Peter. Also, there's no video here, Peter, so this is just audio. You don't have to sit up and smile the whole time if you don't want to.
Jake Leach
All right.
Peter Simpson
Awesome. Scott, it's a pleasure meeting you. My name is Peter Simpson. I'm the Chief Technology Officer here at Dexcom. And as Jake mentioned, I've been here for going on 24 years. So I've been involved in all of the generations of the product and really seeing the evolution both of cgm, but also the impact that all these new technologies have had on people with type 1 diabetes. So it's been amazing to be a part of that evolution and really excited about the future as well.
Scott
I would Imagine, too, you both have a perspective that is almost uncommon in a business like this. Right? I mean, you're. From the beginning till now, you guys have seen it all. You have kind of an institutional memory of it. How does that help you?
Jake Leach
It's, you know, it's, it's. It is really a significant benefit. One because, you know, we're intimately familiar with all the things that have led to the success of the company and the success of the technology. And, you know, it's not like you're trying to learn that while also trying to figure out where the next, you know, place we want to go is. And so I think that is. That is a significant benefit. And, and also, you know, when it comes to my role, a big part of what I do is making sure that we've got the right culture within the company. And, you know, there's some parts of our culture that have made us successful over time that we want to keep, and then there's parts of the culture you want to continue to evolve, particularly as we scale. You know, it was very different days when we were, you know, Peter and I actually used to go down and watch them ship the sensors out the door for that day to the users, right. And we're talking about hundreds of sensors, right? And now we're shipping millions and millions of sensors every week. So it's very different in the things that are important when you scale, you need to have those capabilities, and how you look at it is all continuously improving and all that. So that is a big part of what I do. And I lean on the history and the experience. And the other thing too is things aren't always easy. I mean, it's been many decades of very hard work and really persevering through lots of challenges. You know, when things happen and don't go as planned, you kind of fall back on the. Yeah, this, we've been here before, and this is how we're going to fix it.
Scott
What's it like to try to keep that culture of patient, first worrying about people, worrying about customers, but then scaling up in the way that you did? I don't. You talk about how many sensors you ship a year now versus back then.
Jake Leach
Well, we're well north of 100 million sensors. We have more than three and a half million active users, you know, in using the sensors today. And that number grows. You know, we grew that by 20% last year in terms of the number of people that use Dexcom every day. So we're going to continue to grow it. You know, it's actually, it's interesting, the patient centered focus for Dexcom because it's so core and it's always been there. We've always been a customer focused company. That's actually the easy part in terms of it. I don't, you know, you don't have to convince anybody. People come here because they are passionate about our mission and they want to, they know, you know, what we're trying to achieve for our users. I think the hard part is making sure we always show up that way. Like the intent is there, right? The intent is to always be user focused and we meet the needs of user. But when you scale and grow, you have to start changing the way you do things to better match the scale. And there's lots and lots of examples from manufacturing all the way through customer support where we need to keep evolving and improving and we can't keep doing things that what worked for a smaller number of users doesn't work when you get to the large number of users for lots of different reasons. So I think that's the part that is harder but very achievable. And you know, as the new CEO is like, this is something we're going to show up. We're not adding to people's burden, we're relieving the burden of diabetes or at least helping to relieve that. We should never be adding to your burden as a user.
Scott
Can that be a pothole? Like you're, you're focused on who you are, you know, what you think you're doing, but somehow it gets away from you and you don't realize it right away or is it a thing you see, like we got to get to that but we're so busy when those things ebb and flow. How do you keep your eye on them when they're moving? You know what I mean? Because your intention's the same. So.
Jake Leach
Yeah, yeah, I think it's really the feedback mechanisms that's the number one thing that helps you understand how you're doing is real, the appropriate feedback mechanisms. And it's actually one of the reasons why we started the Customer Advisory Council is because we have many feedback mechanisms we've developed over time. But I was, you know, wanted to hear more from community around, you know, all the things that we're doing well that we can double down on but also the things that we're not doing, you know, where we're not meeting users needs, where we're not showing up the way our users want us to. And so that we put the Advisory council together, which You're a participant in Scott, which I really appreciate you doing that. And it was, you know, it's a diverse group, and we wanted to hear straight talk, like, what are we doing? Well, what do we not? So that we can, you know, improve it. And the goal is to make sure that the feedback from that council shows up in the actual experiences our users are having, whether it's product or support or, you know, anything around your experience with Dexcom. We want to. Want to make sure it's meeting everybody's needs. You got to have those mechanisms.
