Insurance is an amazing business, where ‘lifetime value’ (LTV) is often measured in actual lifetime. People typically buy their first insurance policy in their 20s, pay severalfold more for their policies by midlife, and continue to pay premiums well into their 80s.

That makes for an extraordinary LTV, which is why it can be so lucrative to invest in customer acquisition today, in return for that stream of gross profits in the years and decades to come. At Lemonade we see this captured succinctly in the ratio of lifetime value to customer acquisition cost, or LTV/CAC. Our LTV/CAC stands at >3, meaning that, on average, we get back over $3 in gross profit for every $1 we spend on CAC.

Taking account of the time value of money, our estimated internal rate of return (IRR) on our CAC expenditure clocks in at ~50%, a compelling ROI by any measure. But there’s a downside to the elongated timelines of insurance: They create a short-term cash crunch.

It takes Lemonade ~24 months to earn back our initial CAC, and in the meantime that money is spoken for. A dollar spent today to acquire a customer, in other words, won’t be available for the acquisition of the next customer for two years.

This constraint—known as the cash flow gap—is a real impediment to growth. For even as we see opportunities to deploy capital at a healthy IRR, so long as our capital is tied up, many of those potential streams of gross profit will go untapped. Having to forego profitable business because of the cash flow gap, in other words, is a drag on our terminal value.

In the era of frothy stock valuations, many tech companies (us included) bridged their cash flow gap by selling equity. That made sense for a moment in time, but that moment has passed, and in any event equity is not a scalable source for working capital.

Insurance agents

In traditional insurance, the cash flow gap is typically bridged through independent agents. Countless people turn to their insurance agent for any number of reasons, and most—particularly those leery of self-serve technology—appreciate the value their agent adds. 

From the insurance company’s perspective, too, agents add value in multiple ways, though financing the insurer’s CAC remains a big part of the attraction. The agents bankroll the upfront CAC, and in return they partake of the LTV by means of a commission (usually 10-20% of the premiums). The commissions flow for the life of the customer, and they attach to all their subsequent purchases too. 

We see a place for agents in Lemonade’s business too—primarily to extend our reach to customers whom we cannot onboard through digital channels—but the overwhelming majority of our business will continue to be direct to consumers. 

Why? For one, the agent-mediated business replaces the magical Lemonade experience with the agents’ interface, commoditizing our brand and watering down the data we collect. For another, the agents siphon off as much as half of the gross profit of the customer for the life of the customer, greatly reducing their LTV. 

For business we cannot reach through other means, agents make sense; they are accretive when they are not cannibalizing. But for customers we can reach on our own we’re better off going direct… if only we could find another way to span that darned cash flow gap.

Synthetic agents

The positive aspects of working with agents got us thinking: What if we could fashion a capital structure that gave us the cash flow benefits of agents, without their downsides? Is there a way to own the customer fully, yet have someone else finance our CAC? Can we collapse the payback period and unlock capital-light growth, without ceding half of our gross profit for the lifetime of the customer?

We believe we’ve found a way to do just that by creating what we call Synthetic Agents. We conjured up Synthetic Agents together with General Catalyst (GC), one of the foremost technology investors, and an early backer of Lemonade. GC has a dedicated Customer Value strategy, which for the last several years has helped companies across industries close their own cash flow gaps. They’ve proven an ideal launch partner for our Synthetic Agents program.

Let’s start with the mechanics of the program:

  • Our Synthetic Agents (GC) will finance up to 80% of our CAC spend in any given month (Synthetic Agents’ Spend), and will receive a ‘commission’ of up to 16% from, and only from, the premiums paid by the corresponding cohort (that is, all the people who join Lemonade that month).
  • After approximately 2–3 years, the stream of synthetic ‘commissions’ from any one cohort will tally to the Synthetic Agents’ Spend on that cohort plus a fixed return (16% IRR). Once that capped return is reached, ‘commissions’ from that cohort terminate, and 100% of their premiums, thereon out, accrue exclusively to the benefit of Lemonade.
  • Conversely: If, for any reason, the stream of ‘commissions’ peters out before the Synthetic Agents recover their investment, the resultant loss will be borne by them alone, and they will have no recourse against adjacent cohorts, or Lemonade more broadly. 

