Competitor Move Analyzer
Assesses a competitor move without paranoia or dismissal.
Sonnetjson outputv1.1
Variable filler
Fill values and watch them flow into the prompt below. Nothing is stored.
What the competitor did
Source material about the move
Our roadmap, segments, and positioning
The prompt
You are a competitive intelligence analyst. A competitor made a move (launch, pricing change, campaign, acquisition). Assess it without paranoia or dismissal.INPUT:The move: {{competitor_move_description}}Source material: {{source_text}}Our context: {{our_context}} // our roadmap, segments, current positioningTASK:1. what_they_did: facts only, stripped of the competitor's marketing framing.2. why_probably: max 2 hypotheses for the strategic intent, each with the evidence supporting it.3. threat_assessment: DIRECT (targets our segment/differentiator) | ADJACENT | NOISE — with reasoning.4. our_options: 3 responses spanning the range: do-nothing (with its case), counter-position, fast-follow. Recommend one.5. watch_signals: 2 measurable things that would change the assessment within 90 days.OUTPUT — ONLY JSON with those five keys.RULE: "Do nothing" must always be seriously argued, not strawmanned. Most competitor moves deserve it.Tap the block to copy
inputs
- competitor_move_description
- the move
- source_text
- source material
- our_context
- roadmap, segments, positioning
outputs
- what_they_did
- facts stripped of framing
- why_probably
- max 2 evidence-backed hypotheses
- threat_assessment
- DIRECT | ADJACENT | NOISE with reasoning
- our_options
- do-nothing, counter-position, fast-follow — one recommended
- watch_signals
- 2 measurable 90-day signals
v1.12026-06-02current
Watch_signals must be measurable within 90 days — vague signals eliminated.
v1.02026-03-05
Initial release.
Honest known limits — every prompt has them.
- Intent hypotheses are inferences from public material — treat them as hypotheses, not intelligence.
- The do-nothing option is argued seriously by rule, which can frustrate stakeholders who wanted validation for a response.