Your Business Model Probably Sucks

Note: This post is adapted from a recent solo episode where I break down the mechanics of unit economics. You can listen to the full audio above.

In the venture world, we live with a scary reality: you often don’t know if the emperor has no clothes until you are very far down the road.

We know if a company can be capitalized. We saw plenty of that in the zero-interest rate era. But we don’t always know if that company is actually accruing value. Just because a VC wrote a check doesn’t mean the machine works.

I’m an engineer by training. And when I look at how most founders (and many investors) approach their business, I see a lack of rigor. They build complex three-statement financial models—income statements, balance sheets, cash flows—before they’ve even proven the basic physics of their business.

If you are early stage, your financial model is fiction.

What you need instead is a Business Model.

Here is how to tell the difference, and how to fix yours before you run out of runway.

In mechanical engineering, you never build a physical object without first building a CAD (Computer-Aided Design) model.

Why? Because the real world is expensive. Materials cost money. Machining takes time. If two parts bang into each other in the real world, you have a disaster. If they bang into each other in a CAD model, you just hit “undo.”

Your business model is the digital representation of the machine you are trying to build.

A professor of mine, Marsh McCall, used to look in the mirror while writing papers and say, “Marshall, your outline is weak, and therefore you are weak.”

If your business model—your outline—is weak, your business is weak. You can’t out-hustle bad math. If you are selling a dollar for 75 cents (looking at you, WeWork), no amount of “community-adjusted EBITDA” or charismatic leadership will save you eventually.

Founders often avoid modeling because they think they need a CFO to build a massive spreadsheet. You don’t. You need what I call the One Column Model.

Open Excel. Use one column. You only need three ingredients:

1. Independent Variables: The inputs you control (Pricing, Ad Spend, Sales Commissions).

2. Dependent Variables: The outputs (Revenue, Churn, Conversion Rate).

3. The Mechanics: The math that ties them together.

I recently sat down with a founder running a brokerage business. We opened a spreadsheet. We listed the deal size ($30k), the commission (10%), and the payout to sales staff. In about 11 rows—maybe 10 minutes of work—we realized the strategy was totally under-optimized. The volume required to just pay his sales staff a decent wage was impossible at his current price point. So he adjusted the initial target market, and began executing on that.

That 10-minute exercise saved him two years of grinding on a strategy that was mathematically unfeasible.

This is where people get confused.

A Financial Model is accounting. It relies on GAAP standards (Generally Accepted Accounting Principles). Concepts like EBITDA are useful, but they are innovations designed to standardize reporting, not necessarily to tell you the economic truth.

A Business Model is about mechanics. It answers one question: What is the Unit and how does it work?

If you tell me your “unit economics” are good, my first question is: “What is your unit?”

• Is it a customer?

• Is it a contract?

• Is it a node?

The unit is the precise point where the Product Side (Revenue) meets the Cost Side (CAC & COGS).

Take Uber. It’s a complex marketplace. What is the unit? It isn’t just a driver, and it isn’t just a rider. You have to find the least common denominator over a period of time.

• Let’s say a driver does 300 rides a month.

• Let’s say a rider does 30 rides a month.

• The Unit is 1 Driver + 10 Riders = 300 transactions/month.

You have to calculate the cost to acquire 1 driver and 10 riders, and compare that against the margin on 300 rides. That is your unit economics. If that equation doesn’t result in a positive contribution margin quickly, you don’t have a business; you have a subsidy.

Once you understand your unit, you have to look at the quality of the fuel you’re putting in the engine. Not all revenue is created equal.

When I was at NEA, Harry Weller taught me that Enterprise SaaS is the greatest business model in history for one reason: It accumulates and it pays upfront.

You pay to acquire a customer once. They pay you $100. Next year, you don’t have to pay to acquire them again, but they pay you another $100. Then you acquire a new customer. Now you have $200.

This is distinct from a “consumable” business (like hardware or e-commerce) where you have to re-sell the customer constantly.

When analyzing your revenue quality, score it on three vectors:

1. Reliability: Will it happen again automatically? (Subscription > Single Purchase).

2. Profitability: What is the contribution margin? (Software > Hardware).

3. Velocity: How fast does the cash hit the bank? (Unearned Revenue > Accounts Receivable).

You want high velocity, high reliability and fundamentally strong unit economics. If you have low velocity (18-month sales cycles), low reliability (high churn) and low gross margin (high variable costs), you are in the danger zone.

People like to say “Data is the new oil.” I disagree.

Data is blood.

When you go to the doctor for a blood test, they stick a needle in and take a sample. That sample tells them how the organs are functioning. Is the liver failing? Is the heart pumping? Is there an infection?

Financial data is the blood of your organization. It is the diagnostic tool that tells you if the organism is healthy.

It is the leading and lagging indicators. The independent and dependent variables. All in one place.

If you build the right business model (getting the mechanics right) and you monitor the blood properly, you shouldn’t have to make hard decisions.

I tell founders: “Great analysis results in non-decisions.”

If the data is clear, the choice should be obvious. “Oh, we lose money on customers under $10k ACV? Great, we stop selling to them.” That’s not a decision; that’s a realization.

Stop guessing. Do the math. If your outline is weak, you are weak.