Josh Clemente (00:00):
All right. Just go ahead and do it. Welcome to Friday Forums, September 2022. Okay, recent achievement side. So we’ve got the OKRs project super well underway by now. A ton of work has gone into this across all functions and groups and from the top down in the company, and we’re now ready to operationalize this. We’ve got some more info on this coming in a minute, but just big highlight, big move in the direction of having more insight across the teams and how we’re stacking up resources and objectives. Content side, we had a new ultimate guide on vitamins and supplements that went live. This is a new addition to the ultimate guides that our members really know and love and are a major source of education and access through the organic search that is our main driver to awareness of Levels. So awesome work on this. And then Caitlin has started work on a new Metabolic Health 101 project.
(00:57):
The objective is to revisit how people get familiarized with not just the company, but also metabolic health, which is the fundamental mission that we’re working towards. And we have sort of two ends of the spectrum, sometimes with extremely detailed content, and extremely approachable content and building the ramp into that, that awareness is going to be something that we’ll have some new work on in the next few weeks. The research team is working with the content team on planning the various types of content that we can start to introduce into the product as well. So looking forward to that. We had some updates this week to the IRB protocol in the form of new procedures for deviations from the protocols when the standard protocol procedure is not followed by a participant, and then also deletion request.
(01:45):
So how we deal with that? And then research with Dr. Terry Wahls is coming together, and we have a CGM Fat Burn experiment, which I want to plug everyone taking some time to try that out. It’s internally live. We’re testing some theories on the association between CGM Data and Fat Burn, so definitely check that out and join us. Casey’s episode on Honestly has been just a standout success. We had a couple notes and threads this week, but essentially it’s had a ton of secondary effects, from outreach with huge podcasters and people like Jillian Michaels reaching out for a podcast episode, to a major spike in cash receipts this week, and a huge one in basically all the Google trends. You can see this vertical spike, vertical spike on Google Discover, which is how people, if you scroll in the Google app on iOS or Android, this is the trend line this week. So just a ton of organic interest in what we’re doing.
(02:41):
We had three new affiliates onboarded on the partnership side, Dulce, Elissa, and Beth. Tons of organic content, which you can see here, which is, we’re getting used to this at this point, but we should never take for granted how much the organic sharing is happening without us having to prompt. We had a new pod episode or a podcast ad tests with Cleaning Up the Mental Mess with Dr. Caroline and the Art Of Being Well with Dr. Will Cole. And then Peter Diamandis, we’re nearing activation on his rollout. So you can see the landing page here. He had a two-minute spot about Levels, which he put together super organic, super heartfelt, and a real testimonial. So getting ready to roll that out. On the product side, we had a new logging shortcut go live. So if you haven’t tried that, if you update the app, it’s a long press on the logging button, and that will boot up the camera directly, and then you release, you basically just release the button, and it snaps a photo and logs.
(03:37):
This is a really convenient shortcut. Definitely need to get some internal feedback on this. Up next is Scoring V2.2, and Levels Levels is in the concepting phase with product, which I’m super stoked for. On the inside we’ve got Labs V1.1, Weight and Height data in the app, and Healthier Food Choice features all coming together in eng now. The Prisma migration on the backend is very close. I think there’s six more repost to migrate. We’ve got NextJS v12 upgrades, UK beta launch, which we’re receiving feedback on in preparation for a round of iterations. And then just across the board on the eng team, metrics like pickup time, cycle time, review time, deploy, all just humming along this week. So really great stuff on the eng side. So you can see here a few other really exciting images or items we’ve got in touch with the Zumba team.
(04:27):
Learned a ton from them. We had a really awesome testimonial from a member who shared their story with CBS about how Levels transformed their perspective of food and the sort of health transformation that they experienced, including all signs of pre-diabetes disappearing, drop in blood pressure, stabilization of heart rates, and really amazing stuff. So great week. I’ll jump ahead from here. All right, I want to welcome Peter. So Peter is a Levels member, CrossFit gym, and yoga studio owner. He’s also a startup investor and entrepreneurial guy and he and I met a while back, and I’m super excited to have him joining us here on Friday Forum and taking some time here on this Friday to share some thoughts about his journey with Metabolic Health, his journey with fitness, and what you’re excited for in the industry. Thanks for joining us, Peter.
Peter Brasovan (05:15):
Yeah, super honored to be here, Josh.
Josh Clemente (05:17):
Yeah, we would love to hear just from your perspective how you’re seeing metabolic health and the relationship with just true whole-body health and fitness for you, and what you’re looking to see in the general population as we see a transition towards individual personal empowerment taking over health.
Peter Brasovan (05:38):
Yeah, my wife and I are actually hippies at heart. We have a huge garden in our front yard, and try to eat as local as possible. So when you talk about this type of stuff, it really hits close to home. I remember seeing Josh wearing a continuous glucose monitor in 2018 or 2019. I thought he was some crazy guy with a fake tattoo of some sort. And he’s come so far, and when this opportunity came to invest in Levels and see it grow to this point, it was just a really eyeopening experience that how many people still need guidance on their food and they think McDonald’s or fast food has no effect on their overall sleep or anything like that. So when I look at metabolic health and coming from the Crossfit world, it really came to the forefront for me right away back in the days when the paleo diet was insanely popular for a hot minute and just the results it was having on people’s sleep and their skin tone, and just things like that that you don’t even think about when you think about health.
(06:32):
People think about, oh, I need to lose weight or look stronger, which is the immediate thoughts on food, but really for me, it comes to mental health. And so back from 2010, 2009 when I really started on this journey, and then probably spiking in the pandemic when people were stuck at home and you saw a lot of fast food ordering and the things that were called essential at the time really opened my eyes just to how crazy this is. And when we talk about metabolic health and having something like Levels to really look at that, it made it a no-brainer for me to be an investor and be a part of this journey. It’s amazing to see what you guys are doing.
