Revenue Quality Scorecard

Ask What the Money Required

Revenue proves that somebody paid. It does not, by itself, prove that the same customer will pay again, that another customer will buy the same offer, or that the product can deliver the promise without consuming the founder.

Use this scorecard once real money is involved: a paid pilot, subscription, implementation fee, expansion, renewal, or services engagement attached to the product. Score each meaningful revenue source separately before looking at the total. A large custom contract can otherwise make a small product look validated while quietly turning the founder into its delivery team.

The scorecard should answer a decision, not certify a business. Write that decision at the top:

REVENUE QUALITY REVIEW — [DATE]
Decision this review must inform:
Evidence window:
Target segment and recurring job:
Standard product and price being tested:
Natural renewal or value cycle:
Accounts included and excluded:
Founder-hour decision rate:
Evidence sources:

The founder-hour rate is a planning assumption, not an accounting claim. Use the same reasonable rate across reviews so custom delivery and support remain visible. Also record cash revenue and ordinary gross margin in the business’s normal accounts; this scorecard is for product judgment.

Build the Account Ledger First

Use one row per account, or per distinct contract when an account has bought materially different work. Link every judgment to billing, usage, retention, sales, support, and delivery evidence.

account_id | segment | offer_and_price | revenue_type
contracted_revenue | normalized_monthly_revenue | first_paid_at | last_paid_at
renewal_or_next_cycle_due | value_cycles_completed | current_usage_evidence
sales_hours | discount_or_special_terms | setup_hours | recurring_delivery_hours
support_hours | custom_code_or_manual_output | direct_recurring_cost
refunds_or_credits | expansion_or_contraction | evidence_links | note

Keep cash received distinct from recurring revenue. A twelve-month prepayment is welcome cash, but it does not create twelve observed renewals. Keep an implementation fee distinct from subscription revenue, and label services as services even when they helped win a product customer.

Time boundaries matter. Setup effort belongs in the acquisition and payback view; recurring manual work belongs in delivery cost. Do not quietly move an ongoing export repair, data cleanup, report assembly, or customer chase into “onboarding” every month.

Score Five Claims

For each account, mark every claim 2 — strong, 1 — mixed, 0 — weak, or ? — unknown. Add one sentence of evidence. The number supports comparison; the sentence carries the judgment.

1. Retained value

• 2: The account has paid through at least one natural renewal or value cycle and continues to use the product for the promised job.
• 1: Payment and use are real, but the next meaningful cycle has not arrived, or retention still depends on founder prompting.
• 0: The account paid without sustained use, failed to renew, received a refund, or bought a one-off outcome.

2. Repeatable sale

• 2: The target segment bought the standard offer at the intended price through a sales motion that another comparable account could follow.
• 1: The offer fits, but a founder relationship, material discount, long negotiation, or unusual term helped close it.
• 0: The sale required a different product, exceptional pricing, or a promise the standard offer does not make.

3. Product-delivered value

• 2: The product carries the recurring job with ordinary onboarding and bounded support.
• 1: The product creates value, but manual operations or founder rescue remain material and plausibly removable.
• 0: Custom code, analysis, data work, implementation, or personal service is the main delivery mechanism.

4. Economic contribution

Calculate two views:

product contribution
  = normalized monthly revenue
    - direct recurring cost
    - refunds and recurring credits

decision-adjusted contribution
  = product contribution
    - (recurring delivery and support hours × founder-hour decision rate)

• 2: Both views are positive with room for normal variance, and setup effort has a credible payback period.
• 1: Contribution is positive but thin, volatile, or sensitive to founder time, usage cost, discounts, or slow payback.
• 0: The account loses money on a recurring basis or only appears healthy when founder delivery is treated as free.

5. Expansion from value

• 2: The account has expanded because more people, usage, workflows, or outcomes receive the product’s value at standard terms.
• 1: There is concrete room to expand, but no paid expansion has occurred.
• 0: There is no credible expansion path, or proposed expansion is another custom engagement.

An early account can honestly have an unknown expansion score. Unknown is a request for evidence, not a zero and not permission to fill the cell with an optimistic forecast.

Read the Pattern, Not Just the Total

Record the five marks side by side because the comparison is the work of the scorecard:

account | monthly revenue | retained | repeatable | product-delivered
economic | expansion | blocking fact | decision

Do not average away a zero in retained value, repeatability, or product-delivered value. A profitable services contract may be a good services business, but its margin cannot turn it into product-market-fit evidence. Likewise, a standard subscription that has not reached its first renewal is promising revenue, not retained revenue.

After scoring accounts, summarize both dollars and account counts:

Total normalized monthly revenue:
Revenue with no zero in retained, repeatable, or product-delivered value:
Revenue not yet mature enough to judge:
Revenue dependent on custom work or recurring founder delivery:
Product contribution:
Decision-adjusted contribution:
Founder sales, setup, delivery, and support hours:
Renewals, contractions, churns, refunds, and expansions in the window:
Strongest segment-and-offer pattern:
Largest fact the headline revenue conceals:

A $4,000 Month Can Contain Three Businesses

Consider a modeled reporting product with three paying accounts. Aster pays $600 per month for the standard plan, has completed four weekly reporting cycles without help, and uses half an hour of support. After $60 of direct cost and founder time valued at $100 per hour, its decision-adjusted contribution is $490.

Beacon pays $1,000 and uses the core product, but the founder spends five hours each month repairing an unsupported export. With $160 of direct cost, its decision-adjusted contribution is $340. The willingness to pay is real; repeatable delivery is not yet proven.

Cedar pays $2,400 for a pilot. The contract includes a custom connector and eighteen recurring hours of founder-produced analysis. After $300 of direct cost and the same founder-hour assumption, only $300 remains. Cedar supplies 60 percent of headline revenue, yet the largest payment buys a materially different offer and most of the delivery still comes from a person.

The business received $4,000. The scorecard shows $1,130 of decision-adjusted contribution, only one independently delivered account, and two different questions. Beacon may justify a bounded import fix followed by another unassisted cycle. Cedar requires an explicit choice: price and run it as services, redesign the work into a repeatable product capability, or decline the work. Calling all three accounts recurring product revenue would erase that choice.

Turn the Review Into One Refusal and One Test

Finish with a decision record:

REVENUE QUALITY DECISION
Revenue pattern worth protecting:
Revenue pattern that contradicts the product thesis:
Account or segment to pursue:
Account or segment to stop pursuing:
Custom promise, discount, or founder task to refuse:
One product or process change to test:
Accounts exposed to the test:
Evidence expected by the next natural cycle:
Result that would reject the current explanation:
Review date:

Growing acquisition is justified only when the standard offer has evidence of retained value, repeatable sale, product-delivered value, and positive economics. When one of those claims is weak, choose the smallest test that can strengthen or reject it. When the same weakness appears across accounts, fix the offer or delivery system before buying more demand.

Protect the history. Do not rewrite a pilot as a subscription after it converts, delete founder hours after automation, or rescore an old account using facts learned later. Date a new review and let the movement show whether revenue is becoming easier to sell, deliver, retain, and expand.

The scorecard is complete when the founder can say which revenue demonstrates the product, which revenue pays for learning or services, and which kind of revenue the next month must contain less of.