The Cunning Angle

The thesis

The structure of your questions dictates the value of your research. Generic questions return generic findings, the standard-playbook material everyone already has. The Cunning Angle is a framing gate you run before the research, not after: it transforms a target into five questions aimed at five distinct dimensions of competitive whitespace, so candidate angles surface across the whole space instead of clustering on the one axis you already had in mind.

The five questions

Drop these on any target, then score the candidates each one surfaces:

1. Long-tail demand gap. What does this niche's top-attention content actually want that nobody answers well?
2. Integration leverage. What two tools, methods, or perspectives become meaningfully more useful combined than either is alone?
3. Authority arbitrage. Where does the niche's credibility hierarchy have a gap that a rigorous outsider could fill?
4. Polymath cross-pollination. What framework or extraction from an unrelated domain would an outsider see that an insider misses?
5. Emotional pain point. What is the boring vertical's real emotional pain that nobody articulates well?

Each question produces candidate angles with a one-line description and a note on why it is whitespace. Then you score the candidates on reach, fit, and effort, which forces synthesis instead of a pile of undifferentiated ideas.

The anti-trigger

This is the part most people get wrong, so read it first. Cunning for its own sake is overcomplication. Do not run this when the baseline is empty. If the niche has no real coverage, if the inbox has no competing pitches, if the segment is genuinely uncrowded, then the cunning move is to show up and do the standard thing well. Asymmetry is the tool for saturated spaces. When occupation is the question, occupy.

Also skip it mid-build. This is a research-framing instrument, not a pivot tool you reach for in the middle of execution. And do not run it to ratify a decision you have already made; that is asking the framework for permission, not perspective.

Two modes

The same five questions run in two modes that differ only in the tail. In build mode, a major effort like a flagship overhaul or a multi-week sprint, you score the candidates and ship the top three, parking the rest as future extensions. In deploy mode, a single deliverable like one article or one outreach pitch, you ship the top one and hold the rest. Build mode runs once per major effort. Deploy mode runs as often as you need it.