February 3, 2023

Friday Forum is an All Hands meeting for the Levels team, where they discuss their progress and traction each week.


Josh Clemente (00:00):

All right. Let’s go ahead and do it. Welcome to the first Friday Forum of February 2023. Time is already moving faster than I’m comfortable with. That’s a trend with me. Quick reminder about what the Friday Forum is all about. This is our weekly synchronous time. The goal here is to look across the various functions inside the company, celebrate achievements, hear from members and partners directly, talk about culture and updates to how we do business here at Levels, and generally just align. It’s not a deep dive into business analytics, the underpinnings of why we make decisions. All that stuff is done asynchronously in long form and elsewhere. So just a recap on that.


Okay, recent achievements this week. So firstly, I want to highlight something that we’ll dig into just a little deeper later, and there’s also some stuff in comms that you can dig into here. But we’ve merged the demand functions, demand capture and demand generation, back into a single function called Growth, which Tom is going to take the lead on. So the history of how growth became demand gen and demand capture and back into growth is all well captured in that retro. Ben’s going to give us a little bit of a dive into that in just a few minutes, but that’s a big change and one that I think is going to improve our effectiveness.


Major focus right now is on the interface between product and growth. So right now, our goal is to validate market strategy. We have upcoming product changes, pricing changes, marketing strategy, all of which we need to validate, and we need very close alignment between the product and growth functions. So that’s the key focus between those two teams right now.


We’ve recently shipped an update to the UK signup process. So we’ve brought that in to align with the visuals of the recent US signup changes. Current focus is on building that Crystal Clear Pricing. We’re calling this signup 3.2, and the How it Works sections, which will really explain… It’s more of a product marketing effort. So a lot of great work being done right there. And then we’re also focusing on performance marketing campaigns. So continuing to get very interesting detail on how our various efforts on user-generated content versus retargeting versus sort of de novo capturing attention for the first time through these performance marketing ads is going.


On the eng side, lots of work going into Labs 2.0. This is just about ready to launch. A lot of backend work to get the various partnerships and logistics’ interfaces in place in the labs. The eng work and the app work is getting close to ship ready. And also Trends 1.0, which you can see some visuals up here. The Trends 1.0 process is also nearing ship ready. So we should have some great new features in an app soon, and I’m very excited for Labs 2.0 to release. We’re also putting a lot of eng resources into the growth infrastructure, and this is paying dividends. So all this stuff that I highlighted on the market validation strategy, all of that is powered by analytics that the growth team is really benefiting from some awesome eng work on.


On the member support front, we’re continuing to experiment with leverage. The Otonomee pilot is going really well, I think much better than expected. And we’re also looking to leverage various new and exciting tools, like ChatGPT. So I think we’re going to have a bit of a dive from Chris today. We shipped, on the editorial side, The Month in Metabolic Health from January and how to talk to your doctor about blood sugar, which I think is a really excellent article and one that is quite intuitive in retrospect, but we hadn’t written until now. And I think it’s such a powerful guide for our members to know how to approach these subjects with their primary care physician. And then lastly, we’re writing a response to the pediatric obesity guidelines, which recently came out. And frankly, there’s a lot of misalignment, of course, between the strategy that is laid out in those guidelines versus what we obviously understand to be the underpinnings of the childhood or pediatric obesity epidemic. So can’t wait for that to come out.


A big one, and Dom is on this call, but just want to shout out that our sponsored research on low/high-carb isocaloric diets, so calories staying the same, the effect of carbohydrate content on performance, cardio-metabolic health, glucose and lipids, that publication was accepted into Frontiers in Nutrition. So this is an effort by Jeff Volek, Andrew Koutnik, Dom D’Agostino, Philip Prins, the folks over at Grove City, and one that Levels sponsored. So we’ll have a dive on that. I think Andrew Koutnik is going to come on to a future Friday Forum to dig in with that. But that’ll be one of our first publications, which is exciting.


And then the performance review system. So the process by which we do continuous performance review feedback for our team, this is currently under a remodel. So we’re kind of auditing the processes, how various managers and employees are using this set of tools. Our goal is to really improve the amount of effort required, the clarity and effectiveness of each of those little sub-check-ins just to make this very consistent across the entire organization. So really appreciate everyone continuing to work together with Nicole and Miz as we… This is really V1, and our goal is to obviously continuously improve, so appreciate everyone’s help there.


And then lastly, we shipped the Whole New Level episode 200, which was between Dr. Molly on her new book, Spark Factor, and Lauren. It’s crazy we’re at 200 episodes. 50,000 plays in January alone, which is almost 10% of our all-time plays. We also did a YouTube Thumbnail sensitivity project, so basically playing with background and visuals on YouTube to see exactly how, through A/B testing, those resonate. And then in-app notifications, this was a big success. So the project to suggest that people rate our app was basically at 4x our total ratings and moved us from a 4.3 to a 4.7 average, which is great. So a lot of great stuff this week. And then we got a bunch of coverage on our, it actually happened last year, but our series A extension. So a bunch of publications in the metabolic space and just general VC news picked that up, which was great. And we’ve been on a couple podcasts and had a couple podcast ad reads go live. I think that’s kind of the main points here. Obviously, lots of other exciting stuff going on, but I think that touches on most of it.