Josh Clemente (07:07):
I love all of that. Yeah, I think the mental health element in particular resonates because it’s so hidden. It’s behind the scenes, and most people just, we compartmentalize everything into physical and aesthetic, and then mental is a different thing, and you treat that entirely differently. And so, being able to, Casey really touched on this in her episode with Barry, but we have a number of different labels for the fundamental, foundational dysfunction that we can’t seem to recognize and address. You’re taking a front-row seat in this, driving fitness into the daily lives of people at your Crossfit gym and yoga studios. I’m curious, from your perspective, what you think that we need to see in the industry, whether it’s fitness tools, technology. What do we really need to see in order to bring about this shift in reversing the metabolic health crisis?
Peter Brasovan (08:01):
The first big thing I want to see is somehow normalizing that wearing something like Levels or just understanding of food is normal. We’ve always probably seen it. Maybe someone who doesn’t drink in your life, and not that I’m saying you should or shouldn’t be a non-drinker, but you go out somewhere and someone that doesn’t drink and doesn’t order something, you’re like, whoa, why aren’t you drinking? And it’s like they’re the oddball for not drinking when, realistically, we know in a sense like this, well, that shouldn’t be a play, we shouldn’t be worried about that.
(08:33):
So when I look at where we’re at and what should be talked about first is how do we normalize that understanding what we put in our body causes us to have these peaks and valleys and make people comfortable to either wear a device like Levels, which is really, I think, a great, easy solution. Let’s just see the data, but then just making what we should be eating and putting into our body shouldn’t be looked at as something that’s negative or positive. It just should be normal. This is who I am and it affects me every day in my life. And that to me, as I don’t know how to do that, and that’s where I look at you guys, but how do we have that conversation and make it easier?
Josh Clemente (09:10):
I absolutely love that. Yeah, can’t really add anything to it. I’m curious if there’s one thing in your experience with Levels thus far that you think we need to add or improve in order to continue moving that direction. Do you have any ideas there, any feedback that we could take from this meeting?
Peter Brasovan (09:27):
My lens comes from this group fitness industry. Our gym was pretty large. We had about 500 people. So a lot of times, when you’re introducing people to try to make nutrition a thing, we like to talk about the lifestyle of nutrition. It’s not a paleo diet, it’s not a fad diet, it’s the lifestyle. So figuring out to me how you introduce something like Levels into the lifestyle and making it a component of just what you’re doing and not that tracking is the end all be all because not, it’s just trying to get, again, use that word normalized. So I think that there’s an opportunity to introduce Levels, a lot of the gyms will do kickstart nutrition challenges or just straight-up nutrition challenges. So if you could make it a part of an introductory challenge to somebody to help them understand why and what this would do for them, I think there’s an opportunity there to introduce it in that level and these Levels in level to introduce it into gyms or people’s life in a nutrition challenge or in a way that’s short-term that can make it more long-term.
Josh Clemente (10:31):
I love it. I love the idea, and certainly, as we’re entering general availability and starting to explore new ways to build awareness in new communities, it’s certainly super relevant to us right now. Peter, thanks for joining us this morning. We’d love to have you stick around for the rest of the meeting if you’re interested. It’s awesome to reconnect, and I’m just really excited for your support and for you putting not only your financial but also your mental support behind what we’re doing and thinking through the challenges we face and sharing them with us today. So the team really gets a ton out of these interactions. Thanks a lot.
Peter Brasovan (11:05):
My pleasure. Thank you all.
Josh Clemente (11:07):
Awesome. All right, next, I want to welcome Evan. So Evan has joined us on the R&D team as our first intern, actually the first intern here at Levels ever. He’s straight out of University of Georgia, where he has a biological engineering degree, and Evan is hitting the ground running, helping us get the facility up and running and moving towards getting some really cool research underway. Evan, I’d love to hear a couple of words. I’m not going to read your slide entirely. Love to hear some thoughts.
Evan Adams (11:35):
Hey, everyone, it’s me. I’m Evan. I’m here at the Austin site today; actually, I’m in one of our many office rooms. Like Josh said, I’m directly out of the University of Georgia. This is my first job, and this has been my first week at Levels, and I’ve absolutely loved it. I’m learning new stuff every day. About a month ago actually, I was more or less completely ignorant over the topics of metabolic health, but every day that I spend here, I learn so much more and I’m so much more convinced, and I’ve actually made a lot of personal life choices here recently concerning my intake, and it’s really been a benefit overall. I can’t complain. And the company motive and values, I love all of it. I’ve loved this first week at Levels.
Josh Clemente (12:18):
Awesome. Yeah, Evan and his girlfriend Aaliyah have relocated to Austin from Georgia just in the past week. So uprooting everything, coming aboard, getting started real quick. Appreciate the work so far, Evan. Looking forward to many, many more projects and challenges up ahead. All right, team, please help Evan get ramped up. He’s getting started fresh with all the communications tools that we’re using and all the various platforms and systems of work. So best practices are always appreciated. Jumping ahead to culture and kudos. So yeah, just wanted to highlight once again all the effort that went into the preparation, the relationship building, which was actually multi-stage, essentially first starting with one of the producers on Barry’s show and then migrating to capturing the attention of Barry, and then getting onto the episode.
(13:08):
And obviously, the preparation work and the execution by Casey just a phenomenal job and it affected every element of the business and also will continue to have a ripple effect. So I just want to surface that that was a, I know we talked about it last week, but it’s hard to cover everything. So thanks to the whole team who supported on this. And then on the eng side and product side, so we’ve had a really great, we’re getting some great feedback, although the sample size is not complete, so the whole team did not respond to these surveys. The data on the velocity and the scoping and scope creep on the product design and eng side is looking really good. So it’s trending in a positive direction over the past few months.