Scott
Well, I can say for sure that you wanted to hear the feedback, because I didn't just bring my own. I brought feedback from the audience, which was pretty extensive. And I was invited back to the second meeting, so I figured you were
Jake Leach
actually wanted to hear about it. It's working.
Scott
So that's what we're here to talk about a little bit today. Right? You guys are putting out a report based on these meetings. Can you explain to people what all this is and what did it turn up?
Jake Leach
Yeah. Hey, Peter, do you want to take. You want to take that one? Yeah.
Peter Simpson
We've had two of these meetings so far. First one is to make sure that we were listening, and then the second one was where we played back what we were hearing and gave a little bit of information about how we were responding to some of those commentary. We're about to publish this report out that basically captures that and talks about some of the commitments that we're making as a company to our customers about the product performance, the transparency, the communication, and even things like our replacement policy and such. I think it's very important for us to listen to our customers, but also to know that they are listened to. And I think that this is just a piece of that, that we can communicate out to make sure that they know that we are listening to them and really want their feedback.
Scott
How can you kind of close the time between when the customer realizes something's going on, but you don't realize it, and then you have to react and I would imagine figure out what to do and move forward. And there's no communication in that time, which I don't even know what you would be communicating. I use the. Like, the goosenecking is a great example. Right. Like, I think arden was using G7, like, maybe week one, and one of our first sensors, it happened to. And I don't know. I just. I said, call it in, put on the next one. I didn't even think twice about it, really. I just. And then Suddenly you hear other people talking about online, but how long does it take you to hear rumblings like, and what would you say to people who would say, well, why is that not something you already knew was going to happen? How does something in a manufacturing supply chain, how does that happen? Bring people into that idea?
Jake Leach
So what I would say, Scott, first of all, is for that particular deployment issue, we saw it, we see, you know, because our devices are connected and we constantly monitor feedback from the performance. We actually saw it start increasing early in 25, so beginning of last year we started seeing it increasing and we saw that signal. What we didn't immediately understand is what was causing it. So we jumped on it and the tiger teams put together and specifically I'll let Peter describe a little bit about what we found there. But that was an example though of something that we monitor. But what we've also continued to do is that experience taught us that there's other aspects of the product that we can be monitoring in real time in the field. And so we've continued to enhance our analytics capabilities so that, you know, with all of, if you think about data technologies, artificial intelligence, ability to spot patterns that maybe we don't immediately spot as a data scientist, that is some of the technology we're trying to apply to be able to monitor performance. Because, you know, we have so many users, we can see when something is going out of, out of range and then we can jump on it. But that one sensor deployment issue challenge that we had, happy the team jumped on it, happy that, you know, we've corrected it moving forward, but that was a hard one. Peter, do you want to describe a little bit about that?