So with that, I want to welcome Mark. Mark is a Levels member, performance psychologist. He works at the University of Denver, and I’m very intrigued by his line of work, which is primarily focused on performance in both the sport and, I think, in the work and life elements that we all kind of are constantly enmeshed in. And besides being an outdoors person and a metabolic health aficionado, Mark is, I think, particularly interesting to talk to in this sense because I’m very curious to hear how you feel about products like Levels and how that interfaces with the psychological aspects of performance improvement, which is really what this is all about, right? We’re trying to get better every single day and be sort of optimizing our health. And although we may not be Olympic athletes, which you have worked with extensively, I’m very curious to hear how you look at these sorts of tools and hear your experience so far with them.

Mark Aoyagi (06:57):

Well, thanks very much for that. It’s a pleasure to be here and an honor to be speaking with you all. To be honest, I wasn’t really sure what to talk about. So I have a couple of thoughts I’m happy to share and then would love some prompts or questions if that works.

Josh Clemente (07:10):

That’d be great.

Mark Aoyagi (07:12):

Yeah, as Josh mentioned, I’m a performance psychologist, work at the University of Denver, and also work with the Los Angeles Dodgers. I’ve worked with other professional sport franchises in the past, so mainly in the pro sport and Olympic space, a little bit in corporate as well and a little bit with musicians. I guess my own experience with Levels primarily started out as just curiosity. Just being in the performance space and being an avid learner and a person that also likes to try and mess around with myself a little bit and see if I can improve a little bit in certain areas, it was just one of these things that just seemed like a no-brainer to try out.


So I guess the biggest takeaways for me on an individual level, and then I’m happy to talk more about my work as well, but on an individual level, the biggest thing for me was really just awareness in terms of… I know these are things that I’m sure you’ve heard many times before, but awareness really in terms of being able to map how I felt and awareness of how I was feeling to spikes, in particular, but also crashes in blood sugar. And the biggest behavioral change that’s had for me is in terms of on when and what I eat prior to bed, because I was noticing wake-ups in the middle of the night when I was having desserts, basically, after dinner. Not that I did that all the time, but when I would do that, I would notice… I would get this spike after I would eat, and then it would level back, but then around 1:00, 2:00 in the morning would get this other rebound spike and was waking up at that time. So that was the biggest behavioral change for me. I know time is short, but that was what I had prepared. And then like I said, if there’s any prompts you want to give me to speak in any other direction, I’m happy to.

Josh Clemente (09:06):

Yeah, I think many of us have had similar experiences with seeing the interface or the interaction between various actions that we used to kind of consider as independent. Like, I eat certain things, and then I sleep, and those are independent almost. And then seeing how that interaction actually plays out with something like molecular feedback is really fascinating for us. I think to your unique perspective on performance optimization, I imagine a lot of that is habit formation and continuous repetitive improvement on a daily basis. I would love to just hear, from that perspective, what you would hope to see Levels work on, whether this is like a feature or a product set or suite of services or something along those lines. But what do you see an opportunity for us that might play on some of those elements of behavior improvement or continual performance improvement?

Mark Aoyagi (09:56):

Well, I think behavior change starts with awareness, just like I mentioned with my personal journey with Levels. The biggest thing that I do… Well, maybe not the biggest thing, but a significant part of my work as a performance psychologist is really just breaking down barriers and removing myths essentially. As an example that’s hopefully not too much of a tangent, having had the privilege of being with world-class performers before they stepped on the World Series stage or the Super Bowl stage or the gold medal event, whatever that was for them, we have this image that they’re calm, cool, and collected because that’s what we see when they step out. And don’t get me wrong. There are a few that are lucky enough to occupy that space prior to stepping onto the biggest stage of their life, and those people don’t need me. They just say, “Hey, go have fun.”


But for the rest of us, they’re throwing up, they’ve got butterflies in the stomach, all those things that all of us experience. So the biggest part of my work is actually telling performers that, “Hey, it’s okay to be nervous before a performance.” The issue is if you believe that you’re supposed to be calm, cool, and collected and then you’re fighting against that before you go out on the field, that’s what then takes away from performance, not the feeling of nervousness itself. So anyways, all that is, like I said, hopefully not a tangent to get to the point of it’s really just awareness and accurate information that allow for behavior change, habit change like you’re talking about. So unfortunately, I don’t have a great idea for a new product suggestion or anything like that. But to me, that’s what your product does, is it supplies awareness and accurate information, and then that allows for people to make actionable changes.


