Unit Economics Validation SOP

A unit economics validation proves that one customer, order or subscription earns more money than it costs to win and serve. It takes a few days to model and two to three weeks to check against real payments. If one unit loses money, more units lose more money, and no amount of growth fixes it. This SOP turns a spreadsheet promise into numbers backed by invoices, with a clear pass or fail.

What is unit economics validation and when should you use it?

Unit economics is the profit and loss of a single unit of your business. The unit is whatever a customer pays for repeatedly: an order for a shop, a subscriber month for software, a booking for a marketplace, a seat for a B2B tool. Two numbers sit at the centre. Contribution margin is revenue per unit minus the costs that grow with each unit. Contribution after acquisition subtracts what it cost to win the customer, spread over their lifetime.

Use this SOP:

• Once you have paying customers or a paid pilot, so real costs replace guesses.
• Before you spend on growth, because ads and sales multiply whatever your unit earns or loses.
• When revenue grows and cash shrinks. That pattern almost always traces back to a unit that loses money.
• Before a fundraising round, where unit economics is one of the first slides investors check.

This page combines inputs. The detailed method for acquisition cost sits in the customer acquisition cost validation SOP and for lifetime value in the customer lifetime value analysis SOP. How many units cover your fixed costs belongs to the break-even analysis SOP.

What do you need before you start?

• 30 to 90 days of real transactions, exported from Stripe, Shopify or your invoicing tool.
• Supplier and tool invoices for the same period: hosting, APIs, packaging, shipping, payment fees.
• A time log for support and delivery, even a rough one, so people costs per unit are visible.
• Your current CAC, ideally fully loaded and per channel.
• A spreadsheet with one tab for the model and one for actuals.

How do you run it, step by step?

1. Pick one unit and define it in a sentence

Write it down: "one paying subscriber for one month" or "one delivered order". Choose the unit closest to how customers pay. If you sell both subscriptions and one-off add-ons, model the subscription first and treat add-ons as extra revenue per unit.

2. Calculate net revenue per unit

Start from the list price, then subtract discounts, refunds, chargebacks and sales tax you pass on. Use the average real price paid, which is always lower than the price on your website. Founders who skip this step overstate revenue by the size of their launch discount.

3. List every variable cost per unit

Include anything that rises with one more unit: cost of goods, packaging, shipping, payment fees, hosting and third-party API usage per user (AI model calls count here), and support or delivery minutes priced at a real hourly rate. Leave out rent, salaries for the founding team and other fixed costs. They belong in break-even, and mixing them in hides whether the unit itself works.

4. Calculate contribution margin

Contribution margin per unit equals net revenue minus variable costs. Express it in dollars and as a percentage. A negative number ends the analysis here: fix price or cost before anything else.

5. Add acquisition cost and payback

Take CAC from your acquisition tests and divide it by monthly contribution margin per customer to get CAC payback in months. Then compare lifetime contribution with CAC. David Skok's widely used SaaS metrics thresholds are payback within 12 months and lifetime value at least 3 times CAC.

6. Check the model against actuals

Rebuild the same numbers from real invoices and payment exports for the last 30 to 90 days. Write the gap for each line: modelled versus actual. A gap above 10% on any large line means your model has a blind spot, frequently in support time, refunds or payment fees on small orders.

7. Stress test the unit

Run three what-ifs: price 10% lower, variable costs 20% higher and churn 50% higher. Then ask what changes at 10 times your volume: which supplier prices fall, and which costs rise (support, fraud, returns). A unit that stays positive under all three tests is ready for growth spend.

Template you can copy

UNIT ECONOMICS SHEET            Model     Actual (last 30-90 days)   Gap %
Unit definition: ________________________________

Revenue per unit
  List price                    ____      ____
  - Discounts and refunds       ____      ____
  = Net revenue per unit        ____      ____

Variable costs per unit
  Cost of goods / delivery      ____      ____
  Payment fees                  ____      ____
  Hosting and API usage         ____      ____
  Shipping and packaging        ____      ____
  Support minutes x hour rate   ____      ____
  = Total variable cost         ____      ____

Contribution margin $           ____      ____
Contribution margin %           ____      ____

CAC (fully loaded)              ____
CAC payback (months)            = CAC / monthly contribution margin
Lifetime (months)               = 1 / monthly churn
LTV:CAC                         = (contribution margin x lifetime) / CAC

STRESS TESTS                    Contribution   Payback
  Price -10%                    ____           ____
  Variable costs +20%           ____           ____
  Churn +50%                    ____           ____
Verdict: pass / fix price / fix cost / fix acquisition

Worked example: an AI bookkeeping assistant for freelancers

This is an illustrative case with invented numbers. A founder charges $29 a month. Her model, checked against 60 days of real data, shows:

Her fully loaded CAC is $210, so payback takes 12.2 months, just over the threshold. The model had assumed 4 minutes of support per user; the time log showed 12, mostly questions about connecting bank accounts.

She writes three help articles and adds an in-app setup checklist. Support falls to 4 minutes ($2.00), contribution rises to $21.16 (73%) and payback drops to 9.9 months. Under the stress test with costs 20% higher, payback stays at 10.7 months, so she starts paid acquisition.

What should you measure, and when do you stop?

Stop when the actual numbers match the model within 10% and the unit passes all three stress tests. If it fails, pick the weakest lever (price, a single cost line or acquisition) and fix that one before you rerun the whole analysis.

Which mistakes should you avoid?

• Mixing fixed costs into the unit. Your salary does not grow with each order; keep it in break-even.
• Using list price instead of the price people pay. Discounts, annual plans and refunds lower real revenue.
• Pricing your own time at zero. Support and delivery minutes need an hourly rate, or the unit looks healthier than it is.
• Forgetting usage costs that scale. AI model calls, SMS fees and storage grow with heavy users; check your top 10% of users separately.
• Treating a forecast as validation. A spreadsheet with no invoices behind it is a hypothesis.

How can AI speed this up?

Give an AI assistant your Stripe export and supplier invoices and ask it to calculate average net price, payment fees per transaction and cost per active user. It can build the stress-test table and explain which line moves payback the most. Check every formula it writes by recalculating one row by hand, because a single wrong cell reference spreads through the whole sheet.

To turn validated numbers into a plan you can share, the free business plan generator builds the financial section around your unit figures.

Where does this SOP fit in your validation plan?

Unit economics is the checkpoint between "people buy it" and "we can grow it".

1. Feed in acquisition cost from the customer acquisition cost validation SOP.
2. Feed in lifetime from the customer lifetime value analysis SOP.
3. If price is the weak lever, run the pricing sensitivity testing SOP.
4. If a cost line is the problem, go back to the cost structure validation SOP.
5. Roll the validated unit into the company-wide financial model validation SOP.

See every Business Model Validation SOP. The free investor playbook module of the F/MS startup school shows how investors read these numbers in a pitch.

FAQ

What are unit economics in simple terms?

The money one unit of your business brings in minus the money that unit costs, including what you paid to win the customer. Positive unit economics means each sale makes you richer.

What is a good contribution margin for a startup?

It depends on the model. Software sits high because serving one more user costs little, and physical products sit lower. What matters is that the margin pays back CAC within 12 months.

Can I validate unit economics before launch?

You can build the model before launch, and a paid pilot with 10 to 20 customers gives you enough real costs to validate it. Without any paying customers, treat the numbers as a hypothesis.

What is the difference between unit economics and break-even analysis?

Unit economics asks whether one unit makes money. Break-even asks how many profitable units you need to cover fixed costs such as salaries and rent.