June 30, 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):

Welcome to Friday Forum, June 30, 2023, the last of this half of 2023. And getting into it, recent achievements. This week, we had shipped the Signup 3.5 version of our website checkout, and it was a controlled experiment, meaning we were comparing a new update that included various improvements to the CGM descriptions and visuals, combined with shifting the research study. If a person participates in the research study, the consent form to, after checkout.


So this was a positive result. We we saw a conversion rate uplift of 26% on the web, desktop, and a 40% improvement on mobile versus that control. So it’s a great outcome. So we’ll be shifting 100% of traffic over to that new 3.5. Great work on the growth production team there. We’ve got a new transactional emails campaign live, so those are all going out. It’s upgraded to match our new product experience.


And we have some really nice improvements in our CAC, or customer acquisition costs, from performance ads. So that’s moving in the right direction. Still aren’t at the best targets that we can hit, but seeing some good improvement.


Site visits are way up this week, so we had actually the highest signup starts since March, and we are also on track for ad conversion rates for the week, or the month, rather. So all good things. We have some major updates coming to the performance marketing side of the business starting in mid-July, through our work with the agency, Sharma Brands. So we should see some great work there.


On the docket for growth and Signup, we have a single home page, so we’re going to be testing and experiment, basically A/B testing a new single unified home page, and that will precede and run basically in parallel with pricing and mobile upgrades. So we’ll be working on pricing and positioning, and then upgrades to the mobile checkout experience.


We’ve got a Q3 growth memo as we’re coming to an end of Q2, and we’ve hit most of the targets there, if not all. We’ll now be setting new targets for Q3 on the growth project side. Got an abandoned cart campaign that is in work, and then really nailing the metrics dashboards that will allow us to run validated test campaigns across all the surfaces that growth touches. So a lot of work queuing up a big Q3.


On the product sides, we had our unified app, so basically the project we’ve been calling App Surface Area Reduction. We’ve been really leaning out a single experience of a version of the hybrid, basically the app experience that can work whether or not someone has a CGM on. So we’ve leaned that out. We’ve now shipped that to the internal team and to 5% of our members. So this is a big one.


Been working on this aggressively over the last month, and really need the whole team to give this a shot. This is a big improvement as well for our members, as we now have a single-app experience. Our team is going to be able to more effectively triage issues, we’ll be able to move faster on product feature updates, and generally work towards that hybrid experience that can allow someone to flex in and out of CGM use.


We had a nice survey that went out to all members, members who had done a Labs V2 experience, so that survey will definitely inform our next update for Labs, which will be coming later this year. And then we’ve restarted the shaping on the accountability 1.0 project. So Cosima’s got some great work there, and Victor and Sam had done quite a bit leading up to this.


So we’ll be modifying and executing on accountability 1.0, and then there’s just a ton of other projects that are in shaping for products. So now that we’ve gotten the Surface Area Reduction and the unified app out, we’ll start to turn our focus to the next iterations of the app experience.


On the validated project side, so we had our first, or one of the first, validated product features. And validation, as a reminder, is when a project has shipped and we’ve received or we’ve taken in measurement on the success or lack thereof of, of that feature, that’s when a project is actually validated. So you can ship something, but we don’t actually know whether that worked until we measure its impact and report on it.


So we use the tappable chart v1 experiment as the first of these product validations. Jhan reported on that. So that is now done. We’re not going to move on to a v2 until we improve the baseline metrics on tappable charts, but Jhan’s going to give us an update there. And then we’re validating the Amazon and G7 experience, is how we’re shorthand referring to this, with a Dexcom G7.


So that is now live to 5% of our members. It’s also live to an internal team. If you haven’t tried this, please do. We definitely need to really dig out the issues in this conversion chain that has to happen for someone to be able to go from Levels website all the way through to having a G7 fulfilled through Amazon and our partners that enable it.


We have a new logs user experience update. As we all know, we had a two-step logs feature go out and a lot of people were getting hung up on the complexity of that and just jumping out. So the new UX fix allows you to, just in a single step, log a meal quickly. And then if you’d like, you can add additional information. So we’ll be, again, validating the success of that project.