The last thing I’ll say quickly is that in my experience, once people have that awareness and that accurate sense of what’s going on, they can make the change themselves for the most part. And again, I work with a privileged population. But for the most part, they know what to do once they have the right information. Again, it’s just breaking down those myths and getting rid of the inaccurate information. And again, I think your product does a wonderful job of that.

Josh Clemente (12:05):

Well, that’s amazing. And just as good, I think, is that the recognition that awareness is the first step towards any sort of improvement is really key. And I think it’s a great reminder for all of us that this is a tool, first and foremost, to allow that awareness. And obviously, in future phases, we want to be able to help people who are maybe less familiar with the concepts in terms of guiding in the proper direction. A lot of stuff in the hopper there. But the underpinning of this is really just providing that objective data and helping people to contextualize what they’re experiencing with it.


So Mark, thanks so much for joining us this morning. I really appreciate you, A, supporting us as a member, and B, taking some time to spend your Friday morning with us. This is gold for us, and we really appreciate everything that you’ve shared with us. So if you’d like, we’ve got kind of a full meeting, we’d love to have you stick around and check it out. Otherwise, if you have to jump, thanks again on behalf of everyone.

Mark Aoyagi (12:59):

Yeah. And for what it’s worth, I’ll pop my email address in the chat, and if anybody wants to follow up, I’m happy to chat more.

Josh Clemente (13:06):

That would be awesome. Much appreciated. All right.

Mark Aoyagi (13:09):

Thanks, everybody. Appreciate you.

Josh Clemente (13:10):

Thanks a lot, Mark. Okay, I think Jeremy made it on the call. There he is. Yeah. So I want to welcome Jeremy formally. We talked about this last week, but Jeremy Hernandez has joined the R&D team down here in Austin, Texas. Jeremy’s got an awesome background, a lot of startup stuff, doing rapid scaling of a huge range of assays, diagnostics, laboratory operations, and is coming to help us out with a lot of the fun stuff we’re doing down here in Austin. So Jeremy, over to you. We’d love to hear a few words from you and say hi to the team.

Jeremy Hernandez (13:43):

Hi, everyone. Thanks for the warm welcome. Very much appreciated. Yeah, I’m excited to be here, work more on the R&D front. I come from, like Josh said, the diagnostic world. So happy to be a part of this team and focus on the preventative side of medicine for a change.

Josh Clemente (14:01):

Love that. You’re in good company, Jeremy. And everyone, please reach out to Jeremy. Although he’ll be on site here, he’s going through the virtual onboarding right now, the process we all know and love. So please reach out, offer some tips and tricks, and hopefully, we’ll get everybody down here to meet in person at some point soon. Looking forward to working together, Jeremy.

Jeremy Hernandez (14:22):

Thanks, Josh.

Josh Clemente (14:24):

All right, culture and kudos. So somehow Maxine is hitting one year already. This seems crazy fast to me. I don’t know why. I know Max has been doing some incredible work across basically every function of the company. But congratulations, Maxine, really appreciate you. And then had a nice little group in SF this past week, I think. So you got Scott, Azure, Miz, Zac, Hui, and Cissy all getting together in San Francisco last week. Miz is correcting me, which we love to see. All right, Ben.

Ben Grynol (14:55):

All right, stay with me on this one. This is an update on the growth transition, and I apologize if it’s a bit noisy in the background, but we’ll do our best here. So, the startup of a cymbal. What exactly is this? Well, Zildjian is one of the oldest companies in the world, a 400-year-old company, and they make some of the best cymbals. And if you look on the very left, those are what cymbals start out as. They’re these little, flat, metal discs. And they take so many different tools and so many different parts of the process to keep hammering and hammering and hammering until you get something that resonates and something that rings true.


That is very much where we’re at with this growth transition. We’ve got all these different tools. We’ve got lathes, we’ve got hammers, and we’ve got heat, and you’ve got to keep pounding and pounding and pounding. So probably in the fall, probably around the start of Q3, we started having conversations about the effectiveness of demand gen and capture and some of the confusion, some of the questions that came up and how we could be more effective in that. And when Tom took over… I was going to say growth. It’s so natural to keep reverting back to growth. But when Tom took over demand capture in November, we had the initial conversation of should we go back to a single-threaded function and have a single-threaded leader? And we both agreed we thought it was a great idea, but we punted it forward.


So a couple of weeks ago when we were all in Austin, we recognized that we need different tools at different times in any part of the process so that we can really get to… Very much like the cymbal, we take this raw material that looks, we’ll use the adjective haggard at any given time, and we want to make it look beautiful and ring true. So when we talked about who should lead growth, Tom and I were doing the Spider-Man thing, and we were sort of pointing back and forth. I have a lot of confidence in Tom, and I think he’s been doing such an incredible job with what we need right now at this time for acquisition and retention, and to disrupt that would be a misnomer to what we need as a team.


