2019 | 2020 | ||||||
Price: | 16.87 | EPS | 0 | 0 | |||
Shares Out. (in M): | 111 | P/E | 0 | 0 | |||
Market Cap (in $M): | 1,876 | P/FCF | 0 | 0 | |||
Net Debt (in $M): | -340 | EBIT | 0 | 0 | |||
TEV (in $M): | 1,529 | TEV/EBIT | 0 | 0 |
Sign up for free guest access to view investment idea with a 45 days delay.
Introduction
LVGO has been caught in the recent indiscriminate shorting of IPOs following the WeWork fiasco, and short interest has risen to about 25% of the float despite zero fundamental changes. Misconceptions around its Total Contract Value ("TCV") metric have also contributed to this opportunity. Despite reporting strong 2Q19 earnings in September, the stock sold off dramatically after the company reiterated full-year revenue guidance that implied a deceleration of revenue growth in 2H19 despite record TCV bookings in the quarter. Tech investors misinterpreted the lack of a guide-up to 2019 revenue as higher implied churn in the back half and sold due to a misunderstanding of the TCV metric and how implementations ramp-up and generate revenue.
We value the company using a DCF at $32/share, which translates to a 7x 2021 revenue or 10x 2021 gross profit, for 95% upside. The stock trades at about 6.5x consensus 2020 revenue, which we believe is beatable. As shown below, LVGO trades at dramatic growth-adjusted discount to other fast-growing recurring revenue businesses:
LVGO operates a leading data-driven digital health platform which optimizes patient management of chronic health conditions, mainly focused on Type II diabetes today. The platform consists of a closed-loop continuous feedback system which incorporates connected technology, free, automatically-delivered testing supplies, AI-powered personalized “health nudges”, real-time 1-to-1 coaching and 24x7x365 monitoring designed to guide diabetic enrolled members to avoid danger states and expensive hospital stays. LVGO’s solutions are sold to employer clients as a no-cost benefit to their employees, and 20% of the Fortune 500 are clients.
LVGO’s solution provides individualized care that members love (Net Promoter Score (“NPS”) score of 64), reduces costs for employers (average ROI of 3.7x), while generating gross margins of 70%. All revenue is recurring and annual client retention is 96%. The company has a significant first-mover advantage in a very large and underpenetrated market, and has been growing revenue greater than 100%. Livongo’s sales approach resulted in lifetime value-to-customer acquisition cost (“LTV/CAC”) ratio of 23x in 2018, and high efficiency with sales and marketing spend should yield attractive margins longer-term.
How LVGO makes money and value proposition
LVGO’s platform today focuses on addressing Type II diabetes which affects ~31 million Americans. Diabetes is estimated to have accounted for $327 billion of healthcare costs in 2017. Type II diabetes sufferers have traditionally been forced to manage their chronic illness largely on their own. The average compliant diabetic sees a healthcare provider a few times a year for a 10-minute appointment, and isn't able to access timely individualized expert advice on their glucometer in the context of their particular health condition. Livongo (LVGO) bridges this care gap with a disruptive approach consisting of free, automatically-delivered equipment and supplies, constant monitoring and tailored “health nudges” delivered through connected devices, powered by continuous data flow and an AI engine. Health coaches are available anytime to provide empathetic advice. Members enrolled in Livongo for Diabetes experienced an average 18.4% decrease in the likelihood of having a day with hypoglycemia (blood glucose less than 70 mg/dL) and an average 16.4% decrease in hyperglycemia (blood glucose greater than 180 mg/dL).
LVGO generates revenue by charging a monthly subscription of ~$75 per member per month (“PMPM”) to employers. Self-insured and fully-insured employer clients offer LVGO's solution to its employees at no cost. On average, LVGO's solution generates medical cost savings of $1,900 per member per year, or 3.7x ROI. During the IPO roadshow, LVGO provided case studies of companies like Target that saved much more. Fortune Brands achieved year 1 ROI of 5.0x with $235 PMPM of gross medical savings and a 66% decrease in inpatient hospital spending. Several checks with self-insured clients corroborate a strong ROI and dramatically lower cost CAGRs for enrolled (+3% pa) compared to unenrolled (+15% pa) diabetics. Client satisfaction is further supported by the high 96% client retention rate. Some clients have mentioned the possibility of shifting LVGO's solution from opt-in to a requirement for diabetics, or raising premiums for those that don't enroll.