Then we’ve got the new landing pages that are being built out. We’re also migrating from WordPress to React on the back end. Lot of work going on there to be able to enable a better infra. All right, new refund policy is out. So thanks to the support team for continuing to really find areas of opportunity to improve and lower costs as well while continuing to give people an exceptional experience. We’re at 93% happiness this week. We’ve got six weeks under our replacement rate target of 7%, and eight weeks under our contact rate target, meaning that we’re really doing great work on leveraging our existing information base.


All right, think that’s everything. With that I want to welcome Ari Tulla. Ari has been a member of Levels for quite a long time. I don’t exactly remember, I think it was maybe in 2019 or 2020 when you first came into the Levels orbit. Also an investor in Levels, CEO of Elo, which is a smart nutrition service, which basically enables nutrition supplementation based on blood work and I’m sure many other things.


The goal is to transform food into a form of medicine, which I know we all recognize and appreciate. Ari’s a pretty prolific investor as well and a huge portfolio of startups that we all know and love. And so Ari, thanks for taking some time this Friday morning. Really appreciate you joining us to share with us. I’d love to kick off just the things you’re most excited about in the world of metabolic health.

Ari Tulla (06:46):

No, thank you. Good morning, good afternoon. It’s great to be here. And I think, I’ve been following you guys for long time and I think you guys have been a huge inspiration for not only for me but many people. And I’m an operator, so I look at you guys as a wonderful example of a good operating system and how to build a startup. And all the stuff that you guys been done publicly and sharing and so forth.


I think, I mean, I don’t know how many people, I don’t know if you know how many people you inspired. So that’s, I think, a big thing. Of course, you’re building a great product and helping people, but also helping the startup ecosystem become more transparent, I think. Huge kudos. You’re the only company who is doing it that way and it is pretty unique. And I hear about it a lot.


So yeah, I’m a biohacker, since long time ago. Turned into health entrepreneur because of family health issues. My wife was sick 22 years ago. She had a thyroid tumor, led into autoimmune diseases, hormonal imbalances, and a verdict that we can never have a family. And we spent about 10 years hacking her health. And, of course, I went along to find a way to overcome the issues, get out of medication, and then finally have a family.


And it took us 15 years and a big battle, and we did it. So I became a massive believer in idea that food is medicine, and metabolic health, as you guys have now coined it, more or less, is very critical piece of that. And I think for me the most interesting thing that, and I want to throw a ball a bit for you guys. I don’t know if you have heard any other people or if you guys have experienced same thing, but for me, after now trying out Levels for, I don’t know, maybe two years, and I’m involved in Supersapiens.


I mean the very guys are my old friends from Finland. I’m from Finland. So knowing this market pretty well and talked to a lot of Dexcom and Abbott and others. But I think, for me, personally, I realized that early on, the things that everybody has, the oats were not good for me, rice and corn that are not part of my native diet of my people. I come from northern Finland where you have basically game and berries. That’s all you eat and maybe some bread and milk.


But, for me, I learned a habit, maybe year ago, I realized that I can actually now feel in my body when my glucose is spiking, not when it goes down, but when it goes up. And I created, it’s like an inner ear feeling. I don’t know how to explain it, but I believe I have it. And to me that was a fundamental thing. I’ve been wearing lot of different things over the years. I have a lot of data. But first time with Levels and with the glucose monitors, I learned something new.


I developed a new habit, or maybe it’s like new sense in my body. And to me that is a fundamental thing. We all talk about data can be suffocating, it can be disruptive, scary, but in this case, what if we can learn to be better humans by using the data in the proper way? So that, I think, to me the most exciting thing that I’m thinking now a lot, can the data, can the sensors make us not cyborgs, but can they make us better humans? And that’s pretty powerful notion. I don’t know if it’s real, but I’m trying to think about that a lot.

Josh Clemente (10:01):

I absolutely love that section right there. We refer to that development of a sixth sense as interoception, the ability to perceive, learn to perceive additional things about your internal state. And I certainly have the exact same experience, so I can now, I can pretty accurately predict my blood sugar within a few points, at any level now, after using a CGM for multiple years. And that’s just repetition. Did I lose my sound? Sorry, folks. Yeah. Okay.

Ari Tulla (10:29):

No, you’re good.