So I proposed, despite many people pushing back, I proposed, “Tom, you should take this on. We all have a lot of confidence in you to do this. And by doing that, that’ll free up my bandwidth to go take on more special projects and do some things like growth, experimentation, and hopefully, we can hammer the cymbal away.” So anyway, that is an update and wanted to give more context on the why and the how we made this decision. But yeah, it’s something that we’re very excited about. And on behalf of the team, it is the right direction to head. So let’s all push for growth, push for experimentation, and we’ve got lots of work ahead. So we’ll leave it there.

Josh Clemente (17:38):

Thank you, Ben. You pulled the analogy off. I’m impressed. Looking at the slide, I wasn’t sure how you were going to do it, but you did it. I just want to boost what people are saying in the chat and also, just from a first-person perspective, really shout out the good faith that went into this process between Tom and Ben just working together to figure out what was best in terms of outcome. Ben is very much going to be there supporting Tom, as he mentioned. He’s going to be taking on some special projects, but primarily focused on that growth function and making sure that we hit our objectives. And I really do think that we couldn’t lose in either of these situations. Both of these people are amazing, and we would’ve ended up with a great outcome. So I just appreciate the two of you working out to pick the better of the two good outcomes. So yeah, great work, team. Okay, the rest of today is basically the Chris Jones show. He’s got two segments I’m excited for. Chris, over to you.

Chris Jones (18:34):

Thanks, Josh. Maybe a month ago when Miz asked me if I wanted to share some of the maybe classic war stories over my tenured career, as those of you who know me, I didn’t hesitate for one second. I don’t claim to have the right answers or the right strategy, but I claim to have lots of war stories that are fun and entertaining. And hopefully, if people take things away from it, all the better.


So just a quick reminder, I’ve been in tech for almost 30 years, with my first internship at Gateway 2000 Computers in the mile-long cow palace in Sioux City, Iowa, with Ted Waitt riding up on his Harley Davidson in blue jeans when… Most of you are like, “I don’t even know what Gateway 2000 is.” It’s an old computer company against Dell. So the gray hair is real. It’s not my highlights for my TikTok account.


I’ve been at 14 tech companies over that. I’ve managed a ton of people. I’ve been through a lot of reorgs. Essentially, you get used to being in reorgs. Especially a place like Google, if you don’t reorg every nine months, you’re like, “What’s going on with my group?” And I’ve been through a lot of layoffs, company-wide. Most of these happened during the dot bomb, where every quarter my friends and coworkers were getting cut, like, quarter after quarter after quarter. Doing pink-slip parties is actually making fun of it. So I just bring that up and clear… And actually, I also had to fire myself at one of these times.


So it is kind of part of the industry. We go through cycles. I know it’s scary right now with a lot of the layoffs going on. I just want to make sure that this is just kind of a natural part of it. And I’ve gone through two IPOs, which are super fun. So from the analogy of the O linemen, I brought this up both for Flanagan as well as for those of you on trivia night. I played offensive line in high school football. I was the smallest lineman in the entire state. I’m not saying I was good or not. I’m just stating facts.


All right, Josh, next slide. I’m going to go through in history of some of the jobs and bring out some of the lessons that I had while I was there. Some of them are good, and some of them are bad. First, I’m going to start with Intuit/TurboTax. Intuit, as a company, is customer obsessed. Scott Cook, one of the co-founders of Intuit, was one of the really first early adopters of the Net Promoter Score. You can’t talk to someone who’s worked at Intuit and not have them know what Net Promoter Score is and how to use it, and you use it as your North Star. It came up in every conversation. The amount of energy around UX studies, surveys, Follow Me Home.


We had teams of people every tax season that would go into people’s living rooms with all the tax receipts and watch them do their taxes one receipt at a time. And we were looking for inefficiencies and ways to reduce cognitive load. So how that manifested itself a little bit in the No ROI Required was when we were sitting around at TurboTax every week during tax season, and I would be reporting to leadership around the top reasons of our call drivers around people can’t log in, they can’t download, they have this error code, they’re running into issues, there was no questions around… It was the, “What is every team doing to reduce their areas of customer friction or customer pain points?”


There was no conversation about, “Well, how many engineering hours is that going to take? What’s the expected savings from support cases? What’s the expected increase in retention?” It was the, “If a customer is being frustrated or has friction points, we should do everything we can to remove that, and that will pay dividends.” So it was never this, “Let’s ROI everything.” It was the, “If we’re letting down our members and our customers, we need to fix that, and let’s get it done.” So that type of attitude just really streamlined the team taking action for this.


And then where I saw it come to life even more was for the last tax week, you can imagine… The software shipped. We are about to go through the April 15th filing of people like me that file at the last minute, and the volume spikes up the wall like a rocket taking off in that last week. We’re not shipping any new code. There’s no more code to do. Every update, every feature is in. We would get the entire company into the cafeteria, and everyone from GMs to product leaders to engineers would be sitting on taking phone calls, answering customer chats, doing emails, because that’s the only thing we could do to help them get over that burden of filing their taxes. So the entire company became a support org for that last week.


