# Campaign Strategy
Turn an existing market or customer-research brief into a focused campaign plan. This method produces a recommendation and a ready-to-review plan; it does not publish, contact prospects, change advertising spend, or write to a host system.
Inputs
Use a supplied research brief, ICP artifact, or a concise written brief. It needs a target market, buyer and buying context, competitive alternatives, differentiation hypothesis, capacity, and relevant constraints. Ask only for the missing facts that would change channel selection. If research is thin, label the plan provisional rather than filling gaps with assumptions.
For external claims, retain the source URL beside the claim and include a Sources section. Recheck each citation before presenting the plan.
Workflow
1. **Recap the market and ICP.** State firmographics, triggers, jobs to be done, buying constraints, where buyers seek information, exclusions, and evidence gaps. Test whether the segment is distinctive, accessible, measurable, and economically plausible. 2. **Set positioning.** Identify alternatives, distinctive attributes, value themes, best-fit customers, and the market category. State one positioning statement and three to five evidence-linked differentiation pillars. 3. **Build the messaging stack.** For each meaningful buyer archetype, name the push, pull, anxieties, and existing habit; then provide a value proposition, proof requirements, and likely objections. 4. **Choose channels from buyer behavior.** Select one primary channel and no more than two supporting channels. Explain why each fits the buying motion; do not default to search, paid social, or content simply because they are familiar. Map the chosen channels to the buying jobs they cover and flag gaps. 5. **Define the activation.** When named accounts are appropriate, build a small, sourced account list with fit, dated signal, tier, contact angle, and research date. Omit accounts that cannot meet the evidence bar. Specify the offer, assets, sequence, owner, capacity, budget guardrail, and measurement needed for each channel. 6. **Make the plan testable.** Set a practical calendar, leading and outcome metrics, decision dates, stop/continue/change criteria, and the one next action. Use a 0–30 day discovery and asset-building phase, a 31–60 day execution and iteration phase, and a 61–90 day evidence and adjustment phase unless the supplied motion calls for another cadence. Reconcile effort and spending with stated limits.
Named-account research and activation discipline
Use named-account research only when the motion and capacity justify it. Define the inclusion rule first, then record public source URLs, access dates, fit evidence, signal status, tier, and a proposed angle. Mark unavailable, conflicting, or deferred signals rather than manufacturing a contact or trigger. Use Tier 1 for roughly 5–25 strategic accounts, Tier 2 for roughly 25–200 clustered accounts, and a broader programmatic tier only when the evidence and capacity support it. Stamp the account list with its last-researched date. An activation calendar must name an owner, effort, dependency, hypothesis, and review point for each action. It proposes activity only; it does not contact, publish, spend, or establish consent.
Output format
```markdown # Campaign Plan: [market or offer]
## ICP and market recap ## Positioning and differentiation ## Messaging by buyer archetype ## Channel plan and rationale ## Activation calendar ## Metrics and decision rules ## Assumptions and evidence gaps ## Sources ```
Quality rules
- A channel is a recommendation only when its rationale connects to buyer behavior and evidence. - Separate sourced facts, inferences, and hypotheses. - Keep account research reproducible; do not fabricate signals or contacts. - Treat capacity, approval, privacy, and budget constraints as design inputs. - Propose host-system updates or execution steps separately; never imply they have happened. - State each metric's baseline, denominator, as-of date, measurement method, review cadence, and continue/change/stop rule; a forecast is not a result.