An updated primer on the formula, for startup founders.
I’ve taught CAC<LTV to my students at Stanford for more than a decade. In that time, it’s gone from being a little-known concept to being an often-misunderstood concept. So here are a few updated thoughts.
A quick review of the concept:
A business is an engine that attracts customers, delivers something of value to them, and then extracts that value in the form of profits. That’s what a business is. And so it logically follows that the cost of attracting a new customer needs to be less than the value we can extract from that customer. If it costs us $15 in advertising to get a customer, and we can only make $7 from them then Houston, we have a problem. But if it costs $15 to get them, and then once they are a customer they make a bunch of repeat purchases that yield $75 in profit, then we are happy. Ultimately every venture of every kind has to have a Customer Acquisition Cost (CAC) that is less than the Lifetime Value (LTV) of a customer. It’s a simple, self-evident concept.
Yet it’s a leading cause of startup death.
A high percentage of startups die because their cost of getting customers turns out to be higher than they can make from them. Partly this is just because we’re all optimists — we all think our startup is so awesome that people will flock to become customers and they will remain customers forever. But eventually that optimism fades as we realize that marketing is expensive and no customer stays forever. The immutable laws of economics set in, and at some point many startup founders find that their LTV/CAC ratio is slowly draining the bank account. To paraphrase Ernest Hemingway, Startups go broke two ways: gradually, and then suddenly.
Investors tend to obsess on the LTV/CAC ratio.
Obviously, investors care about your LTV/CAC ratio because it’s the essence of a successful business. But it’s also a proxy for the potential ROI of their investment. If you have proof that you can spend $1 on customer acquisition activities and get $5 in value back (a LTV/CAC ratio of 5.0), investors will want to shovel as much money as possible into that engine. As a VC friend of mine says “What I’m looking for is a just-add-money opportunity”. Having a business with a LTV/CAC ratio of over 5.0 looks like a “just-add-money opportunity” to investors.
But it’s a blunt tool that is better if sharpened.
Let’s say that during the quarter we spent $10,000 on sales and marketing and got 1,000 new customers — a CAC of $10. But probably some of those customers came through word-of-month, some came as referrals, some came from our PR efforts, and some came from paid advertising. So we had a blended CAC of $10, but that doesn’t tell us anything about the relative effectiveness of each of our different customer acquisition efforts. Which leads me to the next point:
Not all customers are created equal.
With every business I’ve ever run, I’ve realized at some point that 80% of our profits were coming from 20% of our customers. It’s amazing how this tends to be true with almost all businesses. So if we look at LTV (Lifetime Value) of our entire universe of customers, we’ll probably see that 20% of them have a much higher individual LTV than the rest. Wouldn’t we want to focus our CAC efforts on getting more of the high-LTV customers? Yes, we would.
Therefor, cohorts matter.
The two points above would indicate we really want to track LTV/CAC ratio by customer cohort. For example, what’s the ratio for customers acquired through Facebook advertising vs those acquired through Google advertising? Knowing that would tell us a lot about how we should allocate advertising dollars. What’s the LTV/CAC ratio for customers acquired through our referral program? Knowing that would tell us how much we can afford to offer in a referral fee. Knowing your company’s blended LTV/CAC tells you the health of the overall engine, but it doesn’t tell you how to optimize the engine’s performance for next quarter. Tracking customer cohorts tells you that.
Also, velocity matters.
One afternoon recently, I sat in the back yard of longtime Silicon Valley venture capitalist Tim Connors as he drew graphs for me on his whiteboard (only VC’s have whiteboards in their back yards). He explained that he doesn’t care about the LTV/CAC ratio, per se, what he cares about is the velocity with which invested CAC comes back in the form of LTV. So he’s developed a metric he calls CACD — the D is for “doubled”. CACD answers the question, “If we spend $12 in customer acquisition activities, how long does it take for us to get $24 back?” As an investor, he wants to see a a business with a CACD of less than eight months. Tim’s formula gets to the heart of an inherent flaw in the LTV/CAC ratio: it doesn’t include a time factor. A business with an LTV/CAC ratio over 5.0 might seem good at first, but if you have to service a customer for 10 years before you make back the money you spent getting him, then it doesn’t seem so good, right? Velocity matters. So think about how you can measure CACD for your business. Putting $12 somewhere where it returns with a high velocity will accelerate your engine of growth (and make Tim happy).
The CAC<LTV concept applies to every business of every kind. Every venture must have a sustainable way to get customers at a cost less than the venture can make from them. It’s an immutable law of economics. Ignoring the formula (or misunderstanding it) remains a leading cause of startup death. As legendary venture capitalist Bill Gurley once wrote in “The Dangerous Seduction of the LTV formula”, “the formula can be confused, misused, and abused, much to the detriment of the business”. So don’t do any of those things, or it will make Bill mad.
Sharpen the tool by tracking customer cohorts, improve the formula by adding a time factor, and remember that not all customers are created equal. If you do those three things, you will have an engine of growth that makes you happy, and investors eager.