And the amount of touchpoints that related when someone said, “Hey Chris, you’ve been telling me that people can’t download the desktop app every week this entire tax season. I just spent two hours on the phone trying to walk someone through downloading a product that I built, and that was super painful. I’m going to fix that immediately.” So it gave them that sense of like, “I can now feel the pain.” It was no longer a number on a chart. It was the “I had that uncomfortable conversation with a member” to get it done. So a lot of my obsession for the customer and voice of the member and relentless beating of the drum came from my time at Intuit and TurboTax.


Next slide, Zynga. And for context, I joined Zynga one year pre-IPO and was there one year post-IPO, so two years. The fastest rollercoaster going up and the fastest rollercoaster coming back down in a two-year period. The first thing I noticed was employee empowerment. During my onboarding week, I had heard all these rumors about this massive Vertica database that they had that had access to every app click, how many friends people have in FarmVille or CityVille, how much money they spend. I’m like, “How do I get access to it, because I’m a SQL junkie?” And like, “Oh, go to this room, and they’ll give you access.”


I walk into this room. It is standing room only with 75 people in a conference room. The person asks the question, “How many of you in here know how to write SQL or have written it in the past.” Only 25% of that room raised their hand, and I was shocked. I’m like, “Hold on. 75% of these people don’t even know what SQL is, and you’re about to hand over the keys to the kingdom and give them access to this massive database of member information?” And it was all about empowerment. They didn’t want to create bottlenecks of data scientists or business analysts. It’s like, “Let’s get the data to the people making decisions as fast as possible.” And that was incredible to see, and it was a machine.


Ville It. This is probably one where it was actually not a great success story. With the success of games like FarmVille, it convinced Zynga that they could A/B test their way to a successful game in any genre. We’ve got the playbook. We’ve got the data. We’ve got the infrastructure to basically… Ducks, crops, whatever, we could just A/B test our way to it. And then what they realized over time was they were a victim to their early success. Things that worked for them before, they just kept repeating and rinsing and repeating and rinsing, and eventually just it wouldn’t work anymore. And the industry had moved on to mobile. They’d moved on to bubble games. They’d moved on to puzzle games. And Zynga was like, “Where did we miss the boat?” And it’s just because they were so successful early on, they became kind of drunk on it.


And the last one from Zynga is incentives matter. As in the customer insights org, and I was the one driving all the insights from the NPS surveys for every game around why people loved every game, why they hated it, and trying to drive product improvements. And I couldn’t figure out for the longest time why a lot of the changes we were recommending from our team were never getting implemented. And as we popped the hood, we found out later or actually during it that the engineering teams were very ROI driven, the exact opposite of Intuit. They had rigorous spreadsheets around engineering efforts, estimated revenue, return on investment, and they would sort by ROI. And that’s exactly how they did product roadmaps.


It was a machine, and it was actually very good and predictive. But I realized that all those engineers at all those product teams, their bonus was also tied to revenue growth, so they would only listen to you if you had an idea that would drive revenue. They didn’t care about retention, new users, whatever. They just cared about revenue growth. They did not care about how happy the game was. They did not care about all the other metrics that we kind of try to balance it. It was all about revenue because that’s how they were incentived. And 100% of their bonus was driven on revenue targets. So be careful how you reward your team because that will drive behavior and often the wrong behavior.


Next slide. Beartooth, for those of you that don’t know, this was a small hardware startup in Bozeman, Montana, that I spent another two years at. You’re seeing a theme and two years and doing the math of my average. Miz, my two years at Levels is coming up pretty shortly, so you’d better watch out. A couple of things I learned at Beartooth. I learned a lot. The first one is emails do not equal product market fit. Similar to Levels, we came out… We launched our product with a database of 100,000 emails. I was doing ad campaigns. I could get an email for 25 cents all day long. So if my boss said, “Chris, I want 25,000 emails,” I’d be like, “All right, here’s how much money I need, and I can get you that many. And I can do it by any industry, any genre, any graphic,” just because Facebook was that good at targeting.


So when we did our launch, and we did a big unveiling launch event: light the fireworks, the press release at the same time, the Kickstarter campaign, and we started blasting our list of emails. We needed 4% conversion for us to break even, for this thing to work and to kind of get out of Earth’s orbit. We barely hit 1% conversion on that. And over the course of several years, we hit that list over and over and over and over again, trying to get as much juice out of it as we could because we were in survival mode. I think we eventually maybe got 1.5 after hitting that list probably 20 times.


And this is a time where we went from a layoff where I had to fire myself. We realized we were not going to have this massive hardware Kickstarter event and paint our own picture. It was the fight for survival, and we fired 75% of the company within about a month of us trying to do like, “Okay, what can we fix?” And we realized our issue was we had not solved for product market fit, and that kind of gets into lesson number two of who are you building for?