Peter Simpson
Yeah. So what you're referring to is, I think you've called a gooseneck. This is where our sensor, during the deployment process, it brings the sensor and the needle down to under the skin and then spring retracts that needle, leaving the sensor behind. What happens in those cases is the sensor gets pulled back with the needle and then it's no longer in the body. So you essentially have an out of box failure, a failure for the sensor to connect or to sense the glucose. Now our algorithms are able to detect it. So it's not a safety concern, but it is a usability concern. It's a very, it's a serious one. You're expecting this product to work and that's your expectation. And we need to live up to that expectation. So when we see those failures, even if it's a small amount, we take them very seriously. When we detected into the field was very early on and we put together what we call tiger teams and these are groups where we take them off of their projects. They're dedicated to figuring this thing out. They're meeting on a day, sometimes twice a day basis to understand what's going on in the manufacturing floor. Is there any changes that were made that could cause this to spike up? And over time we did figure out what was causing it. And a lot of it, similar to other issues that we've experienced in the past, is when you're scaling, you get more variability in your process and when you identify that variability, it's pretty straightforward, or let me not say easy, but you can, you can fix it once you've identified where it's coming from. And that's what what happened in this case is the tiger team was able to figure out what the root cause was and gotten it back down to a very low level. We still have some improvement to do and we've got technology that's going to come out later this year that's going to make it even, even, you know, to the best level that we've ever had in G7. And so while these teams work really aggressively, we have to find the issues quickly and we have to respond to them quickly. And then the most important thing is to get those improvements out into the field so that the customers are getting the best product every time with every sensor that we're making. I think what Jake mentioned and through this process we've learned a lot as a company in terms of being just really on top of everything that's going on in the field. And Jake mentioned some of the improvements that we're implementing where we're doing real time monitoring on not just that one, but dozens of other attributes that we're monitoring in real time. And as soon as we see something that's, that's out of just slightly off, we'll react, we'll get a team that's on it and really jump on these problems before they hit the field. So I think that's what we're, we're continuously improving and we're going to be working harder and harder and doing more of that going forward to make sure that every product is meeting the customer
Jake Leach
expectations because that's what's required a couple of follow ups.
Scott
So what you're saying is it's only really a problem if you don't learn something from it. And so this, it showed you a way that you could actually track other things as well. So yes, it was a problem, but you're going to get something out of it.
Jake Leach
Yeah.
Scott
And I want to tell you and Peter, if you have any kind of time in the future, I saw the explanation of how the goosenecking happens. I would sit down and geek out for an hour talking to you about what you figured out and what was happening. I find it really fascinating and I don't know if everybody would, but, you know, in that council meeting, when you guys came back in that second. In that second meeting and said, oh, here's the answer to how that happened, I thought that was fascinating. I think the other thing that was fascinating was a description of how the interstitial fluid and the wire works. If you want, I'll do a retrospective
Jake Leach
series with you about that.
Scott
I'd be so interested to talk about that, but I don't want to pressure you. So. Okay. So you guys figured that out still, improvements made. Does that mean some people still might see this happen?
Jake Leach
Yeah, I think, you know, one of the things I loved about this project, you know, while it was challenging and we were. It was, we were trying to solve this issue, it led to new innovations that are going to drive the improvements even farther. So we're well back to kind of historic levels for this failure mode, which is very, very low, but there's even more technology teams working on to drive it even so, even further improvement beyond where we've historically been. So I think that's one of the things that's exciting about anytime you learn something new ways to improve the product and apply it. And so, you know, it does happen very rarely still, but we want to try and drive as close to zero as possible.
Scott
Let's move on to like, support and replacement stuff, because I know you have a bunch and I have a couple of questions here. I don't want to keep you guys too long. What's changing about how this is going to work? So break it up into two pieces for me. Is support changing, improving, being different going forward from stuff you learned from this council? And what about how you're going to replace devices when they. When they fail?
Jake Leach
Sure. So a couple of things I'd add there is that, yes, our support teams are continuing to evolve. One of the things that we've done is the new MyDexcom account, as well as the tools that our agents have to be able to interact with our users around their experiences are continually improving. And the goal there is to ensure that the user has the right experience from kind of an empathy perspective and also the information provided to the user by our agents. The tools that we've been launching and we're continuing evolving, but we've done some pretty major new launches in the last six months here. Should result in more consistent experiences and more experiences meeting expectations. There's a lot of great experiences out there, but we tend to hear more about the ones when they don't go right. And that's one of, you know, there's always a learning there for us in terms of what we could do better. Also, sometimes patients felt like they're being