Not only do LVGO’s clients save money by making the switch, but their employees are much more satisfied with their benefits packages and are likely also more productive at work. Diabetic enrolled members love LVGO’s solution as well. LVGO has a NPS score of 64 (which is unheard of in healthcare), compared to Amazon’s score of 62 and Netflix’s score of 68. This high-quality experience results in member non-job loss-related monthly churn of 0.5%.
Competitive advantage
LVGO benefits from first-mover advantage and a multi-prong go-to-market strategy. In addition to selling direct-to-employer, LVGO is years ahead in aggressively investing in educating consultants (e.g. Mercer, Willis Towers Watson), health insurer (e.g. HCSC, Anthem) and PBM (Express Scripts, Caremark) channel partners on the advantages of LVGO's solution. Our checks confirm that LVGO spent years investing in Mercer's sales force, and is also partnered with 4 of the top 7 insurers and 2 of the largest PBMs. PBMs and health insurers generally promote whoever they have a partnership with and know well. They invested in training their people on LVGO’s product and are unlikely to invest valuable time in understanding a competing solution for a slightly lower price. Moreover, since many of these partners are paid on a revenue share, they have a preference to partner with the company that is most likely to drive the highest enrollment rates at their clients. LVGO’s longer history and larger scale with respect to enrollment data provides it with an advantage in using AI and data science to drive enrollment rates. Despite several start-up competitors in the space (Omada Health, Virta Health, Onduo, Glooko, etc.), LVGO’s pricing has been relatively stable.
In a nascent industry with a number of unproven offerings, LVGO is widely regarded as the gold standard for digital health platforms. It has the greatest number of quality 3rd party actuarial studies and reference clients. The effectiveness of their sales approach can be seen in their client LTV/CAC, which was an astounding 23x in 2018, compared to best-in-class SaaS companies at 4x. Diabetic patients are typically diagnosed with other chronic conditions (e.g. 73.6% of diabetics suffer from hypertension, 87.5% are overweight/obese), and LVGO has already launched solutions for hypertension, weight management, behavioral health and pre-diabetes, which are in-demand by employers and will help drive ARPU growth as well. In members with a starting blood pressure greater than 140/90, LVGO’s hypertension solution demonstrably reduces systolic blood pressure by 10mmHG, associated with a 22% reduction in coronary heart disease and a 41% reduction in stroke. Its first-mover advantage means that it has the most comprehensive platform for chronic diseases, which is increasingly attractive to employers.
TCV metric clarification; consensus estimates are beatable
TCV is defined as contractually committed orders to be invoiced under agreements initially entered into during the period, assuming a 25% enrollment rate and a full term of service, and includes both new agreements entered into with new clients and renewals with existing clients. TCV grew 200% y/y last quarter to a record $74.2m. Given an average contract length of 18 months, this equates to new annual recurring revenue of $49.5m. This revenue should have 70% gross margins and recur for many years. Note that the company spent just $17.7m in sales and marketing in the quarter for these bookings.
The timing of TCV conversion to revenue and the conservative enrollment rate estimate incorporated into TCV is not understood by the market. Unlike typical SaaS subscription bookings that contribute to recurring revenue shortly after being booked, LVGO's TCV wins may not launch and generate revenue until the next year if the client chooses to launch its LVGO offering on Jan 1. This dynamic is misunderstood by tech investors and is precisely why strong bookings in 2Q19 did not result in a guide-up for remaining-year revenue. LVGO further clarified that new contracts on average require an additional 3 months to launch, and reach 12.5% enrollment 3 months after launch and 25% enrollment 9 months after launch. Note that the TCV metric is conservatively estimated on 25% expected enrollment, compared to 34% actual average enrollment and 47% enrollment for optimized clients. The company mentioned on the 2Q19 earnings call that it believes enrollment rates of over 60% are achievable.