We had a divide amongst our two co-founders. One co-founder wanted to build the device to take to Coachella or a music festival, where you communicate with your friends, or go hiking off the grid, very much more of a consumer lifestyle product. Our other CEO, who was a smoke jumper, wanted to build a badass radio that would save lives. So his interest was building something that worked with the police, the fire, safety, and all these proprietary radio systems. And they never really had the conversation. They kind of thought they could do both, that we can build this radio that will solve all these needs and be sleek and sexy and affordable and that you’d want to take and show off on Instagram. And that was a big miss. The consumer market just wasn’t there. Cell phone companies were getting better and better, and people didn’t want another kind of accessory in their thing. So we missed product market fit by a mile, into who really wanted this product.


And then lastly for Beartooth is persistence. Two weeks ago, I just had dinner with the CEO, Michael, who was in town. So we had dinner, and I asked him how it’s going. He’s now in year nine of being a hardware company with three employees, and they just hit their first profit. So he did not let up on the gas. It’s like, “Wow.” He’s like, “I thought this was going to be a huge exit,” and then he ground it out. And all right, what is our market? And they’re basically selling radios to black ops. Things like the Ukraine War is their biggest customer right now, people going into harm’s way, and it is the radio saving lives. But it’s because he was persistent of like, “All right, this is not the exit I planned for, but I want to see this through.” And he finally started to kind of round the corner after nine years in hardware. Next slide.


And lastly, Nest, the company just before actually becoming Nest/Google-ish. But the first one was support is a product feature. There are a lot of companies and probably a lot of us here have worked for companies where support is an afterthought. It’s a cost center. I’m looking at you Riley, where it’s like, “How do I make that cost go down? How do you make it cheaper? How do you reduce it? Because it’s hitting my COGS and my profitability.” And usually, support is a big dollar amount, so a lot of companies look to optimize it. They offshore it. They automate it. They’re like, “Why do people need to call us?” So they hide the phone number.


And at Nest, and a lot of this was due to people like Tony Fadell, of we lean into support, just like Apple kind of leans into the Genius Bar. It was stamped on every box of 24/7 support. It was a reason that a lot of our members would say, “I’m going to buy a Nest thermostat over the competition because I know you’re going to be there for me when I’m trying to rewire my house.” It’s something we had to lean into, and we treated it just like part of the product experience. It wasn’t a, “Well, how do we reduce this?” It was a critical part of it. Similar to Levels, our contact rate was about 50% in that first month for the thermostat. It was a critical thing. It’s a hard thing to rewire your house, especially when you’re talking about something or putting a smoke alarm that could cause damage.


The next learning was Empower the Front Line. So what you’re looking at is a demo board that I asked our call center. Now, these are not full-time employees. This is a typical vendorization, where when we merged Nest and Google, we talked in theory about we’re doing it because it’s better together. You have all the great hardware products from Google. You have all the great products from Nest, and they’re going to work better together. And we were waiting for the marketing and product teams to tell us what the better together look like, and they never really did.


So we turned to the agents who deal with the products every day and said, “Hey, can you guys actually build your own demo boards and actually come up with cool ways that these products will work together?” So we built competitions where they actually would say, “Who can create the best scenario?” So you can imagine people building domino boards around like, “All right, I’m going to walk in front of a camera. The motion sensor turns on the door lock. The door lock turns on this light, and then Google Assistant speaks to me.” So they would create these super-long scenarios using all the products. And a lot of the ideas that we ended up then creating with the product came from the support team, because they were the ones closest to how these products actually work and what problems our members were trying to solve by using them.


And then on the right, this kind of six-by-six block of 60-inch TVs was behind my desk in Nest. It was our command center. It was how we looked at our cases coming in minute by minute. What’s blowing up? What articles are shooting up? What’s the weather in Austin of the cold snap? Because that’s going to drive a lot more contacts when there’s a cold snap or a heatwave. So this was our command center of realtime data of our agents answering the phone. “Are they on leave? What’s going on?” And part of this was so we actually knew what was going on across our org with all these different datas real time. Another was to promote support within the company as a function.


So the number of times people like Matt Rogers, the other co-founder of Nest, would walk by my desk, would look at the wall and would stop and say, “Hey, Chris, what is this?” And I would walk him through the data and what we’re looking at and how it drove support actions, and he’d be like, “Oh my God, this is amazing.” So to some degree, we did it because we wanted to know what was going on, but we also did it to promote what support does and the data that we look at to drive great support within the company. So we were kind of doing it as an internal marketing to make sure… And the goal was so that people like Matt could sleep well at night knowing that we were on the ball. That was the end goal, that we know what we’re doing, and we can provide confidence, and you don’t need to worry about it. And when you do, we will let you know.