asked the same information over and over. And a big part of that is just making sure we understand what the issue is. But we shouldn't be placing that burden on the user. Our system should capture that and then transfer it to the next individual who's working, working with the user. So there is a lot of change going on there and we're going to continue to innovate there. One of the three pillars I put forth when I took over CEO, as I said, we want to set the standard for excellence in customer service for the whole category. Right. So we want to be the best and we want to continue improving on the best. And I feel like that is a really important part of you making sure that customer. We have customers for life, Right. They should trust XCOM forever. We should always be improving, living up to that. So that is going on. The thing about the replacements, I think the main thing here is just being much more clear in our messaging around replacements. And, you know, unfortunately, there are instances where we have people calling us who are not users who are trying to get replacement sensors that they can resell on the open market. And that happens. Like you, you would be surprised, Scott, but it doesn't happen. I mean, I am surprised. It happens, right? And it's one of the things that we, we don't want that to happen, but we also don't want to cause an experience at the expense of a user. Right. So sometimes I think users feel like they're being interrogated. But a big part of that is we need to make sure you're actually a customer who needs an actual replacement. And so that's just one example of coming up with new systems to make sure we can authenticate users with less friction so that we can get them the sensors they need. And then the last thing I'll just mention is one of the things I think that's clear is the performance of the sensors. Some of the, what would be considered a sensor failure and is really just physiology. We're at the, you know, we're continually pushing the Envelope here on CGM technology with the most accurate, longest lasting sensor. Not all sensors last. When they don't last, we need to make sure the user gets a sensor. But you know, it's often not necessarily the quality of the sensor. So when it says sensor fail, there was nothing wrong with the physical sensor. It was just, it wasn't able to sense glucose anymore. For example, the body started, you know, encapsulating the sensor over those, that period of time that patient's wearing it. And that happens to people at different rates. It's uncommon, but it can happen. Right. So, but our number one goal is to make sure patients always have sensors. Customers always need their sensors. We don't want anybody to go a day without it. So we're continuing working on systems on how to enable and facilitate sensor replacements so that they always have that sensor and we've made some improvements and we're going to keep making improvements to reduce friction.
Scott
So not that it would be comforting to a person who it was happening to, but what's the reality versus the Internet? Meaning, like how many people put it on? I'm just gonna, I've said this before, I almost feel bad saying it out loud. My daughter wears a G7 for 10 days. Plus the bonus time works great the whole time. I always just assume that means her physiology pairs well with the device. And I just, I count it lucky. But for someone who says, hey, mine shuts off on the seventh day. I never make it past the eighth day. It gets fuzzy on the, whatever they say. I don't know how to ask the question, how many out of a hundred people is that happening to? Do you know those numbers?
Jake Leach
Yeah, well, it's basically, I mean, in our clinical trial data, which is just a sample of patients. Right. But generally our performance in the field, because our clinical trials are large enough we could generally see, you know, we track sense of survival. And I'd say the 15 day sensors are doing a little better than we did in the clinical study in terms of survival. You know, most sensors do make it out to 10 days, but not all right. Occasionally you have an experience where one doesn't. And most of the time, again, it's based on the physiologic environment. But our goal is kind of making sure that at least 80% of the sensors make it to the full sensor wear time. And you know, a lot of them, you know, be day 14 or maybe day 13 where we start to detect that it's not sensing accurately. But the vast majority today of users are getting, you know, the full, full sensor wear. But, you know, it's, it's not, you know, occasionally, you know, sensor falls off, something happens with the adhesive, something happens with the sensor not being able to sense glucose anymore because of the body. So I think for us it's around whatever experience you have as a user, whether, you know, every sensor goes 10 days or maybe or 15 days or some, you know, end early, we got to make sure that people always have that sensor they need. And that's when you sign up for Dexcom. That's, that's our goal to make sure you get, get the sensors you have and that you need.
Scott
So is this a little bit of the idea that people don't go to the Internet to celebrate things going well and that when you see somebody complaining, it's mostly because they're having probably part of it?
Jake Leach
Yeah, I mean, I think that those, you know, we don't often, I mean, myself as a consumer, I don't often celebrate all the successes I have with products. I'm usually, you know, more vocal about the ones that didn't go right. So I think that's, that can happen. It's valid.
Scott
I'm just trying to understand it even just from a personal experience, because, I mean, I'm telling you on one hand, I can count the amount of these sensors. It hasn't lasted 10 days. And I'm not saying we're doing something right or better. I just, you know, I just assume that's the situation. But. Okay.
Jake Leach
One thing I have seen though is over time, people who have experience with sensors figure out what works best for them. And so one of the things I do notice is that like, and we see like earlier when you're brand new to sensor therapy and technology, it, you know, you, you may have some learnings around how to make sure that that sensor stays adhered, stays functioning the whole time.