LVGO expects the seasonal effect of TCV wins being back-half weighted to continue in 2019. Based on a simple waterfall translating TCV to revenue, company guidance and consensus revenue numbers are beatable for 2H19 and 1Q20. We estimate 15% upside to 2H19 consensus revenue and 6% upside to 1Q20 consensus revenue. Full year 2020 revenue will partially depend on bookings in 2H19 and 1H20 but also appear easily beatable using relatively conservative assumptions for bookings growth. We currently model 2020 revenue about 7% higher than consensus but believe that there is a higher probability that our numbers get revised upward rather than downward. For example, we only model TCV growth of 24% in 2020 which is a very large deceleration vs the 200% y/y growth the company posted last quarter. We also conservatively expect TCV to grow at almost half the growth rate of sales and marketing expense in 2020. Despite efficiency of sales and marketing or LTV/CAC improving in 2019 vs 2018, we assume that it comes down going forward.
TAM and revenue projections
LVGO has a long run-way for growth. The estimated TAM for the diabetes solution is 14.3 million diabetics or $12B of potential revenue. This diabetes TAM excludes diabetics managed by United Healthcare (currently not partnered with LVGO) and Humana (partnered with Omada Health). LVGO currently has 193K diabetic enrolled members, or 1% of its addressable opportunity in diabetes. Hypertension represents an additional $14B opportunity. We expect LVGO to grow revenue at a 74% CAGR for the next 5 years driven by member enrollment.
Risks and catalysts
While there are admittedly some questions around long-term competitive dynamics given the number of other start-ups in digital health, LVGO will experience tremendous growth for the foreseeable future. LVGO's outrageous discount to comparable high-growth companies is unwarranted. We believe LVGO will continue to beat revenue expectations and particularly 1Q20 and FY2020 expectations as TCV wins from 2019 start to ramp-up into recurring revenue.
Consensus revenue expectations outperformance
show sort by |
# | AUTHOR DATE SUBJECT |
---|---|
17 | |
Ok 8.3x, but it'll prob be 10x by its anniversary. Nice idea. | |
16 | |
Really well done here. Understand wanting to take the quick gains. Anything that keeps you from continuing to own this? Feels like it's working. TAM is very large. | |
13 | |
Light62, thanks a lot for arguing the opposite side. How does the Blue Cross Blue Shield Federal Employee Program affect your suspicions about the product's efficacy? It would seem that there would be a healthy amount of rigor applied before granting 5.3MM lives access to a new diabetes management platform that's essentially powered by Livongo. | |
11 | |
thanks for the feedback from the blogger. a lot of this is "he said / she said" stuff which is difficult to argue against, but it is always good to understand the other side. it seems like the short thesis here is that mgmt is promotional. i agree that the company is promotional. they are a silicon valley based company with silicon valley VC's behind it. most young saas companies are promotional. Many SAAS companies pitch their sales based on ROI that is usually higher than what the end customer actually achieves. i talk to many customers of saas and i have to say that the percentage of companies that are satisfied with LVGO is much higher than the average SAAS company out there. difference here is that LVGO is in the wellness space and so is covered by healthcare analysts who care a lot more about the claims being perfectly accurate than tech investors. from talking to PB desks who have a more realtime estimate for short interest, the short interest has now climbed to 45% of the float. it is up significantly since i posted the idea despite the stock rising 19% since then. I know you covered but I think it is quite risky to short this stock now in front of management presenting at several upcoming conferences and in front of a quarter where TCV will likely be good again given the large federal contract that was announced. afterall, management is promotional. I usually stay away from promotional management, but i think that in this case, it is actually a potential catalyst given the high short interest and no obvious short term issue.
| |
9 | |
LVGO announced on Monday it was awarded a contract with the Blue Cross and Blue Shield Federal Employee Program ("FEP"), which is part of the Federal Employees Health Benefits Program ("FEHBP") and covers ~5.3m federal employees, retirees and their families out of the ~8 million people covered under FEHBP. This is a 2-year agreement which will launch the Livongo for Diabetes program on Jan 1 2020, and is expected to add ~25K members in 2020 and grow to ~45K members in 2021.
