![\"\"](\"https:\/\/www.lemonade.com\/blog\/wp-content\/uploads\/2022\/08\/NASA.jpg\")
LTV1 through LTV3 were humdrum in comparison. For instance, they allowed us to estimate claim frequency, but not claim severity. We were able to differentiate high-risk users, but overall the models weren\u2019t as granular or accurate as we\u2019d like. <\/p>\n\n\n\n
In June 2021 we were able to take a quantum leap forward with LTV4, which leveraged AI models that were credible enough to be allowed into production. That iteration of our LTV model had the benefit of data culled from the almost four years we\u2019d been in business at that point.<\/p><\/blockquote>\n\n\n\n
Machine-learning modules thrive on data, and it\u2019s a steady supply of new information that keeps them sharp and worth listening to. We\u2019ve had a surplus of data over the past 12 months\u2014over the past year alone, we\u2019ve handled twice as many claims as we did in those first four years of business combined. And during those same 12 months, we pushed 12,000 software builds into production. <\/p>\n\n\n\n
LTV6’s lessons, put into practice<\/h3>\n\n\n\n
Once they reach statistical significance, each product, each market, and each campaign will be scored using LTV6. Lemonade Car\u2019s forthcoming application of LTV6, for instance, will greatly benefit from the wealth of data we\u2019ve inherited from our acquisition of Metromile<\/a>. <\/p>\n\n\n\n
After adding the customer acquisition cost (CAC) into the formula, the resultant LTV\/CAC, together with the CAC Payback Period, determines which product, market, and campaign receives the incremental dollar spend.<\/p>\n\n\n\n
This flexibility equals strategic optionality<\/em>. It means we can adapt quickly, shifting resources from one product or market to another, and from one marketing campaign or channel to the next. <\/p><\/blockquote>\n\n\n\n
What makes LTV6 special, and worth getting excited about? Well, first off, it marks a genuine and dramatic improvement over LTV5. It comprises machine learning and deep learning models<\/a> using almost a million parameters, incorporates state-of-the-art CAT (natural catastrophes) models, and it was trained on Lemonade’s ballooning, proprietary, and highly textured data. <\/p>\n\n\n\n
One fairly dramatic insight from LTV6 was the extent to which certain homeowners insurance business in California was riskier than originally expected. LTV6 flagged that many policies in the state\u2014which had been scored favorably<\/em> by LTV5, and indeed previously appeared to have compelling marketing efficiency\u2014would actually prove loss-making over time.\u00a0<\/p>\n\n\n\n
What each new iteration of our LTV model gives us is a greater degree of magnification and granular perception. So while LTV5 operated at a certain level of magnification, LTV6 is able to zoom in, subdivide an area, and spotlight good pockets and bad pockets in terms of risk. <\/p>\n\n\n\n
So suddenly, LTV was revealing to us that within the mix of profitable pockets within California, it was really made up of better and worse. There’s white blood cells and red blood cells\u2014it’s not just one monolithic blob of blood. <\/p>\n\n\n\n
Again, countless parameters went into LTV6, but in this particular case, with California, it was arguably a greater granularity around catastrophic event modeling\u2014particularly fire events\u2014that helped sharpen our predictive model.<\/p>\n\n\n\n
Another illuminating LTV6 takeaway is that our overall Pet business promises to be more profitable than LTV5 had assessed\u2014welcome news, for sure. <\/p>\n\n\n\n
Having this real-time data and predictive modeling\u2014and being a multi-product, multi-market, multi-channel business\u2014means that we can pivot and shift priorities quite quickly. So in Q2, we dramatically slowed our California homeowners business in several locales, while accelerating our Pet business across most states. <\/p>\n\n\n\n
The systems we have in place both capture the data needed to train our neural networks, and are built to act on their outputs in real time. Each new policyholder brings fresh data into the model, making its predictions better and sharper. We’re not aware of any other insurer who has these capabilities. And it\u2019s not something that an old-fashioned company could simply adopt and adapt; these tools and techniques are difficult to graft onto a company that wasn’t built with them as a core design principle. <\/p>\n\n\n\n
A less distorted signal<\/h3>\n\n\n\n
Another point that\u2019s worth digging into: For a fast-growing and developing business like Lemonade, lagging indicators and leading indicators often commend opposing actions, and the leading<\/em> indicators are almost always right. <\/p>\n\n\n\n
Consider loss ratios. One of the component models of LTV6 predicts the lifetime loss ratio of each customer. As the term suggests, this anticipates how much each customer will pay in premiums, and how much they will cost in claims, dividing the one by the other.<\/p>\n\n\n\n
As you’d expect from a probabilistic model, the outputs can be hit-or-miss for any individual customer, but when you total them all up they become highly predictive. <\/p>\n\n\n\n
It is this predicted loss ratio, rather than the \u2018rear-view mirror\u2019 kind, that we use in managing our business. Why the preference for a prognostic lifetime loss ratios over the traditional kind? Simply put, our company is evolving too fast to rely primarily on a feedback loop that lags as much as statutory loss ratios do. <\/p>\n\n\n\n
So far, so unremarkable. What is remarkable, though, is that 73% of Lemonade’s premiums in Q2 were from customers who have been with us for fewer than 24 months. For incumbents it’s the other way around: the overwhelming majority of their customers are comfortably beyond their second anniversary. A health check that fails to account for the relative newness of our customer base will yield an unduly grim prognosis. <\/p>\n\n\n\n
It follows that, for companies like Lemonade, traditional loss ratios can send a distorted signal, an outdated one, or both. We\u2019re not saying to shrug and disregard loss ratios. But it\u2019s important to look at them with a bit more nuance\u2014and that\u2019s where our machine learning capabilities kick in.<\/p><\/blockquote>\n\n\n\n
By using big data and machine learning to predict lifetime loss ratios, we timeshift a ‘lifetime’ of future data into the present, allowing us to optimize our current business for the long term. (Admittedly, we also really enjoy saying the phrase \u2018timeshift a lifetime of future data into the present.\u2019) <\/p>\n\n\n\n
This is in stark contrast to traditional loss ratios. These also timeshift data to the present, but they do so from the past, rather than from the future. <\/p>\n\n\n\n
Indeed, it\u2019s this deep understanding of our loss ratios that gives us the confidence to reiterate our expectation that our business will operate on a multi-year average loss ratio below 75%. In a deep sense, it already is, once we pay attention to the superior modeling capabilities of LTV6.<\/p>\n\n\n\n
What\u2019s next? We\u2019ll continue to improve and elevate our LTV models, leveraging AI and machine learning with a nuanced understanding of the responsibility those technologies require. <\/p>\n\n\n\n
Lemonade is well aware of the pitfalls and limitations of AI, and we keep the ethics of artificial intelligence at the forefront. It\u2019s why we\u2019ve hired Tulsee Doshi<\/a> as our AI Ethics and Fairness Advisor. She\u2019s also the co-host, along with CEO Daniel Schrieber, of Benevolent Bots<\/a>\u2014our podcast series investigating the ethical implications of AI.<\/p>\n\n\n\n
If you\u2019re interested in how Lemonade is thoughtfully applying cutting-edge tech to one of the most conservative industries, there\u2019s no better place to start.<\/p>\n\n\n\n