(13:47):
And so I just want to highlight that and appreciate all the work that’s going on behind the scenes to improve systems, and lock in scope, and be able to execute on a consistent cadence, week over week. We had an onsite interview this week for the R&D team and had a couple of people in town from out of town, including Ben and Pam, oops, and Sam, and [inaudible 00:14:08], and we had a little dinner here with the Austin crew, so that was a lot of fun. And lastly, we’re getting some great feedback from people on the mobile app improvements, including spike detection. So I want to make sure we don’t lose that in the noise. This was a big transition on the app, and honestly, all things considered it went super, super smoothly and that’s a testament. So thanks, team. All right, Sam, over to you.
Sam Corcos (14:34):
All right, the culture slide for this week is focusing on, we were a team, not a family. And this is one that can be kind of controversial for people because it can feel to some people like we are mercenaries, but that’s really not the intent. The intent is to recognize that at some companies where they use the language of a family, they can really create pathological expectations and incentives for an organization where you feel like you can’t leave a toxic environment because you can’t leave your family, but you should be able to leave a company or a team if it’s not suiting you.
(15:17):
And so we really want to use the analogy that they use in Netflix culture around sports teams, which is, we are here to win a championship. I think if you intend to build a high-performing culture that focuses on performance, you have to have this mentality, and we have to always continue to seek to improve. So the way that we approach this is we have to recognize that we are all here focused on the same goals. I think the OKRs project is going to help a lot as we talk about that later, making sure we’re all aligned towards the same goals and hold each other accountable to our objectives.
Josh Clemente (16:07):
Yeah, love it. I mean, this has been integrated into the content. We’re producing the handbook, and we’re going to continue to surface these sort of culture pearls here in this format. If you have any questions or any recommendations, definitely raise those, but I think this is a major component of the Levels culture. We’re here to accomplish something pretty staggering in scope, and so we need to have that focus and lead execution. All right, over to Hui Lu.
Hui Lu (16:38):
Yeah. Hello, everyone. Yeah, just would like to give a quick announcement about this new memo. What is DRI database? It is aiming to provide clarity around what and how we use DRI database for to help us better keep track of responsibilities. DRI all boils down to accountability. And in order to be accountable for something, you also need to have decision-making authority over that thing to move things forward. It also answers some commonly asked questions, such as what should go into the database, what is the process to hold this accountability model?
(17:16):
What if a conflict arises, and et cetera. The ask here is, yeah, please take some time and review this memo, digest and intake what it means for you to be DRI or accountability person. There’s AMA, ask me anything, sessions scheduled on the Monday, next one. If you have questions, come to ask me questions and be actively participating in upcoming auditing process, which is try to sanitize and update all DRI items to be currently in the database to be up-to-date. And after that, we’ll also kick off recurring auditing process to make sure things are always up-to-date and kept track of. Yeah, that’s all. Thanks.
Josh Clemente (18:07):
Awesome, thank you, Hui. Yeah, great memo. Really helpful to continue to evolve the DRI concept as well as improve the functionalization of the database. So thank you very much for that. Definitely, if you haven’t seen this memo yet, it’s in general. Please turn your notifications on for the general thread so that you can see these types of important updates. Okay, company objectives. The main thing hasn’t changed. Levels shows you how food affects your health. We’re all working towards this in some capacity. If you feel you aren’t, please raise that concern. And then this is how those top-level objectives stack up. As well as we now have the measure of success here in an additional column. So main focuses are member retention, new member acquisition, and member health improvement. Actually, I’m going to hand this over to Lauren.
Sam Corcos (18:54):
I’ll take the first slide, and then I’ll hand it off to Lauren. Go to the next one. Cool. So as many of you are aware, we’ve been since about late July, we’ve kicked off a new set of OKRs, and as is always the case with these sorts of things, the first time you do it, it takes a lot longer and there’s a lot of de-conflicting, and recognizing, and gaining alignment around what our goals are. And so we have the OKRs ready to go. So, Lauren has really taken lead on this. So a huge thanks to Lauren for that. Sonja’s been in support, especially around some of the procedural things and building systems to make sure we have ways of reinforcing it.
(19:42):
Everything flows into these top-level objectives. So if you remember how we have the functional groups, those groups have objectives, and each of those objectives is tied to a top level company objective, and that’s how everything ties in together and works. So these are new, but in many ways, a lot of them are projects that we’ve had ongoing. Sometimes we have had to readjust, but this is an effort to make sure that everyone is rowing in the same direction and that we have clarity on what we’re focused on and where we’re taking things. I think this is probably, as I’ve talked about in the past, we tried to do OKRs some time ago, and they weren’t as informative because the team was so small. I think this time it is going to be quite valuable in gaining that alignment. So I will hand it off for the next slide to Lauren, and she can take it from here on some of the specifics.
Lauren Kelley-Chew (20:48):
Hey, everybody, super, super excited that we are officially kicking this off. This is really just a preview because we’ll be doing a fireside next week as well. So looking forward to all of the lively conversation. And like Sam said, many thanks to Sonja and everyone who helped, especially the functional leads, who really did a lot of work and thoughtful reflection in defining these. So from my perspective, what success means for this project is that every team at Levels and every person knows exactly how the work that they’re doing contributes to driving the company towards our company objectives. And so what you see here is really just objectives and key results, and you can think of objectives as the goal that the company has; as we know, they’re presented each week, the company objectives, but that also each function has their own objectives and that those tie directly to the company objectives.
