Run-Rate Revenue Explained: What an AI Company’s Number Means

Business Pill 33 · What a revenue run rate really says

A run rate is the most recent month multiplied by 12. It shows how fast a company is moving now, not what it earned.

A short explainer video. The numbers in it are round numbers for illustration.

The Short Answer

Fast-growing companies rarely quote last year’s sales. They quote a run rate. Take the most recent month and multiply it by 12. That is the run rate.

A Worked Example

The video uses round numbers. A company made $10 million in January and $100 million in December. Its run rate is $1.2 billion. But what it actually earned over the year was far less, about $500 million.

Both numbers are true. They answer different questions. The run rate says how fast the company is moving now. The annual figure says how much it actually earned.

Run Rate and Recurring Revenue

A related term is annual recurring revenue. It counts only revenue under contract that repeats, like subscriptions. Run rate is looser. It can include usage that may not repeat next month.

Why AI Companies Use It

AI companies favor run rate because they are growing so fast that last year’s total is already out of date. That is fair.

But much AI revenue is usage-based: customers pay for what they consume, and consumption can fall as easily as it rose. So a run rate is a snapshot, not a promise. One unusually good month, multiplied by 12, becomes an unusually good year that never happened.

Three Questions to Ask

1. Which month was multiplied?
2. How much of it is under contract?
3. What did the company actually earn over the last 12 months?

See It in the News

OpenAI’s Revenue Run Rate Nears $70 Billion. A news piece on a run-rate figure and what it measures.

Higgsfield’s $1B Run-Rate and the $4M Model Bill. A news piece setting a run rate against the cost of the models behind it.

More Business Pills

• The Memory Wall: Why a Faster AI Chip Is Not Faster AI
• Tokens per Watt: What an AI Data Centre Actually Produces
• Why AI Can’t Be Both Instant and Cheap: Latency vs Throughput
• The Model and the Harness: Why Same-Model Products Differ
• Context, Not Capability: Why a Smart AI Model Gives Poor Answers
• Discardable Software: When Code Is Cheap Enough to Throw Away
• Human in the Loop vs Human on the Loop: Supervising AI Agents
• The AI Audit Problem: When Making Work Is Cheaper Than Checking It
• AI Evaluation as Acceptance Test: How to Know It’s Good Enough
• Extensibility Is Control: Who Holds the Power in an AI Product
• RLHF Explained: How an AI Model Learns What People Prefer
• Pretraining Explained: How an AI Model Learns Before Anyone Teaches It
• Fine-Tuning Explained: How to Adapt a General AI Model to One Job
• Tokens Explained: The Unit AI Reads, Writes and Bills In
• Embeddings Explained: How a Machine Compares Meaning
• RAG Explained: How a Model Answers From Your Documents
• AI Hallucination Explained: Why Models State False Things
• Distillation Explained: How a Small Model Learns From a Large One
• Reasoning Models Explained: What Changes When AI Thinks First
• Tool Use Explained: How an AI Model Goes From Text to Action
• Capex and Depreciation Explained: Why Chip Lifetime Drives AI Profits
• Backlog Explained: Revenue That Is Signed but Not Yet Earned
• Switching Costs Explained: Why It Is Hard to Leave an AI Supplier