forecast-competitive-response

name	forecast-competitive-response
description	This skill should be used when the user asks "how will competitors react to this," "will this move invite retaliation," "war-game this decision," or has a planned strategic move that needs competitor response forecasted from their actual incentive structure, not generic assumption.
allowed-tools	Read, Grep, Glob
metadata	{"version":"0.3.0","owner":"market-intelligence practice","review_cadence":"quarterly","work_shape":"hypothesis-driven-analysis","permission_tier":"advisory","output_class":"draft-for-review","sourcing_policy":"volatile-facts-must-be-sourced"}

Forecast Competitive Response

When to use

CI analysts and strategy leads war-gaming a planned move — ground predictions in competitor incentive structures, not generic "they'll probably react."

What this skill does not do

• Does not build incentive maps from scratch — route to /market-intelligence:map-incentives if structures are unknown.
• Does not map the landscape — use /market-intelligence:map-competitive-landscape for strategic groups.
• Does not recommend whether to proceed — surfaces retaliation scenarios; strategist decides.

Preconditions

Input	If missing
Planned move stated precisely	Ask once to narrow scope, scale, timing
Competitor set or strategic groups	Halt — route to map-competitive-landscape or ask user to name competitors
Incentive structures per competitor	Proceed with [PROVISIONAL] — flag predictions as low confidence; route to map-incentives

Provisional mode

Without incentive maps: label CONFIDENCE: structured first pass; tag every retaliation prediction [review]; do not present generic industry assumptions as grounded forecasts.

Trust spine

• Confidence bands (hypothesis-driven-analysis):
  • High: Move precise, incentive maps available, visibility/threat classified per competitor with falsifiable retaliation hypotheses.
  • Medium: Some incentive gaps; predictions stated as hypotheses with explicit assumptions.
  • Low: No incentive data — scenario sketch only, not a forecast.
• Failure modes:
  • Strategic advice vs. support: "Plan survives predicted response" is a test result, not a go/no-go; strategist owns the decision.
  • Client confidentiality: Competitor intelligence may be sensitive — CONFIDENTIAL header per plugin CLAUDE.md.
  • Accountability gap: Each retaliation hypothesis tagged [review] where incentive grounding is thin.
  • Analytical Rigor: MECE competitor coverage; falsifiable retaliation levers named, not "they'll react."
  • Incentive Gaming: N/A — external competitor focus.
• Escalation triggers: Move threatens multiple core-position competitors simultaneously; plan only works in no-response case — flag explicitly.

Workflow

1. State the planned move precisely.
2. For each relevant competitor (use map-competitive-landscape strategic groups if available), get or build incentive structure (map-incentives) and assess: does this move threaten something they'd defend hard?
3. Classify visibility and threat level per competitor: visible + threatens core = retaliation likely; quiet + low priority = unlikely; in-between cases reasoned explicitly.
4. For every "retaliation likely" competitor, state the plausible response — specific lever (price, copycat feature, channel, PR) grounded in incentive structure and capability.
5. Stress-test the plan against predicted response: does the move still make sense if retaliation happens? If only works in no-response case, say so directly.
6. MECE/falsifiability check before output: every named competitor covered; each retaliation hypothesis states what would falsify it.

Output format

CONFIDENCE: [defensible recommendation | structured first pass]
PLANNED MOVE: [precise statement]

[Competitor] — Strategic group: [...] — Threatens their priority: [yes/no, which priority]
  Visibility: [visible/quiet] — Retaliation likelihood: [likely/unlikely/uncertain]
  If likely: Predicted response: [specific lever, grounded in their incentive structure]

[repeat per competitor]

PLAN SURVIVES PREDICTED RESPONSE: [yes/no] — [if no, what changes]

Worked example

Input: SaaS vendor plans 30% price cut on mid-market tier. Competitor A (same segment, volume-incented sales) and B (enterprise-focused, different group).

Expected output (excerpt):

CONFIDENCE: structured first pass
PLANNED MOVE: 30% price cut on mid-market tier, effective Q3

Competitor A — Strategic group: mid-market workflow — Threatens their priority: yes (core ARR segment)
  Visibility: visible — Retaliation likelihood: likely
  If likely: Predicted response: matching discount on comparable tier within 60 days [review]

Competitor B — Strategic group: enterprise suite — Threatens their priority: no
  Visibility: visible — Retaliation likelihood: unlikely

PLAN SURVIVES PREDICTED RESPONSE: uncertain — plan assumes net win rate gain; if A matches, margin impact may erase volume benefit [review]

Quality checks before delivering

• Planned move stated with scope, scale, timing
• Every relevant competitor covered (MECE)
• Retaliation predictions grounded in incentive structure or flagged [review]
• Plan-survival test answered honestly
• Output does not read as concluded go/no-go

Propose profile update

When a stable convention surfaces during this run (thresholds, naming, tone, output format, or recurring corrections), propose a profile update: show the exact diff against ~/.claude/plugins/config/claude-for-strategy/market-intelligence/CLAUDE.md (org-wide facts go to org-profile.md), ask for confirmation, and write only on yes. Only /market-intelligence:practice-setup auto-applies a full profile write.

Outputs

Follows plugin CLAUDE.md § Outputs. Next: revise move, run map-incentives for thin competitors, or escalate to corp strategy.