# Operator context

The "data" in this demo is the operator context: the two paragraphs that the sharp prompt adds to the generic prompt. These paragraphs describe the situation the strategic analysis should be framed for.

## The operator context used in the demo

```
I'm building a methodology-and-application business at the
intersection of AI craft and operator development. The methodology
is the product I'm developing. The application work is where I prove
it works in practice and where the methodology gets sharpened.

Publishing is one channel of distribution for the methodology,
not my primary product. The strategic question I bring to this
analysis is how AI changes publishing as a distribution channel
for operators with a methodology to teach, not how AI changes
publishing as a primary business.
```

Two paragraphs, each carrying one specificity move. Paragraph 1 names what kind of business is being built and how the parts relate. Paragraph 2 names where the analysis topic (publishing) sits in that business and the specific strategic question being asked. Without paragraph 1, the analysis has no operator. Without paragraph 2, it has an operator but no specific question.

## Why this context works for the demo

The video host (the person running the demo) is actually building this business. The operator context isn't synthesized for demonstration purposes; it's the same context the host would supply when using AI for real strategic work.

That lived authenticity matters for two reasons:

1. **Operator context that's vague or inauthentic produces vaguer output.** AI amplifies what's there. If the operator context describes a generic situation, AI generic-amplifies it.
2. **The viewer trusts the demo more.** Watching the host engage with output that's actually relevant to their work registers as real strategic engagement, not contrived demonstration. The demo is teaching by doing.

## Adapting the operator context to your own situation

Swap the two paragraphs with your own. The territory move stays as is; wording can be adapted to your topic per the variations in [`prompts.md`](./prompts.md). To make the contrast as visible as possible:

**Be specific about what makes your situation specific.** Generic context produces generic output. Test by asking "could this paragraph describe ten different operators?" If yes, sharpen it.

**Name the tradeoff or decision you're sitting in.** "I'm thinking about marketing" is too generic; "I'm deciding between paid acquisition and content-led growth, with a budget of $50K and a 6-month runway" is specific.

**Use lived context.** If you're using AI for a real decision you're making, paste the actual context. If you're trying the demo, pick a real situation you've been in. The more recent and specific, the better.

**Length: 1-3 paragraphs.** Enough to specify; not so much that the operator context dominates the task. Two paragraphs is the median.

## Templates for strong operator contexts

For a strategic-analysis ask:

```
I'm a [role] at [type of company] with [size/stage]. I'm
deciding [specific tradeoff] given [specific constraint]. The
question I'm bringing is [specific decision].
```

For a writing-task ask:

```
I'm writing [specific piece] for [specific audience]. The
piece needs to do [specific thing] without [specific failure
mode]. My voice on this is [specific voice notes].
```

For a thinking-through ask:

```
I'm trying to figure out [specific question] in the context
of [specific situation]. I've already considered [specific
angles]. What I'm stuck on is [specific gap].
```

The more specific the operator context, the more AI has to amplify. The less specific, the more the output drifts toward the generic average.

## What to avoid in operator context

**Don't describe yourself, describe your situation.** "I'm a hardworking entrepreneur with a track record of execution" is identity claim, not context. "I'm three months into building [specific thing] with [specific constraint]" is context.

**Don't include the conclusion you want.** Operator context is the situation the analysis should be framed for, not the answer the analysis should produce. If you tell the model the answer, you'll get it back.

**Don't be vague to feel safe.** Specific operator context produces specific output, which is sometimes wrong in ways the reader can correct. Vague operator context produces vague output, which feels safer but isn't actionable. The demo's value is in the specificity.