What a good LTV to CAC ratio looks like for a startup

Startups that report a 3:1 LTV to CAC ratio are usually reporting a 1:1 business with the awkward costs left off the page.

It is the first number I go looking for holes in when a founder shows me a deck. The version on the slide counts the ad spend and stops.

Rebuild it with everything left off, and the whole ratio moves. So does what you can promise investors, and where next quarter's budget goes.

The ratio does two jobs at once. A founder uses it to judge whether the spending is working. To an investor, it signals whether the money is sane. Define cost differently, and the two aren't looking at the same business.

The 3:1 rule of thumb survives on a cost counted short

I see the 3:1 target copied from a blog into a pitch deck. Nobody asks where it came from. Atticus LeBlanc, founder and CEO of PadSplit, had a blunter read on why it holds up.

CAC, the cost of winning one customer, is almost always understated in the decks I see. Some of that, in his telling, is deliberate. "Most of us often underestimate our CAC," he said, "to some degree intentionally."

The 3:1 only looks healthy because the cost underneath it was counted short. Even the ad-spend line flatters itself, because the platforms reporting it overstate how much of the growth they drove.

Here is the part that trips up the founders I work with: these undercounted businesses still work. LeBlanc's read is that the 3:1 rule took hold because CAC was always counted too low. Add the missing costs back, and a functioning company sits nearer one, or one and a half, to one.

A true acquisition cost is mostly payroll, not ad spend

A true acquisition cost has far more in it than the ad spend, which is the smallest line. When I asked LeBlanc to build it honestly, the salaries came first: whoever runs the campaigns.

After that, the software they run on, and a slice of the founder's own hours.

The diagnostic I hand founders is dull, but it works. Count the payroll, the software and a share of your own hours alongside the ad spend, then divide the total into lifetime value, what a customer is worth over their whole stay.

Opportunity cost: the line nobody actually counts

LeBlanc adds one more line, then admits nobody counts it. "Almost no one that I'm aware of factors opportunity cost into CAC," he said. I think he's right that most never try.

Here I keep his honesty and drop his precision. At seed, opportunity cost is the one line you cannot put an honest number on. So I name it as a known gap, rather than invent a figure that makes the ratio look settled.

Averaging lifetime value buries the customer you lose money on

An honest CAC is only half the ratio. Lifetime value is the other half, and it hides the same averaging better, because one blended number feels like a fact.

LeBlanc runs a marketplace that fills rooms for housing providers, and it shows me the problem cleanly. A single lead can turn into wildly different customers.

Some book a room and never move in. Some stay for years. I still repeat his own line to founders: "did they book a room and never move in, did they book a room and stay for years."

Two very different customers, the same cost to win

I keep coming back to one fact. Both cost about the same to win. Averaged into one lifetime value, they describe a customer PadSplit doesn't have.

The average is the trap.

So the company guesses early, from the first click, how long a lead will stay: one day, ten days, or years. That guess is what separates a lifetime value worth dividing into CAC from a number that averages a real, paying customer with a ghost.

The one measure I'd judge marketing on: net move-in

The single measure LeBlanc would judge marketing on is net move-in: the renters who move in, minus the ones who move out. I would use it for the same reason.

It moves only when someone stays, where leads and bookings just count arrivals.

The crude version: split who stayed from who left

The cruder version works too. Split your customers into the ones who stayed and the ones who left. Look at what each cost to win. One honest split beats a blended average.

A thin ratio doesn't always mean the spending is wrong. But it is the first place to check whether you are spending more than your stage should.

A ratio worth raising on answers whether the next dollar comes back

Rebuilt with the whole cost in it, the ratio finally answers the question it was for. Does another dollar of spend come back as more than a dollar?

A founder looking at 1.5:1, with the team and the tooling counted, knows two things. The business works, and the margin is thin. That is a harder number to raise on, and a safer one to run on.