Synthetic Agents have two important things in common with the garden-variety kind:

  • Synthetic Agents collapse Lemonade’s CAC payback time, enabling us to unlock rapid growth without tapping our cash reserves.
  • Synthetic Agents are doing this specifically, and solely, for the exposure to ‘commissions,’ or Lemonade’s customer lifetime value. GC is not receiving equity or warrants, nor are they placing encumbrances, liens, or restrictive covenants on Lemonade. Like their flesh-and-blood counterparts, they have a contractual right to be paid the equivalent of a commission from the cohorts they brought to Lemonade—and that’s the only place their returns are coming from.

With those compelling upsides, the similarity to the traditional agency model ends. For unlike the standard terms of an agency deal, with Synthetic Agents:

  • The customer’s relationship is directly, entirely, and exclusively with Lemonade. The customer sees our ads, clicks through to our app or website, and has the full Lemonade experience—unmediated by anyone else, at any point.
  • The Synthetic Agent is not a partner of ours for the life of the customer. Their ‘commission’ ceases after about 2–3 years, and the entirety of the customers’ LTV accrues to Lemonade thereon out. Synthetic Agents are our partner only in financing our CAC, not in ‘owning’ our customers.

We think Synthetic Agents are something of a game changer for Lemonade. We can henceforth accelerate growth without drawing down our capital reserves or selling more equity; and we can generate a significantly larger business, far sooner, with more cash in the bank, and with a materially higher return on capital. That’s a lot. In fact, thanks to our Synthetic Agents, we estimate the IRR on our CAC spend will jump to ~90%!

The program is also great for our Synthetic Agents, and we wouldn’t want it any other way. Unlike the conventional insurance agent, Synthetic Agents incur no overheads, interact with no customers, and need no licenses—it’s a purely financial play. 

Insurance is a high-retention business, and Lemonade’s retention triangles are remarkably stable and predictive. These form the basis for the only underwriting Synthetic Agents engage in: getting comfortable that the stream of premiums they’re financing will be around long enough to return their investment, and then some. Investing in Lemonade through the Synthetic Agents constellation is not risk free, but the risk is low and amply rewarded. 

The win-win nature of the Synthetic Agents program is one of its strengths. It means the program can scale and endure.

While the initial agreement makes $150m of committed CAC financing available to us over the next 18 months, the parties intend for this framework to adjust, renew, and upsize in concert with our expanding business. Synthetic Agents, in short, provide the kind of robust capital structure upon which our ambitious future growth plans can take root.

Change agents

To our regular listeners, talk of “ambitious future growth plans” may sound slightly discordant given something else we’ve been broadcasting of late: we’re proactively slowing growth. Let’s unpack that.

Our recent communications around moderating growth implicated two culprits. One was the elevated rate of inflation and its impact on profitability. In our Q1 ‘22 Shareholder Letter we explained that “claims are instantly adjusted for inflation, while premium rates take months to adjust,” and that in response “we have filed 100 applications for rate changes over the past year.” 

Later, in Q4 ‘22’s Letter, we noted that notwithstanding our extensive rate filings, “rate approvals—in some states and products more than others—have not kept pace with inflationary pressures… Our determination to exclude such areas from our growth plans means that for now we expect overall annual IFP growth of approximately 11-12%, though we’ll look to catch up to our multi-year target of 20-25% as our rates come online.”

Synthetic Agents don’t solve the inflation-induced mismatch of rates and risk. Happily, though, we are at last seeing many of our rate filings being approved, implemented, and beginning to earn in, and our best guesstimate is that we’re a few quarters away from loss ratios with which we will be comfortable picking up our growth rates.

Which brings us to our second reason for slowing growth: elevated cost of capital. In our Q2 ‘22 Letter, we noted that: “we’ve seen our cost of capital increase by about one order of magnitude,” and that “with the goal of ensuring we are never forced to raise capital…we have moderated our growth spend.” We added that “we can effectively deploy far more capital than this moderated plan will in fact see us deploy—just not at today’s cost of capital.”

The refrain “opportunities abound, the cash to pursue them does not” became a familiar one. In our Q3 22 Letter we paraphrased it thus: “…more accelerated growth would, we continue to believe, maximize the net present value of Lemonade. Nevertheless, given today’s high cost of capital, we are intentionally decelerating towards our ‘optimal cash burn’ velocity, which we estimate in the 20-25% annual growth range.”