This expected enrollment is higher than the original 20-30K first estimated by Livongo in its S-1 and the 2Q19 earnings release. We think this development highlights the general conservatism of management. LVGO expects the contract will account for $20-25m of revenue in 2020 and $30-$35m of revenue in 2021, and helps de-risk revenue expectations. Sell-side has not revised their numbers as of yet.
The stock had a positive initial reaction, closing +18% after the announcement. The stock has since sold off and is now only +8% above pre-announcement. Short interest has actually increased to 35% of float. | |
8 | |
We asked the company about the questions posed in the link you shared. Given what you said, we were surprised with how quickly they responded to our questions and their directness. With respect to your blogger himself, they noted that the blogger had approached them a number of years ago and solicited compensation to endorse LVGO on his site. They obviously declined, and since the post you referenced, he has been unwilling to respond to their requests to engage in discussion. They claim he has been paid in the past by one of their competitors and is running a pay-to-play scheme. We noticed that he is very promotional of the Validation Institute on his site. It turns out he is listed as a certified consultant/broker for the VI (https://validationinstitute.com/certified-professionals/), which offers a certificate program starting at $1195 (https://validationinstitute.com/certification/). In a prior post, he trumpeted the benefits of the CORA Pro certificate course while simultaneously cherry-picking a couple points to disparage LVGO (https://dismgmt.wordpress.com/2019/05/23/questions-to-ask-livongo-before-signing-with-them/). We find his motives at best ambiguous. Needless to say, numerous 3rd party consultants have raised concerns to LVGO about this blogger and his claims.
While we understand that the clients may not be performing tightly rigorous analyses and may also be somewhat biased because they are justifying a decision they recently made, the vast majority are saying that they are seeing their total healthcare costs across their diabetic population come down even after incorporating LVGO’s cost, comparing pre-launch to post-launch costs. It’s possible LVGO may not have been the only change they made or that their diabetic population may have somehow structurally changed post-launch or other confounding factors have materialized, but the general feedback has been positive and most are willing to be reference clients and seem happy with their decision to go with LVGO.
We think it is arguable whether the statement “Livongo for Diabetes program delivered an $88 per member monthly reduction” is mis-leading in suggesting causation and, frankly, is a minor semantics issue in light of the savings that clients are reporting. LVGO has only used this language in a handful of press releases, and we haven’t seen any more-aggressive causal language from the company.
In addition to the clients we mentioned in the prior post, we just spoke to a client that has achieved >70% enrollment and has been a client since 2017. He believes that the ROI on LVGO is 1.8x based on comparing the same cohort pre-launch vs. post-launch. His savings came from less urgent care, less episodes of diabetic shock and decreased episodic care. He was adamant that LVGO works. We acknowledge the selection bias here, but given the high enrollment rate the <30% unenrolled would have to be pretty bad for overall ROI to be negative.
Curious if your negative view is based on any customer checks or if it is mainly based on the blogger you cited?
We acknowledge that from a design standpoint, a better way to perform a conclusive, controlled study to prove efficacy would be to have volunteers randomly assigned to LVGO test and non-LVGO control groups. This approach doesn't seem very realistic in the context of live clients (imagine being an employee not chosen for the test group). LVGO could have performed a study outside their client base, but it's not clear that the time required and the extra cost (volunteers would have to be provided the solution for free for the length of the study) would have produced much of an incremental benefit.
Regarding the questions around reduced inpatient admissions, the analysis in the blog (and w/r/t the chart you provided) compares admissions related to diabetes to total commercially insured people/the general population. This seems apples-to-oranges to us. 90% of the population does not have diabetes, and presumably there were no non-diabetics included in LVGO’s studies. Hospitalization stats would look much different for the diabetic population. To be clear, we are still waiting on a more direct answer from the company but as your blogger states, “in all fairness to LVGO, they don’t promote the inpatient admissions result anymore.”