(21:41):
So these are the goals, and these really reflect the critical problems that we need to solve as a company in order to get to our next stage of funding, our next stage of success, our next stage of growth. And the key results are just the primarily quantitative metrics that will help us measure how close to those objectives we are and what our progress is. So in terms of what this means for everyone on the team, part of the purpose of this exercise was to begin to make sure that we’re all on the same mindset in terms of problem solving and thinking in terms of solutions rather than thinking just in terms of features to build or projects to complete, or roadmaps to follow. And so setting the stage for being very outcomes-focused rather than output-focused. So hopefully what you’ll see as this kind of framework gets implemented is that there’s much more emphasis on is the work that we’re doing solving our problems versus are we completing this project that we set out to do simply because we thought that it needed to be completed. Next slide.
(22:45):
Like I said, this is just a preview. So prior to the lengthier conversation next week, there’s a few asks of everyone on this call and from the team in general. The first is to take a look at the memo on the fourth quarter functional OKRs, and that will be shared with everybody. And what you’ll see is each function and each functional group defining exactly what their focus will be and how they’ll measure their progress. There’s also a memo that you’ll have access to which explains how we’ll be tracking our progress, how this will be reported, and what you can do to follow along and to always just keep track of whether the work that you’re doing is really driving towards your functions objectives and therefore towards the company objectives. And that’s the final ask of this team is to start really thinking about everything that you’re spending your time on, all of your priorities, and your focuses for you personally.
(23:40):
Are those directly linked to what your function has outlined as being the most critical and important tasks, or I shouldn’t call them tasks, problems to solve. And can you trace directly how your work links to driving the company forward towards our company objectives? Like I said, coming up next week is another Friday Forum update, and you can look forward to those every week. And each week we’ll be featuring one of the functions who will do a deeper dive into their OKRs and how they’re thinking about that. And that will be led by each of the functional group leads. And then you’ll also have access to an OKR dashboard where you’ll see a simple stoplight chart to track progress and to just get a sense of how we’re all doing and how we’re moving towards success collectively. So to be continued at the fireside next week, and look forward to lots of questions and conversation. Thanks, everybody.
Josh Clemente (24:36):
Awesome. Thank you, Lauren. Thanks, Sonja. Thanks, Sam. Really looking forward to the rollout here. All right. Over to JM.
Josh Mohrer (24:46):
Okay, would you mind stopping the share and spotlighting me? I have to be difficult. All right, and on my screen, it’s all listed as backwards, but I assume if I can just get a thumbs up that it’s showing not backwards. Okay. Demand capture. We grow the business by converting our audience into members. I’m going to start with Ben’s last slide from last week. He and I are going to be on an alternating schedule where he’ll present one week and I the next. And I quite enjoyed this last slide, so I thought I’d start with it. It’s about eyeballs, emails, and orders, which are the three things we like to measure here on the growth team, but I’m going to talk about how I think about the world. Even though we run a website and online business, I think of the metaphor of the supermarket and people come in to our store, and we’d like to know how many people there are, how they got there, how they heard about us, are they warmly greeted by this gentleman up here.
(25:52):
Some people will come in just to read the articles. That’s okay. And we’ll gently make the case that they can join Levels when the time is right for them. And if one day we showed up to work and the store was more crowded than usual, we would want to know why. And we want to make sure, when people come into our store that they can find what they’re looking for, understand what we sell so that they’re more likely to buy. And we want browsing to be as convenient as it can be so that when someone is ready, they know where to look. And on our team, sometimes we have to be detectives.
(26:35):
We make sense of the world through attribution. And what I mean by that is we can see where someone comes from when they come into our store or into our website. If they use a code, then we’ll explicitly know if they find us through another website, we’ll get that information in our analytics, but it’s not always clear what happened, and sometimes crazy things happen without any real attribution. So we put on our detective hat and figure out what that was, and this was a very exciting week for detective work. I learned at forum last week after seeing sales numbers really be higher than usual for many days. Last week I learned at forum that Barry Weiss put out her episode with Dr. Casey, and we saw a massive surge in searches for Levels and for Casey and enjoyed a nice bump in sales from that.
(27:27):
But it turned out that wasn’t the only thing, Paul told me later in the week that Google Discover, which I never heard of, it’s the native Google experience. If you get the Google app, not one of the specific ones like photos or whatever, but just the vanilla generic Google app, they randomly show a bunch of stuff there, and they started showing our vitamin supplements and ice cream articles last week, and that’s driving a couple of thousand visitors a day, which is a pretty big deal, and it was a total mystery, and we could have very easily missed that. Generally speaking, SEO, or search engine optimization, is really the result. It’s sort of an industry that’s just the result of people guessing what Google is looking for. Since none of the information about how the algorithm works is public. So we’re in the process of doing a bunch of experimentation with different things to rank higher in Google, and we already rank very, very well, but it’s far from a straightforward thing, and it takes detective work. Another place where we’re detectives is in our international expansion.
(28:41):
As we launch our beta in the UK, we’re going to learn the subtle differences between our two countries, the US and the UK. These two guys look the same, but they’re very different in almost every way. And so we’re going to learn things, the subtle differences like the unit of measure for CGMs, millimolar versus milligrams, but also more complicated things like language and culture, excuse me, and that we’re more the same than different. International expansion is about figuring out those things and embedding those learnings in our product, and growth, and support work. On the partnership side, Jackie and Tom noticed that after liftoff, our partnerships really served a different use. The offer that we had been making, and all of them skip the wait list because that was no longer a thing. There was risk of underperformance there, and so they came up with a couple of offer ideas, how do we make an offer something that makes the ad unique and makes someone interested in following that ad to its endpoint?
(29:50):
We’re about to launch a couple of experiments with offers. Mark Hyman has an exclusive video that he is going to share automatically through our system for anyone that buys. We have another one coming with extra months of membership, and we’re excited to see how that goes. Also, working with DemandGen, Haney, and Ben specifically, we put out a new experiment. An alternate way we’re going to get email addresses on the blog. You may be aware that a popup shows up now on your first visit on Ultimate Guide for Glucose. We created a PDF version of that, that you can be asked if you want it, if you want to save it for later. This is a post that gets something between 500 and 1,000 visits a day, and it’s very long and very interesting, and so thus a stylized hard version of that can be in your inbox for free.