Synthetic Agents may not cure inflation, but they do wonders for optimal cash burn. During our November ‘22 Investor Day, we shared why we believe a compound annual growth rate (CAGR) in the 20% range is optimal for reaching profitability with extant cash, and why, if inflation causes elevated loss ratios to persist, slowing to the 12% range would be better (as, indeed, happened).

A couple of graphics help make the point:

The scenarios aired at Investor Day remain available and viable. But by enabling growth that doesn’t deplete cash, we believe Synthetic Agents meaningfully expand our range of options and elevate their outcomes.

One way to think of Synthetic Agents is as time-shifters, funneling customers’ future gross profits backwards in time, so customers effectively finance their own CAC before we’ve even met them. Such time travel not only breaks the tenet “growth-depletes-cash,” it actually reverses it.

The renowned British statistician, George Edward Pelham Box, cautioned that “all models are wrong but some are useful,” and ours are no exception. Furthermore, while the cash flow gap may be the “longest pole in the tent,” it doesn’t stand alone, and in reality our growth rates and cash reserves will be shaped by a range of factors beyond the reach of Synthetic Agents.

With those disclaimers duly delivered, let me share that our models now predict that—thanks to Synthetic Agents—our ‘Minimum Unrestricted Cash’ would double to >$200m under most scenarios.

This holds true when modeling not only for the 12% or 20% CAGR we showed in November, but for growth in the 30s and 40s too. And it’s not only modeled cash that remains pretty constant across growth levels; so too would the timeframe for breakeven. 

What does change in the models? On a 5-year horizon, those faster growth rates yield severalfold more In Force Premium and Net Income at the end of the period. Therein lies the absolutely transformational potential of the Synthetic Agents.

The Synthetic Agents go live July 1, and should start doing their cash-maximizing thing right away. For the remainder of 2023, however, we plan to continue to moderate our growth, allowing a few more quarters for our new rates to more fully register on our loss ratio. 

That done, we will look to ramp up growth, and we anticipate elevated growth rates in the years to come. That’s how Synthetic Agents enable our “ambitious future growth plans.”

Strategic context for Synthetic Agents

Our game plan for the coming years stands on three legs:

  1. Expense Ratio: AI leverages our digital infrastructure to automate ever more of our business processes, driving down our expense load even as our quality of service (precision, consistency, speed) climbs.
  1. Loss Ratio: Machine learning harnesses our data advantage (quantity, quality, and usability), making us ever better at matching risk to rate. 
  1. Growth: Synthetic Agents and quota share reinsurance minimize our capital drag, streamlining us for efficient growth.

These three are mutually reinforcing. Our digital infrastructure not only enables automation, lowering expense ratio, it also generates the training data for our machine learning, thereby lowering loss ratio.

The resultant improvements in unit economics free up margin for our Synthetic Agents and quota share partners, alleviating our capital burden and unlocking lean growth; growth swells the denominator of our expense ratio—even as automation keeps the numerator relatively constant—driving down our expense ratio and driving up our net profit. Thus, with every turn of the flywheel, our business grows in size, efficiency, precision, and profitability.

Lemonade was built for scale: We expect growth to be the gift that keeps on giving, and for the power of our model to manifest as increasing profitability as we grow.

In their elemental state, though, the capital burdens of insurance—both regulatory capital and working capital—would weigh us down, slowing growth, idling cash, and delaying profitability. That’s why it’s so significant that we have partners with symbiotic business models to ours. Our reinsurance partners relieve our regulatory capital burden through our quota share program, and General Catalyst relieves our working capital burden through our Synthetic Agents program. 

Growth is key to Lemonade fulfilling its potential, and thankfully we’ve plenty of headroom to grow. We could 10x our business and State Farm would still be 10x our size; and even if we 100x Lemonade, AXA would still be 100% bigger. 

That perspective places the Synthetic Agents program in its broader and proper context. We believe Synthetic Agents will dramatically accelerate our cash flow cycles, and those turbocharged revolutions are the key to getting to scale sooner, and with plenty of gas in the tank. That, in short, is what we plan to do.