The premise of blogger’s Question 3, “if I’m seeking to reverse diabetes in my population,” makes us wonder if your blogger even understands what LVGO does. LVGO helps diabetics manage their chronic conditions and has never focused on reversing diabetes. Consequently LVGO hasn’t focused on measuring weight loss or reduced insulin use in the context of diabetes (improved compliance with LVGO actually results in an initial increase in the use of test strips and insulin). In fact, the question reads almost like an endorsement of Virta Health, whose strategy is to actually reverse diabetes by encouraging dramatic lifestyle changes and nutritional ketosis – perhaps the competitor endorsed by your blogger referenced by the company?
LVGO’s study very clearly addresses the obvious concerns around selection bias and other unobserved confounders through the use of propensity score matching: “the comparison group of people with diabetes who were not enrolled in the Livongo program was selected using 1:1 propensity score matching on covariates including age, gender, Charlson co-morbidity index and pre-program medical spending with the Livongo program population to reduce the bias due to confounding that could be found in an estimate of the treatment effect.” In simpler terms, members and non-members were matched on a 1:1 basis based on the characteristics described in a way that makes them “look” similar and consequently provides a valid comparator group in an observational study design. Our conversations with a highly experienced statistician from the FDA confirm that propensity score matching is a standard and appropriate technique for the study. In addition, instrumental variable analysis was used to control for unobserved confounders.
Regarding the other questions posed by your blogger in Question 5, they seem like red herrings and more fear-mongering to entice readers to cling to the “truths” graciously revealed by him. With respect to the JAMA Internal Medicine study cited that suggests there is "no statistical difference between patients who do not self-monitor their blood glucose multiple times per day and those who do self-monitor their blood glucose multiple times per day in glycemic control, nor evidence of effects on health-related quality of life, patient satisfaction, or decreased number of hypoglycemic episodes," there is a difference between “structured checking” and just self-monitoring multiple times per day. The ADA has guidelines on how people with diabetes should use structured checking as a way to get insights into their glucose control. There is a lot of supporting literature to show the effect of structured checking on glucose control (https://www.ncbi.nlm.nih.gov/pubmed/22464874), which your blogger failed to point out. LVGO’s messaging through digital and human coaching is centered around educating members on the benefit and insights from structured checking. The company agrees that random checks provide limited insights into glucose control. | |
7 | |
I'd be interested to know what the customers are doing to evaluate whether LVGO works or not and whether it's a rigorous analysis or just a simple - LVGO using employees are outperforming non-LVGO using employees (which is going to lead to a massive selection bias).
It sounds like you have a good rapport with the company - I think a great way to get more clarity would be to ask them the questions posed in the link I provided in my original post. If they provide substantive answers please let me know because, in the past, they've been unwilling to engage on these questions (which strikes me as a big red flag) - they are fully aware of the concerns and the post.