(30:48):
I’d be remiss to present without at least showing some graphs. You know my affinity for graphs. Here is new members by month. You can see we’re coming up towards meeting last month. Our revised goal for this month is 2,545 new members, and we’re trailing that by about 9% now. It was much higher than that last week, and so I have a good feeling that we’re going to come in pretty close and maybe even exceed that with hopefully another strong week. On a weekly basis, last week we had 616 new members, which was the second highest since liftoff and really the highest ever when we’re not hitting the wait list with invitations. This was driven by, of course, all the things I referred to earlier and just zooming out a bit, a graph that makes me happy. Weekly total shipments over the last year, up, up, up, up, up, up, up. And this one blue is subscribers. Yellow are first-timers. Green, I’m sorry, Chris Jones, are replacements, and red and purple are blood and singles, respectively. Thank you for coming to the Live JM show, and have a wonderful weekend.
Josh Clemente (32:00):
Thank you, JM. Insightful as always, engaging as always, and some great metaphors in there. Continuing to love learning about the growth story. Over to product.
Maziar Brumand (32:11):
All right, welcome to the product update September 23rd. If you remember from last episode of Friday Forum, we talked about the metaphor of velvet cake. That product is like a cake, a beautiful cake that you see here, which is made of layers that really drive what we’re trying to drive for our members, which are effectively engagement, retention, and then delighting them. If you go to the next slide, please. So our bakers are hard at work baking this cake that we hope people will love and it’s delicious, and people want to purchase this cake, but then, when we thought about it more, what can be even better than a delicious cake? Next slide, please.
(33:03):
A delicious cake that rates a 10, a perfect 10 on Levels, which effectively helps you maintain your good health and makes you want to come back because it tastes good and it rates really high on the metabolic scale. So we’re hard at work making a beautiful, tasty cake that rates really well for your metabolic health that you want to come back to over and over again. Next slide, please. All right, so what is this cake made of? What are the layers of this cake, and what are the ingredients? So, as you all know, we’re investing heavily on content. Content is the lifeblood of the app, which drives the insights. And we have three channels where we’re communicating this content, we plan to communicate this content, which is personality-driven content that the team has been hard work experimenting with. And there’s a really good indication that this will be some of our most popular content that people will engage with.
(34:06):
Whether it was the Mallory video that we tested or the video that we’ve done on social that now have shown to be one of the most engaging videos we’ve created. Event-based insights, we continue to grow those, and the pipeline looks really strong and really proud of the work that the team’s put together. Big thank you to Mike D, and people that have been really helping him drive that. And then healthier food choices, which is in production, and hopefully we can ship it soon to our members and see how that works out. The second one, our superpower, really is biofeedback. And currently, obviously, we have the CGM, which helps people see on a minute-to-minute basis how their glucose is responding at a molecular level to the food that they eat. And we’re turning that into a score, and I’m going to talk more about that in a minute.
(34:55):
Labs, which we’re investing in heavily, that gives you more of a picture of the rest of your health. Glucose is not everything. There are a lot more things that we need to track to make sure that we get that feedback from our body. And there’s some more exciting things that are in the hopper that we’re thinking through for feedback, and I’ll keep the suspense for that for now. All right, engagement. So how do we make all this interesting and fun, and personalized? So this is all the other instructural tissue that we’re creating to boost the ability, to boost the motivation, and to boost the rapport to make people want to eat that cake. And so really, with this, which is our superpower of content, biofeedback, and the magic of the experience with our expert bakers, we hope to make this cake that people want to eat and come back over and over and over again for many years to come. Next slide, please.
(35:55):
All right, so what are we working on? Levels Levels. I’m really excited about this, and Alan’s going to talk us through a sneak peek of what this looks like. I hope you really like it. And Labs 1.1 and 2.0. 1.1 is now in production, handed off to engineering, big thank you to Cozimont and team to really making this, and Viktor, to making this a possibility. And then we’re also finalizing the structure of 2.0 and I hope that this will really make a difference in how people engage with a lab product.
(36:26):
In app content, we continue to push forward there, as I mentioned, with Stacy, Mike D, David, and the rest of team really supporting, and really big thank you to the rest of the company getting involved here, whether it’s Casey, Sonja, Haney, and the rest of the team. So really proud of the cross-functional work that’s happening here. Personalization, we ship demographics, so now that’s live in the app as part of the experience to really help us create the personalized experience. And then, scoring, we’re going to give you a little bit of update of the new thinking around adding an additional layer that will help people improve throughout the day and also track their progress over time. With that, I’ll turn it over to David. Thank you.
David Flinner (37:04):
Thanks, Maz. Let’s see, is the next slide up yet? There you go. Cool. So last week I talked a bit about how we had shipped the table stakes version of logging 5.0, and then we had a bit of a faster way to get to that logging screen with 5.1. And this week, I’m excited that we’re launching the logging shortcut. The logging shortcut is if you hold on that action button for about a second, it’ll open up the camera automatically, and when you release your finger, it’ll take a picture, just a really fast way to take a picture-based log. Whereas if you just tap the button, it’s a really quick way to open up a text-based log, and please try it out. It’s live today. The demo here is pretty perfectly a spec that we showed a couple of weeks ago. Great job on that one, John, and great job on the rest of these features as well.
(37:55):
There’s more we’re going to do on logging, but this is, we’re in a pretty good spot right now on the table stakes rails for this. So next slide. And then wanted to quickly recap on how we’re responding to member feedback on scoring from that stability ring and spike detection launch. So with version 2.1, there’s been a few bug fixes in that. But one of the things, the last thing in this project that we’re wrapping up is providing relief for members who are getting all-day spikes. So one of the behind-the-scenes details for this is that if you tend to have a hot sensor or your average is running really high, people who are averaging over 110 can see all-day spikes. And so we’re doing some small backend change tweaks on the engineering side to bring relief to these people so that they’re not getting an all-day spike.