Scott
Right.
Jake Leach
And so whether it's over patches or different types of skin prep. Yeah, I think everyone kind of has, there's different options out there, which, you know, one of the things we can do, continue to do is make sure people are familiar with what the options are out there to help them continue to get the full wired life.
Scott
Can you explain a little bit about accuracy? So this is another thing we did in the council and it's, I think in the report too. So I guess with the 15 day or the seven day, it doesn't matter to me how you talk about it. But explain to people, day one accuracy and maybe give them a little insight on how the algorithm I'm going to say scrubs the data it's getting. I'm sure that's not the word you use, but how does the algorithm make sense of what it's seeing and why does what it's seeing. Why is it more sensible as time moves on after that first day?
Jake Leach
Peter's a expert at this. I'll let him.
Scott
Peter, did I ask that even reasonably clearly? Yeah, Scott.
Peter Simpson
So I'll explain a little bit of the challenges that our sensor can experience specifically on day one. So accuracy in general, obviously accuracy is super important for our customers and having consistent performance and something that we work on all the time. Our sensors on the benchtop are really accurate. I mean we're like in the 2 to 5% me already super accurate and very consistently accurate. It's when you go into that biological environment that you tend to see that variability and the inaccuracies start to occur. And from a biological perspective. And again, I want to. You mentioned it, but you know, what we see is these sensors that work exactly the same on the benchtop. You put them on 10 different people and you can get 10 different experiences. And so part of the technology on our side is to develop that interface, that membranes and the coatings that work on all the different biologies that people have.
Scott
Okay.
Peter Simpson
And again, most people don't have problems on day one, but some people do experience some inaccuracy on day one. And what's going on there is when you insert the sensor, it's essentially creating a very small, small wound pocket or a wound environment. So your body goes through this wound healing response. And what it's doing is it's kind of sending some cells in there to clean up this environment. And when that happens, these cells tend to be very metabolically active. So the sensor itself is working perfectly fine, but the body is interfering with the ability of the glucose to get to the sensor. So that's what you're actually experiencing when you have those challenges on day one oftentimes. And after that process takes its place, the sensor really then stabilize or that wound healing environment and the interstitial fluid is really stable for a long period of time. And then just to close it out at the end, why some people don't get their sensors to work last all 15 days is because the body over long term starts to encapsulate that sensor and basically starve the sensor of glucose and oxygen. The things that it needs to do. And in that case, what we do with the algorithm on both these cases is we are looking at that environment looking at the sensor signal to make sure it's of high quality. And if it detects that it's not of high quality, then it will either blank the data temporarily or it will shut off the sensor, especially at the end, the end use. And that's what it's saying, you know, I don't trust the data. I'm not going to present what may be inaccurate data to the customer. So the algorithm then shuts the sensor off. So that's what we're doing. And we continue to evolve both the sensor technology to make it work better and better in a larger population, but also advance the algorithms so that it can often then correct for those, those issues and improve the performance of the system.
Scott
I have a question I hope is a kind of a thoughtful way of asking about G8. I know we're just getting to 15 day G7, but if G8 was your first sensor ever, none of the other ones would have existed. What of today's conversations and or problems that we have would never have existed if G8 was first.
Jake Leach
I mean obviously G8 is the next big advancement for us from a glucose sensing perspective and we're adding multi analyte to it as well. I think some of the variability that we see in performance, it certainly is going to attack that. It has brand new technology in it that is proprietary, first of its kind, where we have additional signal that we can measure from the sensor to basically self adjust. You can think about it like calibrating, you know, you can finger stick, calibrate when needed or when you Want to for G7. And one of that's one of the unique characteristics of our sensors. Think of it being able to do that on its own. So it's slow, it's self adapting while you're wearing the sensor. You know, this is just an algorithm technology. It's algorithm and trunks that measure an additional signal. So it's very exciting. I think that would, you know, so basically some of the variability that folks have experienced in the past obviously wouldn't be there. You know, obviously the G8 adhesive is pretty phenomenal. We've learned so much over time with adhesives. You know, we're in the middle of launching the third version of the G7 adhesive. It's already starting to roll out to the field. People are experiencing it and we're already seeing the impact out there. We'll roll that across all, all of the portfolio and GA took those learnings and took it even farther. So quite a bit. I mean there's still a Little variability on the first day that you're going to see. You know, we always try to minimize that there'll be an improvement. But that's why we say there's always opportunity for us to continue to innovate here and, you know, provide the absolute best experience.
