Unit Economics Is Not What You Think It Is

Everyone talks about unit economics like it's a binary switch: "good unit economics" or "bad unit economics."

Like you either have it or you don't.

That's where it gets broken.

---

The most profitable company in the world might have unit economics that look terrible.

Here's what happened when I started looking at actual founder math versus what everyone says unit economics should be:

DoorDash went public with a CAC (customer acquisition cost) of $20-30. Their LTV (lifetime value) in year one? Around $40-50. That's only 1.5x.

The venture playbook says you need 3:1. DoorDash looked broken.

Yet they scaled to a $30B+ company.

Same with Uber. Same with Airbnb. All of them had unit economics that looked "bad" by the standard metrics people quote.

Here's the thing: everyone's measuring the wrong thing.

They're measuring whether a single customer is profitable. What they should be measuring is whether the machine itself is sustainable.

These are completely different problems.

---

The Misunderstanding

When people talk about unit economics, they're usually saying one of two things:

"Get your LTV to 3x CAC and you're golden."

Or: "Unit economics are bad — you need better retention."

Both of these are looking at the wrong metric.

What you actually need to know is simpler: Can the cost of serving one more customer be covered by what that customer generates?

Not eventually. Now.

This is called contribution margin. And it's the only metric that matters before you scale.

DoorDash disclosed this when they went public. A $15 order has roughly:

• Restaurant payout: $10
• Delivery costs: $2
• Contribution: $3

That's 20% margin. That $3 is what has to cover marketing, engineering, support, and infrastructure.

Sounds thin. It is. But here's why it matters: that $3 is positive. The order sustains itself. The math works today, not "eventually when we reach scale."

Airbnb's S-1 showed the same thing. Their gross profit per booking was 15-20% in early days. Their total profitability was negative because acquisition costs were high.

But the booking itself was profitable. That's different from a booking that loses money.

One scales. The other doesn't.

---

Where Founders Get This Wrong

There are three places where the model breaks.

1. You're optimizing for the wrong metric

CAC (customer acquisition cost) is useful. But it's not the bottleneck.

What matters is: cost per dollar of revenue acquired, weighted by retention.

A customer costing $50 to acquire is not the same if they spend $100 once versus $100/year for three years.

Stripe publishes their ARR retention and expansion in tax filings. What they found: a customer with 95% retention and 5% expansion is worth 10x a customer with the same first-year spend but 70% retention.

Most founders track churn as an afterthought. It should be front and center because it fundamentally changes your unit economics.

2. You're not accounting for what changes at scale

Uber and Lyft both hit this hard.

Their CAC in early markets (San Francisco, New York) was completely different from later markets (smaller cities, international).

Lyft's IPO filing showed this explicitly: their monthly churn varied dramatically by geography and cohort. First movers in a market had 40% lower churn than newer cohorts.

Why? Because:

• Early adopters cost nothing (they find you)
• At scale, you're acquiring less interested people
• Competition drives up prices
• Your retention changes with market maturity

Your unit economics at 10,000 customers look nothing like at 1M. Most founders project linearly. The market doesn't work that way.

3. You're thinking about the wrong time horizon

DoorDash's Founder Letter (2019) said something most founders miss:

"We focus on contribution profit, not overall profitability."

Why? Because overall profitability is a question about: "Have we reached the scale where fixed costs spread thin enough?"

Contribution profit answers: "Right now, at our current scale, can new orders sustain themselves?"

DoorDash's overall profitability looked bad. Their contribution margin looked great. That meant scale would fix it.

And it did.

Most founders optimize backwards. They make the P&L look good on paper by treating fulfillment and unit-level costs as "overhead." Then they scale, and the unit-level math breaks because it was never actually sustainable.

---

What Actually Matters

Here's what I'd track instead of LTV/CAC ratio:

Step 1: Calculate your contribution margin

Revenue per customer minus direct costs to serve them.

For Lyft (from their financials):

• Ride revenue: $100
• Driver payout: $75
• Payment processing: $3
• Contribution: $22 (22%)

For Stripe (disclosed in filings):

• Processing fees: $3 on a $100 transaction
• Payout to customer: $0.30
• Processing costs: $0.20
• Contribution: $2.50 per transaction

For Airbnb (from S-1):

• Host commission: ~25% of booking
• Their take after payment processing: ~3-5%
• But transaction costs are minimal once scaled
• Contribution margin: grows with scale, starts at ~15-20%

Calculate this honestly. Not averages. The marginal order.

Step 2: Calculate your CAC payback period

How many months does it take for a customer to generate enough contribution to pay back what you spent acquiring them?

Example:

• Total customer acquisition cost: $200
• Contribution per month: $40
• Payback period: 5 months

DoorDash has been public about this: their CAC payback is 8-12 months depending on market maturity.

For B2B SaaS: $50/month subscription, $40 contribution, $200 CAC = 5 month payback. That's aggressive but sustainable.

If your payback is 18 months and you have 12 months of runway, you don't have a scaling problem. You have a business model problem.

Step 3: Stress test your cohort assumptions

Uber's S-1 showed retention curves by cohort and geography. The decay pattern was consistent: sharp drop months 1-2, then stabilization.

When stress testing, ask:

• What if retention is 20% worse?
• What if contribution margin compresses 10% (price competition)?
• What if CAC increases 30% (market saturation)?

Airbnb has dealt with all three. Their margins have compressed, their CAC has increased, but their retention improved. The net effect: the unit economics still work.

Step 4: Ask the hard question

At your current contribution margin and retention, does this business scale to where you want it?

If you want a $1B company but your unit economics only work at 40% margins and you're at 8%, you have a problem. Either:

• You're in the wrong market segment (someone will pay more)
• Your product costs too much to deliver
• You're in a race to the bottom

These aren't fundraising problems. They're architecture problems.

---

Why This Matters

The companies that survived and scaled—DoorDash, Uber, Airbnb, Stripe—all had one thing in common:

Positive contribution margins from day one.

They had other problems. Acquisition was expensive. Overall profitability was years away. But the unit itself worked.

The companies that didn't survive? Negative contribution margins. They were betting that scale would fix the math.

Pets.com had negative unit economics. They scaled anyway. Burned through $300M in two years.

The ones that built real companies are the ones that fixed the unit economics first, then acquired customers.

---

What I'd Do

If you're a founder and your unit economics are fuzzy right now, don't build a perfect model. Just:

1. Calculate your contribution margin per customer action, honestly.
2. Calculate how long it takes that contribution to pay back your CAC.
3. Compare that payback period to your runway.
4. If payback > runway, fix the unit economics before scaling acquisition.
5. If payback < runway, you can afford to acquire while you improve retention and margin.

Everything else — LTV/CAC ratios, cohort analysis, blended CAC — is optimization on top of this foundation.

Most founders know this. They're just hoping the problem solves itself.

It won't.

---

References

• DoorDash S-1 Filing & Founder Letters
• Uber S-1 Filing (2019)
• Airbnb S-1 Filing (2020)
• Lyft IPO Filing (2019)
• Stripe Tax Filings & Public Statements

---

If you're building something and your unit economics are the bottleneck, this is the framework to fix it.

Share this with one founder who needs to think differently about how their business actually works.

— Druhin