Josh Clemente (10:30):

Okay, okay. Yeah, so I have developed a very similar thing. And I’ve heard of it from many people, many types of people, with many different goals. It’s just when you start to get feedback, you connect these small, subtle signals that you’re collecting from your five senses and it refines into a sixth sense. Which I don’t think there could be something more powerful when it comes to behavior change, whether or not you have a device on your body.


And so, that’s really the inspiration behind our hybrid approach to our app that we’re building on, which allows somebody… It doesn’t enforce continuous device use. You can flex in and out as it works for you. I think that’s the direction that this space needs to take, because we don’t want to become better machines, we want to become better humans. And I think that can be supplemented by tools like this.


Ari, you’re building a lot of really cool stuff and I think you have, just having spoken with you and watched a lot of the calls that you’ve been a part of and seeing the business you’re building, you have a really good sense for where these companies need to go in order to attract a large, diverse consumer market who will ultimately prop up and enable us to build very scaled, wide distribution-accessible products that will improve health.


What is one thing that you see Levels not doing? What could we do better that you think, just as an investor, as a user of Levels over the years, what could we do that you haven’t seen from us yet?

Ari Tulla (12:02):

Yeah, I love the fact that you have the coaching component. We have that as part core of the Elo service. We slap it on as a default, but frankly, it’s too expensive to really scale. I mean, we can look at Noom, we can point fingers, and I’ve been involved in some companies that went down already. It’s just unit economic-wise, but difficult to do.


And I think the one thing that I think, I’m sure you’re working on it and doing stuff, but I think deploying AI in more in a conversational manner will be the biggest thing for you. And it actually could make Levels, and many other companies probably also Elo, to actually get, escape velocity at the unit economic points. Because we all know that, and you guys know this better than anyone, that being a slave to a device that you don’t own.


I mean, my company is hundred percent that. We don’t produce any of the food, any of the testing. We are slave to all our partners. And it’s a tough spot to be, because you are living in a margin on top of margin game. But AI might be, and APIs and AI might be the unlock that could actually become the thing. And, I mean, we have more and more people now who are so used to do it, and a billion people have likely tried some type of a conversational AI.


I just downloaded the Pi today. I don’t know if you have tried it. They raised 1.3 billion announced yesterday. Unbelievably good. I mean it’s really cool. Different notion of it. Same, same, but different. Try it out Pi. You can download for free on App Store, on iPhone only. But that’s the thing, I don’t know how we can overcome the fact that in medicine, and we all science minded, we all believe doing the right thing.


I don’t think we can get deploy the AI for medicine, but can we deploy it in the nutrition somehow, coaching somehow? I think we can soon. So that, to me, is the big unlock that I’m most excited about. And I maybe year, two years behind you guys as a company. We are smaller, much smaller. I believe there will be holistic companies coming in five years, a holistic place where maybe Amazon is the one who has the most tools today.


Because they have Whole Foods, they have buildpack, they have lot of AI, they had the wearable things, they have Alexa, and they have One Medical. So they have really good assets that they actually do something unique. But before that happens, it’s time for small companies like us to really sow the way.


And I think many of us will be bought by this big companies, which is great. I mean, there’s no problem of going to big companies and build it there. But it’s tough to be a small play and tough to have some money, but not billions of dollars you might need to solve this problems, and it’s just very difficult.


And I think the one thing that you have done really, really great work, and I think, I really take my hat off on, because there’s no science. And you guys have amassed amazing group of people who are helping you guys on the science perspective, and those are the right people. And the current science, we took all the literature you have really done on the nutrition sciences and it’s not a lot. The corpus is very narrow and even if you use algorithms and AI to mine it, you don’t learn a lot.


So, to me, that’s the tricky part, what to do about it. And I also, I think I really like about the approach you have on the biomarker testing and all things we talked about it, because we are testing the wrong biomarkers. At Elo, we test the normal biomarkers because people want them to be tested. Those are wrong biomarkers. The lipid panel does tell anything. I mean, it’s crap. A1C’s okay, but it’s not really actionable. It’s very slow. So you guys looking at the right biomarkers is really great.


Tough to do on the finger prick at home, because they can only be done with [inaudible 00:15:58] now. I think that’s also the big deal. In three years time, I think, and you probably talk about this internal lot, but biomarkers and biomarker testing will completely transform healthcare. I think that’s the quickest unlock that will happen. Not really the sensors.