Scott
Jake, you're a pro. Because I skipped my adhesive question because of time and you still were able to fold it in there at the end. I thought that was fantastic. I didn't know.
Jake Leach
Sorry. Thank you for great.
Scott
You don't have my questions in front of you. I just. I had to pick something to skip for time and. Peter, I wasn't kidding. If you want to come back and really talk, like, dig deep about this stuff anytime at all, I'll do it while you're having lunch if you want. Doesn't matter to me if it doesn't eat into your day. But I really appreciate you guys doing this with me and continued success. We love the product. It really does change our lives every day here. I can't begin to say enough different ways about how it's made our lives better at my house. So thank you very much.
Jake Leach
Thanks for the time, Scott. And thanks. Thanks for being part of the advisory committee that's really helping us.
Scott
Oh, no, I appreciate it. I'm excited for the next one. It's been an interesting process to actually hear other people's perspectives about what works, where they have rough spots and the reasons behind the parts that they're worried about. It's interesting to come into a space where you think that the thing you're saying is going to be the thing everyone is saying. And then when that's not the case, you go, oh, wow, there's more to this than I even imagined. And I'm thinking about it way too much. Thank you again. I really do appreciate your time. I hope you guys have a good afternoon.
Peter Simpson
Thank you, Scott.
Scott
Take care. Hey, thanks for listening all the way to the end. I really appreciate your loyalty and listenership. Thank you so much for listening. I'll be back very soon with another episode of the Juice Box Podcast. If you're looking for community around type 1 diabetes, check out the Juice Box Podcast. Private Facebook group juice box podcast, type 1 diabetes. But everybody is welcome. Type 1, type 2 gestational loved ones. It doesn't matter to me. If you're impacted by diabetes and you're looking for support, comfort, or community, check out Juice box podcast type 1 diabetes on Facebook. If you're new to type 1 diabetes, begin with the Bold Beginning series from the podcast. Don't take my word for it. Listen to what reviewers have said. Bold Beginnings is the best first step. I learned more in those episodes than anywhere else. This is when everything finally clicked. People say it takes the stress out of the early days and replaces it with clarity. They tell me this should come with the diagnosis packet that I got at the hospital. And after they listen, they recommend it to everyone who's struggling. It's straightforward, practical and easy to listen to. Bold Beginnings gives you the basics in a way that actually makes sense. Have a podcast. Want it to sound fantastic? Wrongwayrecording.com.
Episode #1867: Dexcom CEO and CTO on G7, G8, and What's Changing
Host: Scott Benner
Guests: Jake Leach (President & CEO, Dexcom) and Peter Simpson (CTO, Dexcom)
Date: June 3, 2026
In this episode, Scott Benner welcomes Jake Leach and Peter Simpson from Dexcom for an in-depth, honest conversation about the continuous improvement of Dexcom’s G7 sensor, user experiences, manufacturing challenges, support and replacement policies, and a sneak peek at the upcoming G8 sensor. The discussion focuses on Dexcom’s approach to user feedback, lessons learned from recent deployment issues (like "goosenecking"), improvements in customer support, adhesive and accuracy innovations, and Dexcom’s cultural ethos of putting patients first.
[02:33–04:12]
Both Jake Leach and Peter Simpson have been with Dexcom since its early days, providing strong institutional knowledge.
Dexcom has evolved from shipping “hundreds” of sensors to now shipping “millions and millions of sensors every week.”
Maintaining a patient-first, customer-focused culture while scaling the business is both crucial and challenging.
“We're not adding to people's burden, we're relieving the burden of diabetes or at least helping to relieve that. We should never be adding to your burden as a user.”
— Jake Leach [05:31]
[06:06–08:17]
Dexcom has implemented feedback loops like the Customer Advisory Council (of which Scott is a member) to hear “straight talk” from diverse user voices.