Next slide. And lastly, no regrets because those were the droids we were looking for. My comment on this one is if you get into the, “Oh, if I would’ve just joined that startup, if I would’ve joined at this time, I could have maximized that IPO. If I would’ve worked for this boss or done that,” you start getting into this… Hindsight’s 20/20. I’ve had a lot of experiences. I’ve worked for a lot of incredible bosses. I’ve worked for some awful bosses. And every time, I’m looking at it going, “What did I learn from that situation? What can I take away? What is going to make me a better person, a better employee, a better boss?”


And there’s always something you can learn from every situation that you’re in. So if you look at it in terms of a no-regrets attitude as you go through life’s journey of the career and just saying, “All right, I’m going to make the most of it. I’m going to do my best, and I’m going to take away everything I can from this.” And that’s what I’ve got. Sorry for the long rant, but I could keep going, if you know me.

Josh Clemente (37:02):

That was awesome. Thank you, Chris. Round one of the Chris Jones show or episode one of two. Deep. Yeah, we’re going to have to definitely clip this one and re-share. All right, quick aside into company objectives. Levels shows you how food affects your health. It’s the main thing. Everyone should be working towards this. Please raise it if you don’t feel you are. And product is our top priority in Q1. I’m going to hand it over to Maz.

Maziar Brumand (37:29):

Thank you. Welcome to the product update of February 3rd, and standing ovation for Chris. I’m going to stand in his honor for his TED Talk, which was phenomenal. Thank you. I wasn’t sure when I’m going to go on, and I just didn’t want it to end. All right, let’s get in. First and foremost, just repeating for everybody, so we know the prayer of product for 2023. We’re focused on Maureen. We want to be customer focused. We are customer focused. We are customer obsessed, as Chris mentioned, and Maureen is our customer to cross the chasm with. And we’re creating a product that will improve her health through the loop of guidance, action, and accountability. Next slide.


Okay, we wanted to give you a quick update on what have we been up to and continue really two themes, and we’ll repeat these themes week over week. And we’re going to actually put a process that we can update everybody on a weekly basis through async. But really, what are we learning, and what are we doing to build based on those learnings to deliver on the 2023 vision?


So in terms of learning, we’re continuing to learn a lot based on the tools that we have at our disposal before we even build anything. And as you guys all know, Stacy is doing the Guides V4 experiment, and we have learned a lot from that experiment. And we are actually kicking off a V5 with our very own Sonja, and we want to try a number of hypotheses here. I think one is can we actually create the daily loop and a journey for a member to improve their health? Can we 10x the Guide output by giving them the tools and processes that we’re just so good at creating and systemizing at Levels? Can we actually leverage our company DNA to enable the Guides to create content and learn from it?


And can we actually take an emerging Guide, not somebody like Stacy or Casey, which have 100,000 followers, but somebody like Sonja that only has a few thousand, and can we actually use our channels to market her and have people actually pay to join the experiment? And then finally, we want to know if people are willing to pay for this experiment, because the best thing to know whether you’re creating something that people want is to see if they’re willing to pay for it. So we’re testing willingness to pay based on a new set of variables of a new Guide on a new distribution channel, and that will teach us a lot.


In terms of building, we are putting a lot of thinking through how do we actually do this in a way that’s going to 10x our speed and acceleration. First and foremost, we’re working through a scoping exercise, where we’re actually creating different phases for the product launch. So we can learn quickly going anywhere from the skateboard scrappy alpha to beta through V1 and so on, and really shooting to have something for people in the app in a really scrappy way and build on top of it. So we’re not going to build Taj Mahal and say, “Come try it for the first time,” but we’re really going to try to create the skateboard and have people try it and get feedback.


We are creating our very first engineering pod so that engineers are more empowered: more context, more ability to make and contribute to the progress in a much higher velocity. So we’re going to experiment with that. It’s a new way of doing things for us. It’s obviously been done in industry now for years, but it is time for us to experiment with that. So that’s something that’s coming.


We are boosting cross-functional collaboration. I think for us to really, really be successful, I think we need to have everybody on deck and work together really seamlessly. And we’re really excited to do the deep collaboration with both growth and strategy to answer a lot of the questions that we have and really have everybody pulling the same direction. And then finally, we’ll have a weekly update which captures all the stuff that’s happening in product in a way that it’s easy to digest, to let people know what we’re working on and how they can help. So hopefully, all these processes will lead to even faster cycles for learning and building the product. That’s all.

Josh Clemente (41:45):


Maziar Brumand (41:46):

Thank you.

Josh Clemente (41:46):

Thank you, Maz. Yeah, it’s awesome to see. Okay, back over to Chris for the Voice of the Member and other member experience updates.

Chris Jones (41:56):

I had to get some bulletproof coffee to recharge for this one. Next slide, Josh. I’m covering three things today, really more kind of project updates. So the first one is Otonomee. Quick reminder, Otonomee is a boutique BPO about augmenting our support team. So they are about working with startups like ourselves who are looking for, “Hey, I need two agents. I need six agents. I need 10 agents.” They are not the prototypical convergence of the, “Hey, I want 10,000 agents around the world all the time.” These are a small company looking to meet our needs where we’re at. So it’s all about support augmentation and can we scale and be more effective, is kind of the goal of the pilot.