(38:37):
And it’s a bit of a technical detail, but we’re creating less of an artificial baseline for those members who are at the higher range. And then version 2.2 is providing more movement for the stable time metrics. So one of the feedbacks that we heard internally and externally as well is that the stability ring is super encouraging and it gives people assurance that they’re doing well throughout the day, but they wanted to see a bit more movement in that. And so right now, with our first launch, these unstable time, that red zone is calculated from the moment you go 30 over your baseline to the point that you return to 30 above your baseline, and it’s a very short window to represent the spike. So what we’re doing is going to be expanding that window to more intuitively represent the spike duration. So it’ll be that initial spike start, which is that plus 30 over your baseline, but then we’ll keep the red line going until you recover and get back to your baseline.
(39:33):
So Jason did some really good work on the data analysis side to prove that this would be a good change. And you can see more in threads on the background for this. But looking at the data, it really did make a lot more intuitive sense from a product standpoint to map onto what people expect. And so this should be a pretty positive change for people, but it’s going to take a few weeks for this one to be under development, most likely. So it’s in end review right now to see how long that’s going to take. And then lastly, the last thing we heard was that people wanted to see a bit more visual quick glance ability in terms of how they’re doing day over day. If you go into my data now, you see your spike counts and spike time, but it’s a little bit hard visually to see what’s the difference between a three-spike day and a one-spike day.
(40:16):
And so we’re looking at a scoring abstraction that more visually helps you understand the differences. This is not going to be something that we’re raising to the forefront to replace the stability ring. This is just something that extends the concept of stability but is more of a gradient and helps you see day over day what that difference is. So taking the concept of stability, mapping it onto a simple metric, which is most likely going to be variability, which lends itself to a nice color gradient, and that when you’re in my data, you’ll be able to see how you’re trending day over day. This is going to be a very scrappy test internally. So Galit is working on this now. We want to get the next step is just validating internally that people actually find this useful. So stay tuned next week, where we will turn this on, and we’d love your feedback.
(40:57):
If it looks like it has legs, we’ll continue to design it and release to members as the last iteration, on their feedback from the scoring v2.0 Project. And more to come. Next slide. And okay, next slide. So the main update here is Levels Levels, and we wanted to give you a sneak peek on the concept of this as Josh mentioned. Next slide. So just to set the stage Levels Levels, the opportunity here is Levels is shifting towards member health improvement. We don’t just want to show you where you’re at, although everything we’ve done to date has been table stakes. So that great logging experience, the scores that we were talking about, that zero to 10 zone score that JM was playing around with, but we really want to help people get to improvement, and that is a combination of showing you what the goal is, showing you directions on how to get there, and then also making it really easy and fun on how to get there. So next slide.
(41:51):
And we’ve laid the foundation with the core stuff loggings there. We have that right now, we closed the loop in real time. We made it very understandable and explainable to show you the opportunities with spike detection as to which parts of your day are most opportune to focus on, and we’re shifting towards that improvement. So healthier food choices is really those Google Maps directions pointing you in the direction as a v1.0. You can imagine over time that’ll extend out into programs like pulling on personalities programs, perhaps working with someone like Bridge-IT to bring on step one, two, three, four for that. And then, as well as your labs. Really showing you what the outcome is and showing the proof over time. But next slide. But we don’t just want to show you the directions; we want to make it as smooth as Google Maps navigation, as just turn it on and you go there.
(42:39):
And so what is the equivalent for that for health improvement and for our product? And I think one of the ways for doing this is really making this fun, and making it engaging, and helping you get to your goal in a way that is going to motivate you to do it without it feeling like a lot of work. So next slide. How do we take all these pieces that we’ve laid in the system and glue it all together, and what is that glue that’s going to help orient people towards where the goal is, where they’re at, how they can get there and have a good time along the way. And we think that game system dynamics can layer on top and really help with this. So I hand it off to Alan to talk to you a bit more on the ideas here.
Alan McLean (43:17):
Cool, thanks, David. So yeah, I’m really excited to actually be at this point where we can start exploring this more deeply, and David and I have had a lot of fun looking at the history of games, playing some of them. So I think, if you go to the next slide, games have a really rich history of teaching people something new. Probably none of us remember this, actually, I wasn’t old enough to use this, but the original MAC, even when you first learned how to use a mouse, which was kind of a novel new and concept for a lot of people, you literally drag this little mouse to the cheese. Quite a delightful little experience there. Next slide, please. And of course, the best games, they stack on top of each other, they take these learnings and they progress you through the system, and you gain mastery over it as you’ve progress.
(44:04):
And so you start maybe in Mario with this little jump, and then you start hitting things and you get rewards and then eventually you learn how to jump and maybe potentially kill your enemies, right, in the third section there. So these are sort of learning loops, and we want to start introducing this into Levels too. Next slide, please. So the objective of our game is to better understand how food affects your health, and it’s not really a destination; this is sort of the thing you’re always doing, right? This game is going to be about being engaged in your wellbeing. You’re going to be learning how to take care of yourself or do better maintaining your metabolic health. And sometimes you’re going to go back, and sometimes you’re going to go forward, and that’s okay. This is just all part of savoring the journey.
(44:45):
Next slide, please. And so to move forward in this game, progress will be earned through what we’re calling experience points at this common game nomenclature, but we’re probably going to sort of explore and see what we might potentially call metabolic points. I don’t know. Anyway, so basically, these points will be rewarded sort of commensurate with how much effort you’re putting in. So really large tasks like challenges or experiments, making a recipe that we suggest that would be more of a larger reward, right? Medium-sized ones. Perhaps it’s consistency in managing your glucose variability, and maybe the smaller ones are more everyday tasks, right? Maybe you log something that’s good for you. Next slide. So basically, this game revolves around what we’re calling the metabolic checklist. We know the power of checklists, giving you something to do, giving you some purpose. And really, the goal here is to essentially win the day.