Direct user feedback is driving actionable commitments, including improved communication, product changes, and a new public report on their findings.
“We wanted to hear straight talk, like, what are we doing well, what do we not? So that we can, you know, improve it.”
— Jake Leach [06:46]
[09:04–13:01]
"Goosenecking" refers to a failure where the sensor pulls out with the needle during insertion, resulting in out-of-box failure.
Dexcom detected an uptick in this issue early in 2025 through connected device monitoring and convened “tiger teams” to isolate and resolve the root cause.
The experience prompted even more real-field data analytics and AI pattern recognition to proactively spot and address variabilities.
“When we detected into the field was very early on and we put together what we call tiger teams... They're dedicated to figuring this thing out.”
— Peter Simpson [11:09]
“It led to new innovations that are going to drive the improvements even farther.”
— Jake Leach [13:57]
[14:34–18:16]
Dexcom is improving both tech and human dimensions: the new MyDexcom account and enhanced agent tools aim for seamless, empathetic user experiences.
Efforts to reduce repeated questioning and to authenticate users efficiently—balancing fraud prevention with genuine support.
Replacement policies will be clearer; systems enhancements are reducing friction for sensor replacements when failures happen.
“One of the three pillars I put forth when I took over CEO, as I said, we want to set the standard for excellence in customer service for the whole category.”
— Jake Leach [15:41]
[18:16–21:22]
Most users get the full sensor wear time (e.g., 10 or 15 days), but sometimes sensors fail early due to individual physiology, not device defect.
Dexcom tracks "sensor survival" data and aims for at least 80% full wear time in their user base.
“Our goal is kind of making sure that at least 80% of the sensors make it to the full sensor wear time.”
— Jake Leach [19:38]
Users often discover and optimize wear strategies (overpatches, skin prep) through experience.
[21:22–24:36]
Sensor performance is highly accurate on the benchtop; biological variability introduces real-world differences.
Day-one challenges are usually due to the body’s wound healing response, temporarily affecting glucose transmission to the sensor.
Dexcom’s algorithms monitor signal quality, blanking or shutting off data when accuracy can’t be ensured.
“Our sensors on the benchtop are really accurate... It's when you go into that biological environment that you tend to see that variability and the inaccuracies start to occur.”
— Peter Simpson [22:40]
[24:36–26:23]
G8 is the “next big advancement,” featuring multi-analyte sensing, improved self-adaptive algorithms, and a brand new, proprietary technology.
G8 will further reduce variability and may preempt many current user issues.
G8 adhesives benefit from everything learned with G7, with an even stronger, longer-lasting formulation already mid-rollout for G7.
“It has brand new technology in it that is proprietary, first of its kind, where we have additional signal that we can measure from the sensor to basically self adjust… Think of it being able to do [finger stick calibration] on its own.”
— Jake Leach [25:11]
Institutional experience driving culture:
“Peter and I actually used to go down and watch them ship the sensors out the door for that day to the users, right. And we're talking about hundreds of sensors, right? And now we're shipping millions and millions of sensors every week.”
— Jake Leach [03:34]
Commitment to actual user needs:
“The goal is to make sure that the feedback from that council shows up in the actual experiences our users are having, whether it's product or support.”
— Jake Leach [07:00]
Proactive issue resolution:
“When we see those failures, even if it's a small amount, we take them very seriously... these teams work really aggressively, we have to find the issues quickly and we have to respond to them quickly.”
— Peter Simpson [11:37]
Transparency about real-world variance:
“You put them on 10 different people and you can get 10 different experiences. And so part of the technology on our side is to develop that interface, that membranes and the coatings that work on all the different biologies that people have.”
— Peter Simpson [22:50]
Jake Leach and Peter Simpson offer a transparent and detail-rich look at how Dexcom approaches the real challenges of scaling medical technology for millions of users, with a clear focus on continuous improvement, user-centered culture, and transparent, science-driven communication. The episode serves not only to address current user concerns but also provides a hopeful, science-forward look at the future of CGM technology with G8.
Listeners interested in technical deep dives or process retrospectives are encouraged to reach out for future episodes exploring the "geeky" engineering details!