A couple highlights. So as they’ve come out of onboarding, we are tracking the number of members that they’ve helped. So you can see on the left, it’s gone from they’re averaging 50 to 100. This week, I think there are about 125 members helped per agent in a week. And for reference, the team average is 150. So if I look at the performance of our existing team, we’re in the middle, helping about 150 members per week per agent. So they’re within earshot in terms of the productivity. On happiness, they’re at 85 so far in, which for the same time range, I think the rest of the team was like 87. So a couple points away from the rest of the team. So they are getting really close to being at par with us.


Now, there’s still some rough edges. There’s still some cases that they have harder times with, but every week they’re showing progress. Doing daily looms, letting us know where they get stuck. And they actually really enjoy the async nature. They love the onboarding. They’re really excited about what they’re doing and why they’re here. Some of the lowlights we’ve run into was they’d never used a CGM. Getting them access to the Levels app. I want to do a quick hat tip to Maz, Karen, and others for trying to help me unlock that. When they were saying, “Hey, I don’t know what the app experience is going through it because I’ve never seen it before,” it is identifying some gaps we have in terms of could we scale this and how do we get them access to it.


Another one that I forgot to call out is no access to comms. So the amount of communication. We have to kind of double over to Slack to get them access to it or to Notion is just inefficient. So that’s an area that is also a lowlight in terms of getting them access to data and the conversations. But so far, it’s going really well. We’re excited, and the key thing is they’re learning every week. Next slide.


The next experiment. Mr. Flanagan jumped all into, “All right, tool…” So we use Zapier a lot for kind of lightweight automation, doing tagging, moving data from Help Scout into Notion or into Sheets. And they recently opened up OpenAI, which essentially is ChatGPT, as part of a Zapier class. So Matt’s like, “Great, I’ve already got Zapier hooked up to Help Scout. Let me see what ChatGTP can kind of do against Help Scout.” So he took a couple areas and said, “Can I have ChatGTP summarize the inbound question and can provide recommended answers, not as a chatbot to respond back, but more to augment the agent of getting it, of like, hey, this case has already been pre-summarized with maybe some suggestions for you”? To try to make our agents more efficient was kind of the goal.


Next slide. I’m going to show you a couple examples of what that looks like. So, on the left we have Pretty Good. In the top box, you have the inbound question coming from the member, and then below in the red box, you have what ChatGTP did around summarizing it and with suggestions. So in the Pretty Good, some questions around people asking around what’s the wait list, what’s the commitment, what’s the CGM, and the tool identifying what the questions are and coming up with some simple kind of responses. Great, that’s actually pretty good. It may not be what we respond with, given our snippets, our language, but it’s like, yeah, that’s technically accurate.


The Not Quite in the upper right was it had some things, but it directed them to the [email protected] versus the Levels’ support org. It kind of missed some of them, but it was kind of like, all right, some of these are actually pretty wrong. And the one that we got the most laughs out of was the bottom right, where when the customer said in response to a newsletter, “Please stop, you don’t support my phone,” ChatGTP said, “Does the person not support the phone? Do they not have a good relationship with the phone? Why is the person sending this message?” So this is one that we just really had a good laugh with.


I think Matt’s test said, he said, “Maybe about 33% of the time it was doing an okay job.” So likely, we’re not going to move too much farther with it. But I think it’s good for us to at least kind of test it, experiment, and try to figure out what the limitations, what these things are good at and what they’re not. So I just want to tip my head to Matt. He took this out on his own with no direction and just said, “All right, let’s see what we’ve got.” So thanks, Matt, for kind of taking this and running with it.

Josh Clemente (47:12):

Another awesome segment. Some really cool projects in there and… Yeah, just reading the chat. Yeah, very cool. Thank you, Chris, for carrying the team this week. Great insight into the member support side and some really awesome wisdom density there from your history in deep tech, high tech, rather. All right, hiring updates. So currently, we do not have any open roles. With what’s developed on the R&D side, we are currently at a point where we can take down that general posting and really lean in with the team we’ve got. If you or someone you know is interested in a role here at Levels, definitely still check out levels.link/careers and send in your referral or your personal background, and we will reach out when that changes. With that, we’ve got a couple minutes for individual contributions. And actually, I’m going to ping Nicole here. We might want to just jump into the cafe earlier. Is that better, or what do you think here?

Nicole Miller (48:11):

That sounds great to me. We can do breakouts, and folks can leave when they need to with the breakouts.

Josh Clemente (48:19):

Okay, let’s go ahead and do that. It’s first Friday. We’ve got some extra time here, so we’ll do the breakouts first. I’ll hand it over to you, Nicole.