Because if we can do a $50 price point, 200 biomarkers, in few years, that will be completely game-changer. And doing that every six months will give us so much data that we can really make a difference and have more, I think, a stronger point of view, because the data is your blood. People will believe in that. Blood is almost like religious when people think about it.

Josh Clemente (16:39):

Completely agree on all of this. This is, it’s really fascinating to hear the perspective, because you’re seeing a lot across the industries. I think, I’m sure that comes from being an investor in such a broad spectrum of companies. And we feel very similarly. We’re taking a bit of a risky play, in a sense, with our Labs approach, because we are going outside the norms and testing through things, like APOB, which most people don’t really understand. But we’ve got to push and pull a little bit, make sure that we have A1C in there, because that’s such a respective marker. But then, simultaneously, test for the real underlying lipoproteins.


I also love what you said about AI. I think there’s a generational opportunity here to really leverage the information that does exist, in an easy conversational way, that’s approachable for anyone, because they’re asking the questions. Pull information rather than push information. You don’t have to try to predict what this person’s going to ask, they can ask you and then the model can pull from the corpus. It’s just getting rid of those hallucinations is going to be key.


Anyway, there’s a ton here, and Ari, I would love to really jam on this stuff, and maybe there’s a podcast episode in here somewhere to talk about the future of this space. Really appreciate you coming and joining us this morning. The whole team absolutely loves this stuff. Yeah, appreciate your support over the years as a member, as an investor, and as a builder in the space. If you’d like to hang out and follow along for the rest of the meeting, we’ve got a couple more segments here, but if not, if you have to sign off, thanks again. We genuinely appreciate it.

Ari Tulla (18:07):

Hey, thank you all. Keep up with the good fight. It’s never easy, but the payoff in this market will be unbelievable. I mean, 10 years from now, we can all look back and say this is the best time of our lives, I think.

Josh Clemente (18:18):

I love that. Same to you. Best of luck with Elo and everything else. Bye-bye. Okay, that was great culture and kudos. All right, we’ve got a couple here. So Tom is hitting his three years here at Levels, so that’s huge. That’s yesterday. Already has hit. So congrats, Tom. It’s been awesome to work with you. Tons more ahead and, yeah, just really appreciate how much has been accomplished over those three years. Several podcast tours and many more.


Galit, big shout-out to Galit. She has taken a second… I mean, this is just an example of just classic team-player approach here. We’ve been really trying to press on some major growth initiatives which have required Juan’s support, and Galit stepped in to take on another week of on-call to backstop Juan, so he could take over that and get that pricing experimentation done sooner. I mean, this is just so important. The cross-functional work here and just making sure everybody’s supported has been awesome.


And Galit has also supported some very short-notice work and data generation on some of the research initiatives and DEXCOM initiatives that we’ve got underway as well, on very short notice, and really compelling stuff. Just a ton of awesome work from Galit. Really appreciate that. And then, on the ownership side, we’re in this new culture of measurement and of closing full loops for validated learnings and validated projects.


Just want to highlight this example from Rafael, really making a great show of ownership on one of the new products that we’ve been implementing into our analytics. Basically just raising a concern that they’re not ready for prime time, and that we can’t necessarily rely on this product to do what we need it to do in the timeframe we need to do it. This is the type of thing where it’s very easy to just throw something over the fence, get it shipped, cross it off the list, and move on to the next thing.


But Rafael here is taking the time to put up his hand and say, “I don’t think that this is ready, and we need to make sure that this is going to be able to serve the purposes, because it could cost us a lot more time if we jump the gun here.” Yeah, just really appreciate this. So much of this stuff is just on-the-fly judgment and these are the examples that I think everybody can adapt to and recognize in their own work all the time. Thanks to everyone on this slide for some awesome work.


Okay. Now some more sobering news. As we all know, Nicole is departing Levels. Her last day is going to be this Monday. And I want to have Miz jump in and say some words. But firstly, I just want to appreciate Nicole, first of all, just for everything she’s done over the months leading up to this point, but specifically, in this last few weeks of just really careful, concerned human handling of a difficult part of Levels’ overall journey. Just really appreciate all the work, all the thought, all the heart you’ve put into it, and cannot wait to see what you do next and can’t wait to stay in touch.