That said, I agree that there might be enough momentum here that these concerns won't matter (at least for awhile) but at this point I think the concerns about efficacy are becoming more widely known and if they break out I think it puts the company on the defensive and throws a wrench into DCF's making bet #2 harder to pull off. Personally, I doubt the sell-off has been primarily due to the disappointing guidance (I've seen the analysis you provided on bookings growth vs. revenue timing several times - it doesn't seem to me like the company has failed to get the message out there). This thing's down from the $40's post IPO and was falling even before the earnings release. | |
6 | |
its a risk but also an opportunity. express scripts is likely LVGO's most important channel partner today. however, note that Cigna actually invested in LVGO before they completed their acquisition of express scripts. also, there is generally staying power with these partnerships. if the relationship is ended and customers decide to keep LVGO after the initial term of the customer contract expires, it is our understanding that cigna/express scripts do not get their revenue share. from our checks, omada has a good product for pre-diebetes and LVGO has a better product for diebetes, so it is possible that both actually coexist within cigna/express scripts. | |
5 | |
Gross margins are 70%. The strip cost is in cogs. Ebit is negative cause of OPEX similar to many fast growing SAAS. Given the LTV to CAC, you want them to spend a lot on sales and marketing. Part of the opportunity here is that tech investors dumped it after the lack of guide up for this year and healthcare investors are not used to valuing very high growth recurring revenue businesses that are losing money near term. Many of the companies in the chart above are also losing money on the ebit line and some even have lower gross margins than lvgo. | |
3 | |
We have talked to 8 lvgo customers. Many of them were very large self insurers. They all basically said that the product worked and is producing savings. They also said that their employees are much more satisfied with it than what they were previously using to manage diebetes. Simple things like not having to constantly buy new strips vs just having them automatically sent to you made a big difference. That being said, we do acknowledge that the company may be being overly promotional (most growth companies at this stage are). We also talked to someone from a large consulting firm that set up the partnership with Lvgo. He was pretty optimistic about the market’s growth but was admittedly a bit more skeptical than the customers we spoke to about effectiveness. He said some customers had positive savings and others didn’t see it. He didn’t tell us that the product didn’t work however. The most negative check came from a large insurer that was the most skeptical. He frankly seemed skeptical of any digital health company and is arguably a potential competitor to lvgo longer term. However, he also said that there are many product lines that they offer that have questionable efficacy but still are $1b revenue companies. While he questioned the efficacy, he thought that there was a lot of momentum for lvgo in the market place. We also talked to a former executive of lvgo who was the most bullish but she was arguably biased. We put the most weight with the checks with the actual customers and users of the product. Those were clearly positive, but after your message, we will do more. We also take comfort from the numbers - 96% renewal rate when the avg contract is 18mos, high NPS scores, very high bookings and revenue growth relative to sales and marketing spend, etc. maybe digital health is just different, but if the product was bad, I don’t think the numbers would look like that. There is definitely high skepticism about digital health. that makes lvgo’s sales and marketing efficiency even more impressive.
| |
2 | |
I'm short this and have been for awhile. It's not as clear to me that this product works as your write-up (and the company's promotional material) make it seem - see https://dismgmt.wordpress.com/2019/08/26/are-livongos-outcomes-real/ for more details. Frankly, I think the product doesn't seem to work (at least not nearly as well as advertised)*. That's not to say this can't be a good investment; however, I think the bet is much less attractive than the one detailed. I think you're making two bets: 1) Patients like the product even though it doesn't work and HR departments/insurance co's are so desperate to show they're trying to contain H.C. costs that the business does well regardless of its ability to actually lower H.C. costs. Keep in mind how long this relationship would need to sustain itself for a DCF to justify the current valuation (a question for Mason - how much of the DCF value is post 2030?). 2) Investors bid this up on the basis of (what I think are) non-applicable SaaS-type quality criteria (LTV/CAC) vs. a SaaS valuation framework (P/S vs. growth) and a greater fool bails you out. Those may be good bets and are things I'm monitoring for my short. That said, I think this company is going to be on the defensive given the serious questions about efficacy.
*Just one thing that jumps out to me as hard to fathom: How can a diabetes coaching app drive a 66% reduction in in-patient hospital spending? What % of in-patient admissions are even tangentially related to diabetes? Which of the following are:
I’m pretty sure I can come up with 33% of spending just in the top 20 conditions (47.7% of total inpatient spending) that aren’t going to be solved by improvements in someone’s glycemic control. | |
1 | |
thanks for the write up. how do you think about the esrx relationship post cigna merger? cigna is an investor in omada. |
Are you sure you want to close this position LIVONGO HEALTH INC?
By closing position, I’m notifying VIC Members that at today’s market price, I no longer am recommending this position.
Are you sure you want to Flag this idea LIVONGO HEALTH INC for removal?
Flagging an idea indicates that the idea does not meet the standards of the club and you believe it should be removed from the site. Once a threshold has been reached the idea will be removed.
You currently do not have message posting privilages, there are 1 way you can get the privilage.
What is wrong with message, "".