This post contains forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995.

All statements other than statements of historical fact contained in this blog post, including without limitation statements regarding the anticipated benefits of the Synthetic Agents program and expectations regarding its impact on our capital structure, the expected future results of operations and financial position, our ability to effectively manage the growth of our business, our ability to compete effectively in our industry, and the plans and objectives of management for future operations and capital expenditures are forward-looking statements. These statements involve known and unknown risks, uncertainties and other important factors that may cause our actual results, performance or achievements to be materially different from any future results, performance or achievements expressed or implied by the forward-looking statements.

These statements are neither promises nor guarantees, but involve known and unknown risks, uncertainties and other important factors that may cause our actual results, performance or achievements expressed or implied to be materially different from any future results, performance or achievements expressed or implied by the forward-looking statements, including, but not limited to the following: our history of losses and the fact that we may not achieve or maintain profitability in the future; our ability to retain and expand our customer base; risks related to the “Lemonade” brand; risks related to denial of claims or our failure to accurately and timely pay claims; our ability to attain greater value from each user; risks related to the novelty of our business model; risks related to our Giveback; risks related to our limited operating history and our ability to evaluate our current business performance, implement our business model, and our future prospects; our ability to manage our growth effectively; the intense competition in the segments of the insurance industry in which we operate; risks related to reinsurance, including the availability of reinsurance at current levels and prices and counterparty risk; our ability to maintain our risk-based capital at the required levels; our ability to expand our product offerings; risks related to the operation and confidentiality of our proprietary artificial intelligence algorithms and proprietary technology; new legislation or legal requirements which may affect how we communicate with our customers; risks related to our reliance on artificial intelligence, telematics, mobile technology and our digital platforms to collect data that we evaluate in pricing and underwriting our insurance policies, managing claims and customer support, and improving business processes, and any legal or regulatory requirements that prohibit or restrict our ability to collect or use this data; our reliance on search engines, social media platforms, digital app stores, content-based online advertising and other online sources to attract consumers to our website and our online app; our ability to raise additional capital to grow our business; risks related to security incidents or real or perceived errors, failures or bugs in our systems, website or app; risks related to examinations by our primary state insurance regulators and examinations or investigations by other states in which we are licensed to operate; risks related to our collection and use of customer information and other data, and our ability to protect such information and comply with data privacy and security laws and regulations; our ability to underwrite risks accurately and charge competitive yet profitable rates to our customers; risks related to our ability to generate revenues from new products; risks related to our expansion within the United States and any future international expansion strategy; risks related to combining the businesses of Lemonade and Metromile and the potential failure to realize the anticipated benefits of the mergers; risks related to the historically cyclical nature of the insurance business; risks related to extensive insurance industry regulations; risks related to severe weather events and other catastrophes, including the effects of climate change and global pandemics which are inherently unpredictable; climate risks, including risks associated with disruptions caused by the transition to a low-carbon economy; risks related to increasing scrutiny, actions and changing expectations from investors, clients, regulators and our employees with respect to environmental, social and governance matters; risks related to fluctuations in our results of operations on a quarterly and annual basis; risks related to utilizing customer and third party data for pricing and underwriting our insurance policies; risks related to limitations in the analytical models used to assess and predict our exposure to catastrophe losses; risks related to potential losses that could be greater than our loss and loss adjustment expense reserves; risks related to the minimum capital and surplus requirements that our insurance subsidiaries are required to have; risks related to assessments and other surcharges from state guaranty funds, and mandatory state insurance facilities; risks related to our status and obligations as a public benefit corporation; risks related to significant shareholders and their ability to influence the outcome of important transactions, including a change in control; and risks related to our operations in Israel and the current political, economic, and military environment.

These and other important factors described under the caption “Risk Factors” in our Annual Report on Form 10-K for the fiscal year ended December 31, 2022 filed on March 3, 2023 and in our other and subsequent filings with the SEC, could cause actual results to differ materially from those indicated by the forward-looking statements made in this blog post. Any such forward-looking statements represent management’s beliefs as of the date of this blog post. While we may elect to update such forward-looking statements at some point in the future, we disclaim any obligation to do so, even if subsequent events cause our views to change.

categories: #news