Michael Mizrahi (21:18):

Yeah, not much to add there, Josh. I think you covered all the notes. But Nicole, thanks for being really an exceptional partner for the entire company. I think this role was one that was so hard and so nerve-wracking for us to get right at the time, right when we were about to hit that crazy hiring spurt that we went through, and a lot of the growth that we experienced over the last few quarters over the last year or so.


I’m really thankful we found you. The values alignment, the culture alignment, and bringing someone in that has that people background, the remote understanding was all really exceptional. Yeah, we’re all super, super thankful to get to work with you. So all of us, thanks for being here. Thanks for sticking around and helping us through the last part of this transition, which was particularly tricky. We wish you all the best. We’ll keep in touch, and wish you well.

Josh Clemente (22:01):

Hundred percent. And as Casey said, can’t wait for the book and best of luck with that project as well. All right, another similar theme here. Our Athena team, we’re in proportion with the team changes we’ve made over the last few weeks. We’re reducing some of the support from our Athena partners. Again, just really appreciate the amount of, I mean, just the unbelievable support and partnership that this team represents.


It’s been an absolute game-changer for Levels to be able to have people like the Athena EAs just slot into every type of project that we take on and act as force multipliers. Yeah, thank you for all the work you’ve done. Thank you to the Athena team that are staying on, for continuing to be awesome partners, and we will definitely be working with Athena in the long run, and appreciate each of you on the individual level as well.


All right, and the final culture and kudos piece. This is just a call to help us use and test and share feedback on the app. Part of the validation culture that we’re shifting towards is making sure that we don’t ship something that shouldn’t be shipped, because that introduces a lot of bad data. It’s an unnecessary cycle. The way to do this is for each of us on the internal team to dogfood, the app, basically try what we’re building before it goes out to members. We should be the first line on features, bugs, and suggested improvements, right?


Right now, we have the new lean hybrid app, the single app that is out. It’s out to the internal team, it’s out to 5% of our members, maybe it’s 20% now. Point is, please everyone, get back into the Levels app usage and let’s get back to building anything that we all love. And the best way to do that is to be in there testing it and improving it. We’ll have more on the dogfooding the app.


There’s actually a project that’s going to come together to help everybody do this more effectively, which Cosima’s working on. But, in the meantime, everyone can just jump in right now and try the app experience as it is, and poke holes in it.


Okay, new product ideas. You can see there are a couple more, some new ones on the slide. So people are introducing, they’re using that comms channel, they’re introducing ideas for what we can build that we have not built yet, something that we love. So don’t hold back, get in there, throw ideas on the comms channel.


We will then triage those and put them into the flow, and figure out where they fit in the overall architecture and roadmap. This is super important. Again, we all know what we want, what type product we would love, and this is the way to do that. It’s to raise those ideas and don’t be shy. All right, I’m now going to hand it over to Jhan for a dive into the validated project, tappable charts.

Jhan Cruz (24:48):

Yeah, thank you. So this is one of the first projects that we are using to iterate on this format. The idea is to show how the projects are doing and the results. In the tappable chart items, v1 project, we wanted to provide a shortcut to the entry details screen. So instead of having to scroll down in different data, to a certain item, or doing the same in the MyData tab, you just tap on the items displayed on the… chart.


This works for food, exercise, notes and sleep logs, and it works for anomalies as well. So depending on the item, you would get redirected to the sleep report, the activity details, the food details, with the ingredient breakdown, and… option. Or, if it’s an anomaly, you will be able to add a log directing.


So this is a very small project, but still an interesting case study for all the new things that we are doing on the project side to validate our projects. The hypothesis was that adding the possibility to tap on a chart entry, we will drive more traffic to the entry detail screen. Regarding the validation criteria, we focus on food logs only. So we were looking for positive signal on these two items.


First, for existing users, we wanted them to visit more the entry detail screen after creating a food log. That is 10% from baseline. And for new Levels users, we expected for them to do the same, but 10% more compared to the existing users. Next slide please. So we implemented the project. It took us just a day of engineering workforce, and it was limited to hybrid user only.