(45:43):
And to do that, you’re going to need to keep your glucose stable. You’ll eat some healthy food, and you’ll do one of these healthy habits, which is sort of a catchall for all these great healthy things you can do to maintain or reduce your glucose variability and maintain your health. So you can see there on the right, we’re giving you some points for maybe a stable morning, stable afternoon. For eating glucose-friendly foods, it’s maybe logged three times. That’s part of connecting to that, or maybe you had some chia seed pudding. We want to find a way to actually reward you as much as possible without it being annoying. So it’s not just having stability for the whole day. We’re finding all these little moments that we can build upon to make it fun. And I think part of this also will be there’s a lot of randomization.
(46:26):
You won’t always know what you’re going to be rewarded for and we can continue to stack on this. I’m really looking forward to brainstorming with all of you, and Casey, and so on to find all the interesting things that we can build upon here. Next slide. So every ingredient we hope will account in some way. If you’re eating a bunch of anti-inflammatory foods, you should know about it. You might not even know it. We give you some points. You’re having kale or chia seed pudding, right? We want to make sure that we call that out and give you some acknowledgement. We think this might potentially reframe some of the logging as well. So it’s a little bit less than about what did you have and more like what was in your food, right? Next slide. Thank you. And we’re going to find a way to bring back challenges.
(47:09):
Challenges were a much-loved feature early on. We had a hard time figuring out where to put them. This is more of a structured task. This is where you follow specific protocol that we provide for you. And this is going to give you a much more clear picture of how to close the loop on your choices. And of course, because it’s so much more work, you’re going to get a lot more points for doing them. And we’re going to potentially reveal new challenges as you make your way forward along this journey. So maybe at the start you come with one of the most basic ones, maybe it’s like the coat challenge or something like that. And as you continue to go on, you’ll start to get a bit more involved, a little bit more interesting. Next slide, please. And of course, this list, as I said before, could get really long.
(47:48):
There’s all kinds of things we could call out. We need to find a way to manage this, especially if some of you’re logging every single ingredient in your food. But I think this is actually kind of a good thing. This is a good problem to have where we’re saying too many good things about your day. So we’re going to talk about this, we’re going to brainstorm how to solve this, but I think this is really fun. We talked also about potentially having things like multipliers, right? You get a stable morning, a stable afternoon, a stable evening, that’s a no spikes all day day and we give you a bit of a multiplier for that or potentially having streaks and things like that. Streaks will play a big role in the experience moving forward because we know that there’s going to be users that aren’t logging everything and they’re just observing and following their healthy routines to stay stable.
(48:29):
I didn’t really call that out in this presentation, but it needs to be a balance of both the active rewards and the more passive ones for those who are sort of in maintenance mode. Next slide, please. So, giving you a little bit of a look at this, these are all wire frames, so don’t get too hung up on any of this, but we’re probably going to bring this pretty forward into the dashboard. So it’ll give you a bit, have this little affordance near the ring that will show you how you’re doing towards building progress on that level. We don’t know how many levels you’re going to have yet.
(48:59):
We’re going to explore the math behind this a little bit further along. But as you build this, it’ll start coming forward potentially you level up; there’s the big ring showing you leveling up. And the next slide, please. And so again, these are wire frames, but we’re going to try to take advantage of this big, rich spot at the top. We’re going to be showing when you log healthy food, maybe you have a stable morning as you progress up, and so on. So I think that’s it for this update. It’s all still pretty rough. The game system is developing and hopefully show you more soon.
David Flinner (49:32):
One quick follow-up on this one. I just wanted to say that a lot of this stuff, there’s super deep rails and what we really focused on was the skateboard here. So it’s just the core basics, and do we think that passes the sniff test? And this is just the concept thing, what you’re seeing here, we haven’t even done any of the shaping. So this is just very high level. What we really need to do is dive deep into the mechanics on what deserves a streak, what deserves a multiplier. And like Rob was mentioning in the comments, we need to be very thoughtful about what are we trying to incentivize. And this isn’t about driving everyone to a flat glucose line. This is about rewarding the inputs that are healthy choices. And so you can imagine that the multipliers might exclusively be for chaining days together of logging healthy micronutrients or something like that.
(50:14):
We need to figure out the things that we’re aligned on that we do want to encourage and the things that are sort of off limits. But these are all very deep tracks that are going to require a lot more thinking. But the good thing here is that we can do the smoke tests here on the skateboard fashion, see if any of this resonates. We can quickly double down on what works, figure out what doesn’t. But we’re going to need the whole team to engage and help us with ideas on what are the paths that we want to go here. Really working very deeply with both the data teams to understand what we’re seeing on our member data, the research and medical teams to understand what is the science and what do we want to incentivize. And then everyone else, and the rest of the team, we all are using the product. We talk with customers, what are going to be the things that will resonate.
Alan McLean (50:58):
I just want to add one thing before we move on too. I’ve seen some of the comments there. So like any real-world life game, there’s going to need to be defined guardrails for this. So we’re definitely going to try to find a way to ensure that both the game doesn’t lose sight of the actual health that you have and sort of the things that are adverse to that. But we also need to be really mindful of these things. So I think you’re going to see us unroll anything like we might think of as engagement juice. We want to make sure that it’s squarely focused on you. Missing from this presentation was competition. There won’t be an element of competition in this. This is about you, this is about your health. Less so competing with other members of the team. So just want you to know we’ll call that out. And I think we can also have a more in-depth presentation later too.