And then we got this. So the baseline is 66%. It means 66% of the hybrid users who created a food log then visited the entry details screen within the next two days. As a result, we got a very tiny increase of just three points compared to the baseline, which is not a lot. However, if you remember the validation criteria, we expected to validate this against existing and new users, two different cohorts.


And that’s exactly the first learning of this project. We didn’t have the right tooling available when we collected these metrics. We couldn’t split the data by these two different cohorts. And we also perceived some inconsistencies in the data accuracy. The second learning is that we… very low numbers. As mentioned, we’ve implemented this for hybrid users only. And at that moment, most of our users were classic.


We also learned that despite us not increasing the basics to the entry detail screen, we were able to reduce [inaudible 00:28:20], the way people access it. It means 30% of the people are now using the shortcut. Next slide please.


So for the next steps for this project, the project is done and we’ll have a version two soon. That will happen once we have the right tooling to collect accurate metrics. All this experience was already shared with the measurement product team. In version two, we will let the users know that this feature’s available.


It means that we cannot expect our users to discover a feature by themselves. So we will probably use not just, or animations to let them know that the items on the [inaudible 00:29:17] chart are open. And finally, you can find more information about this project on the star database, Notion. Thank you.

Josh Clemente (29:28):

Awesome. Thank you, Jhan. Yeah, I know we didn’t have great baseline data here, but I appreciate the willingness to trial run the post-validated learnings presentation and forum. So the idea here is that we will have a lot of projects that are going through validation. So we’re going to shape them, build them, ship them, measure the impact, and then that will go into the Monday metrics meeting, which is where we actually look at how these projects move the needle on the important metrics that they support.


And then those validated projects will then roll into this Friday Forum where we can share the learnings in, as it connects to the user journey. So Jhan, thank you for, first of all, shipping this. I personally use this feature. I love this feature and I’m excited to have it back. Maybe I’m in the 66% who are looking at entry details, but I’m confident that we’ll be able to boost this. Yeah, thanks for trial running this. All right, Zach is going to take it from here on Legal function dive.

Zac (30:25):

Hey, there everyone. So it’s been a while since I’ve talked through what it is that Legal is doing within Levels and how I’m ultimately helping the team out in various ways. So I’m just going to spend some time talking through that, and then also describe a little bit of how I think about prioritizing the functions within Legal. So next slide.


Cool. One of the best things about doing legal, especially at a startup, is that I get to work with everyone at the company at least a little bit. But one thing that’s certainly the case is, I’m doing different things across Legal. So you may see me doing one thing over here, but not have any idea how I might be working with, for example, R&D or support.


So I just dropped a few examples here. With engineering, perhaps Legal is supporting by making sure that we have the right agreements in place to actually get data from our members’ devices. So maybe an API agreement, or if we end up gathering data through something like a scraper, there are some nuances there that we’ll definitely want to talk through and make sure that we’re getting right.


On the product side, product certainly knows that my biggest concern is always making sure that what we say and do is something that we’re allowed to say and do. Levels is general wellness. We’re not software as a medical device, but jarringly, what we actually say within the app or say that the app can do, can actually change our status, just as much as adding a new feature. So that’s a thing that we’re constantly talking about, and the product engages with Legal on.


On the marketing and growth side of things, obviously loads of agreements, we have lots of different partners and there are agreements for each one of those. Some of them are very simple, some of them actually get pretty complicated. And then especially as we’re thinking towards growing, one of the things that we’re obviously trying to do is use data better at Levels.


And that sometimes means making sure that the member data that we’ve collected, we’re really getting all of the power out of that data as we can. So making sure that the ways that we use member data is still consistent with our stance as being good data stewards and is consistent with our privacy policy. That’s something that we’re always thinking through.


On the support side, there’s loads here, but most of my work comes down to helping with data deletion requests. And then some of my favorite work is, support does some really hard work for us and tackles some really tricky, unique one-off issues that come up. And every once in a while, I get to be looped in on those and that can be some really interesting work, and often pretty high impact in the context of helping that one person go from having a not very good experience to hopefully a better one where legal isn’t implicated for us.


And then, of course, on the corporate side, there’s legal strategy, there’s employment-related stuff, there’s company equity and cap table management, and then all just of the general keep-the-lights-on stuff. So it’s a lot of stuff and lots of different buckets. So one of the things that I’m always thinking about is how developed should my skillset and our documentation, and our processes be for each of these? And that’s something that we actually have a system for as well. So next slide.


So I use what’s called a legal department maturity model. And basically, it comes down to this. Lawyers really like to be perfect, and in startups, perfect is super bad. If there is a legal function that I am working really hard to make perfect, there’s a really good chance that I’m not using my time well, I’m probably using up other people’s time poorly, and other areas of the company are getting neglected, at least in the legal context.


I have probably 15 buckets or so of stuff that’s Legal. And, so the way I handle this is, I use a zero to five scale, where zero is, I’m not paying attention to this function at all. If something happens here, I don’t know what’s going on. Two, five would be this is totally dialed in, I’m paying attention to this every single day. And there’s a system for the system for the system.


And what I want is for just about every legal function in Levels to be sitting at, between a two and a three. And what that translates into is, we have enough system on the legal side in place to make sure that we can handle it and nothing important falls through the cracks, but no more than that.


So here’s just a snapshot, on this bottom right, of what my legal maturity model table looks like. And you can see that on the right, even for pretty important things, like data privacy and security, contract management, litigation management, we’re still trying hard to not be any more developed than right around a three, which is again, that stage where we have our hands wrapped around it, but we’re not going any further than that.


This is pretty tough, but this really comes down to why inside counsel should be pretty different than outside counsel. Outside counsel are really used to getting everything to that perfect level, but nobody has time for perfection. And even if I did, the moment I made something perfect, obviously we’re a startup, so we would change and we’d have to do that work all over again. So this is one of my most useful tools that sits in the back end and it keeps me focused on being intentionally imperfect. Next slide.


So despite all that, there definitely are still a few things that I consider my big picture, major priorities. Number one on those is just answering your legal questions. It’s so easy for Legal to accidentally be a blocker. So when I see legal questions come in, that typically goes to the top of my list, if possible, to make sure that you have the answers you need to be able to continue to build and not have to redo any work.


Related to that is just helping us find ways through and around legal impediments. We are, I always tell people that Levels is a legal-adjacent business. The FDA is not a regulator, but like I mentioned before, if we make a couple mistakes, they could suddenly become our regulators, on accident. So making sure that the decisions that we make keep us in the proper path. And if there is a legal challenge, finding ways around it so we can still reach those goals. That’s combined with number one as my core priorities.


And really what that comes down to is making sure that we just remain in good standing with all of the major regulators that are in this space, whether they regulate us today or not. And the big players for us are HHS, FDA, and CDRH. Those all are in the health bucket. That’s the organizations on the federal level that manage privacy and medical devices.


But one thing you all may have seen in the news this past few years is the Federal Trade Commission is actually suddenly awake again and is regulating startups more than anyone else. In fact, that’s probably the organization that’s regulating health data far more than HHS, which is what we think of as our health data regulator. That’s HIPAA, but FTC really just cares about, “Hey, are we being honest with our members at all times about what we’re doing with their data?” Next slide.


So to wrap it up, really, at the end of the day, all of that stuff aside and a bunch of buckets aside, Legal in any company, but especially a startup, really just has one true job, that’s to help us achieve our objectives while also doing the right thing. This is, honestly, one of the best parts of working at Levels in particular, because pretty much to a person, and certainly from Josh and Sam, all the way down, we’ve really cultivated a culture of acting in good faith.


And I know you all know this, but that is just not the way it works at every company. It’s nice for me, because candidly, it makes my job a lot easier. I know that everyone at the company is really trying to stay in the good faith zone, and ultimately, that’s my core job, to just keep us there and make sure that what we do not only accomplishes the mission, but is, in fact, the right thing. That’s it.

Josh Clemente (38:34):

That’s it, he says after that segment. Awesome, Zac, I love these function dives. And, in particular, I just really appreciate that model that you brought, the maturity model. I’m going to have to think about that a lot, how that can be used in my own work. That’s awesome.


All right, quick hiring updates. No open roles at the moment. If you or someone you know is interested, check out levels.link/careers. I anticipate some potential updates here on the R&D side, but nothing just yet. Great meeting. Have an awesome time. If you’re taking some time over the fourth, enjoy. Otherwise, back at it next week. Have a great weekend.