Josh Clemente (51:48):
Yeah, I love it. And I think one of the beautiful things about all this is making the rules of the game or the goals of the game very transparent to people so that they can read. Just like when you’re starting a new game out of a box, you read the instructions and the rules and you decide whether it’s a game you want to play. And so I think this is part of that is explaining the underlying incentives and what we’re seeking to achieve, and always making that available to people, and receiving feedback on it. So really love the discussion here today and super excited to see Levels Levels coming together. This is the long-term health improving objective.
David Flinner (52:22):
Yeah. Final thought on this one, just to reiterate, it is a discussion. So I think a lot of great progress here, but we’re going to want your feedback, and we’re going to be iterating on this in the weeks to come as we start to shape this into real designs. Engineering is not going to start in this for several weeks, so.
Josh Clemente (52:35):
Love it. Well, I’m stoked to see the direction and some experimentation starting. All right, we’ve got quite a bit left here. So we’re going to just turn towards the last few minutes here. So on the experimentation side, I’m going to jump out of the way on Athena in Helpscout Pilot.
Chris Jones (52:58):
Thanks, Josh. So we recently are wrapping up a pilot where we allowed Athena to help us out in Helpscout. So as most of you know, our current model is our entire support team is full-time people, which means by the time that we open up a rec, interview, hire, onboard, it can be five months from that posting of rec towards we have someone in the queue really helping, which means we have to nail our forecast in terms of volumes and orders and support to really do this well. And as we were going through, we realized the volume can go up or down. It can be a podcast, it can be something breaking, and we need a much more flexible model to have more help on demand. So we actually said, well, a great place for us to begin is using Athena and treating them like contractors like we would of: how do we make this as short as possible?
(53:50):
So the goals about this is how do we bring someone on board that is not a full-time employee, limited access and basically ramp them up quickly to help out when we have spikes in data. So that’s really the goal we are going for. Next slide, please. So as we’ve all gone through onboarding, there’s a lot of getting familiar with all these departments, engineering, content, go see what everyone else is. Take city training. This is an area where if you’re just in of how much is Levels or what’s a membership and you’re relying on FAQs or snippets, you don’t need to spend time understanding what’s going on in growth, or design, or engineering. So how do we streamline it, and specifically, how do we get rid of the city programming training, which I know everyone probably takes every week because it’s so much fun, but it’s a big time sink?
(54:34):
So we really said, well, what can we do? And it’s really pre-sales and order management. This is the biggest area for focus for this group. It’s not can you solve everything, but just more, Hey, can you take things out of the queue and provide more coverage? Next slide, Josh. So when we look at just these two categories, another area was, well, right now, all of our support members, we’ll go through and answer every question. How do we create a sense of triage? Can we flag things so that if someone’s coming in and their scope is only pre-sales or support, can they actually get ideas of like, oh, this has already been pre-categorized in areas that I can do versus having to read through 10 cases to get to one? So within Helpscout, we created a number of workflows that are looking for basic keyword.
(55:16):
Looking if they mentioned price, if they mentioned cost, if they mentioned where’s the status of my order, it automatically flags it with an order management or pre-sales flag. And those two categories. So if you look at pre-sales, it is 15% of our volume today. And when I actually looked at the accuracy of this AI model, it was 30% accurate, so not great around pre-sales. On the order management side, that is 30% of our volume, and we are 50% accurate in terms of identifying order management. So a little bit, we can tune this, it did a pretty good job on the order management, but this is almost half of our volume on things of what is Levels, how much did it cost, and where’s my order? Next slide, please. Another area was how do we give people that are not full-time employees or even Athena access to retool and limit their access?
(56:10):
So we created a folder that only Athena has access to. We took some of our key dashboards, like the member show, and made a limited version where we removed PII, we’ve removed actions, I think it adds see glucose. So it’s trying to strip down and really limit the data they can see in retool to say, how do we give this as opposed to you have to be a full-time employee to get access to retool? Next slide. Sorry, I’m going through this quickly, but it’s the 1.5x speed version. A little bit of the quick results, we’re going to be doing a deeper dive on this. In terms of the quicker onboarding, super successful. Roland and Nina, I think within three days, were answering tickets in the queue. They helped out on a lot of different areas, so it was very quick to get them up and running.
(56:53):
The AI for triage was a bit of a mixed bag. Not great on pre-sales, but really good on order management. The retool access, even though we didn’t leverage this a ton, I think it created a framework that we could use this for other areas around delegating to Athena or contractors to give them a lockdown version. On performance, over the course of a couple of weeks, they took 167 cases with the 96% happiness, which is incredible. And they specifically helped us on the ask the audience, which drove 150 cases to us during the time we were busy.
(57:23):
And we just handed all that to Athena and said, take all of these which are more cookie cutter and just get them off our plate. And they were super helpful on that one, Last slide. Next steps. We’re doing feedback now from all the team; what went well? We’re going to create a roadmap of how would we onboard a contractor to do this model and lay it out, and then eventually start looking at using the contractor model to scale support expansion, and really in terms of augmentation. Sorry for the lightning-fast version, but thanks.
Josh Clemente (57:55):
Perfectly executed. Thank you, Chris. And I’m very, very interested in the continued experimentation here, and love the direction it’s taking. Thanks, team, for continuing to explore ways to scale ourselves. All right, quick hiring update. So we’ve got Priya and Shawn both starting. Shawn will be next week, and Priya, October 17th; software engineering and support associate, respectively. Excited for their joining. And then we’ve got our roles still up. If these are up, it means they’re live. Please reach out, send your information, or forward on to somebody who would be a great fit for Levels. Okay, we made it. We’re right at the buzzer. Thanks, everybody. Awesome week. Obviously, tons of great discussion, tons of great info and data, and have a great weekend. Enjoy yourselves. We’ll see you next week.