influencer-campaign-brief

Influencer Campaign Brief

Draft the creative brief a brand hands to one already-signed creator for one specific campaign: deliverables, key messages, guardrails, disclosure, timeline, review rounds.

Creator briefing has no named, cross-vendor-adopted framework. Published guidance is a scatter of single-vendor formulations, none adopted beyond its own publisher:

• one publisher's nine-section list
• another's five-questions framing
• a third's do/don't blocks

The one structural idea corroborated across four unrelated publishers is separating non-negotiables from preferences, so this skill is built around it. Provenance labels below mark where a claim comes from:

• corroborated: three or more independent publishers
• single-vendor: one company's published formulation
• survey: one vendor-run survey with disclosed methodology
• inference: a reasoned default, not a published formulation

Never present any formulation here as an industry standard.

Interview

Ask before drafting. One question per message; offer multiple-choice answers when possible; skip anything the user already answered.

• B2B creator program or B2C brand collab? (Drives deliverable types, approval chain, proof style.)
• Campaign goal in one sentence, with the metric? (Awareness / consideration / conversion - this also sets the freedom level.)
• Who is the signed creator, and on which platform or channel type? (Short-form video, professional network, newsletter, podcast...)
• Which deliverables does the signed deal already commit to? (This is the source of truth; the brief may specify them, never change them.)
• What are the key messages - three at most - and the proof behind each? (No claim ships without proof the brand can back.)
• Which brand guardrails and no-go topics apply?
• Do any claims need legal or regulatory sign-off before the brief goes out?
• By what date must the result land, and is it a fixed campaign moment (launch, event, seasonal date)? (A near date promotes high freedom - a framework brief is enforced through review rounds the calendar may not have.)
• One-off campaign, or the first of a relationship you intend to repeat with this creator? (Repeat intent promotes high freedom; granting freedom and revoking it at validation is what loses creators.)
• What is your effort ceiling: review rounds the signed deal includes, briefing hours available, and is there an in-house reviewer? (One round or no reviewer takes low freedom off the menu entirely.)
• Will the content be run as paid media? (Decides audio licensing and, in B2B, the thought-leader-ad mechanics - must be settled now, not later.)
• What tracking mechanics will the creator use: promo code, tracked link, dedicated landing page?
• Who reviews drafts, and who signs off?

Workflow

1. Run the Interview; collect every answer before writing a word.
2. Reconcile the deliverables against the signed deal. On any contradiction: stop, flag it to the user, and never silently rewrite - the deal wins. Renegotiation belongs to the negotiation skill.
3. Enter brainstorming mode for the message angle (next section). Write no brief until the user picks an angle.
4. Set the creative-freedom level from the ranked default below - stated in the brief, never accidental.
5. Draft from references/brief-template.md, checking each section against references/brief-anatomy.md. Keep the brief to one or two pages.
6. Specify every deliverable to the precision fields in references/deliverable-specs.md. If paid amplification is possible, declare it in the brief now - the audio-licensing trap.
7. Write the disclosure section, covering required wording, placement, the platform branded-content tool, and who verifies after posting. Route the pre-publication content check itself to the compliance skill.
8. Bound the review rounds to exactly what the signed deal includes, and state that count in the brief.
9. Confirm the tracking mechanics exist and are tested before the posting window opens.
10. Validate the brief with the user section by section. Run the Quality gate. Iterate until every item passes.
11. Hand off the finished brief with a one-line reminder of who verifies disclosure post-live.
12. If your harness has persistent memory, memorize the approved angle, non-negotiables, freedom level, and (once known) revisions used and outcome - the next brief for the same brand or creator skips most of the interview. Otherwise hand the user a short recap to keep.

Message-angle brainstorming

• Ground candidate angles in the creator's actual content and audience. If you can browse the web, review their two or three most recent pieces; otherwise ask the user for representative posts. Never derive an angle from generic brand copy.
• Propose 2-3 candidate angles, each stated as: hook, key message, proof point, the deliverable it fits best, and its trade-off. Recommend one and say why.
• Ask brainstorming questions one at a time. Draft nothing until the user explicitly picks an angle - the angle decides what gets locked and what stays free.

Non-negotiables vs preferences - the central mechanism

• Sort every brief item into exactly two buckets: non-negotiable (locked, verbatim where required) or preference (inspiration; the creator's call). Corroborated across four unrelated publishers under four different names - the single most corroborated idea in creator briefing.
• Lock the facts, leave the voice (single-vendor phrasing, clearest articulation found). Never lock the voice.
  • Facts: claims, prices, dates, specs, regulated statements, disclosure.
  • Voice: framing, hook, pacing, setting, delivery.

How much freedom to hand over - ranked

State a freedom level in the brief (single-vendor mapping: high for awareness, medium for consideration, low for conversion). The three levels are not equal-cost alternatives to pick by taste - rank them by campaign result returned per unit of briefing effort:

• efficiency (result per briefing hour): high > medium > low

• value (result bought): high > medium > low on awareness and consideration; low > medium > high on conversion, where the offer, code and CTA only land if the brief locks them

• effort (briefing hours, approval rounds, coordination): low > medium > high - low costs about a week of coordination plus the rounds it spends enforcing its framework, high costs about an hour and one round

• compliance cost (sign-off owed before the brief goes out): low > medium == high - medium and high send the same claim set to legal, and only low adds scripted framework wording that must itself be approved

• Default to high, and say so.

• Demote to medium on a first collaboration, or a creator new to the category.

• Demote to low only when the campaign metric is a redemption or signup count and the offer mechanics must land verbatim.

Delete low from the menu outright when the signed deal includes one revision round or fewer, and tell the user why: a framework brief is enforced through rounds that do not exist here. Never park it at the bottom of the list instead - a ruled-out level reappears as scope at validation.

What this order starves: low freedom is high-value and high-effort, so it loses every round and an efficiency-first habit under-briefs exactly the conversion campaigns paid to move a number. Promote it anyway when the metric is a redemption count, or when a regulated category forces script approval.

The ranking is a default, not a law - it shifts with the goal and with who executes it. Re-rank against what you already know about this user:

• a regulated category that forces script approval promotes low, whatever the goal
• a creator who has worked with this brand before promotes high one rung, because the guardrails are already internalized
• no in-house reviewer promotes high, because nobody is free to run the rounds the lower rungs consume

Freedom is also the axis creators report as scarcest (survey - one vendor-run October 2025 survey of its own creator database, n=403 creators, ±4.9% at 95% confidence): 10% report genuine creative control, 53% want more. Cite those figures only with that methodology, never attributed to a named person. Rung mechanics, lock/leave-open lists, and the ranked guardrail menu: references/freedom-and-guardrails.md.

Invocation and expected output

Typical invocations:

• "Write the campaign brief for the creator we just signed - short-form video, launch is March 12, deal includes two revision rounds."
• "Turn this signed deal memo into a creator brief: one newsletter deep-dive plus one professional-network post, conversion goal, code JULES15."

Deliver one brief document (full template in references/brief-template.md):

CREATOR CAMPAIGN BRIEF - <brand> x <creator>, <campaign name>
Campaign goal   : one sentence + the metric
Why you         : why this creator; which of their strengths the campaign leans on
Key messages    : approved angle; up to 3 messages, each with its proof point
Non-negotiables : locked facts - offer details, mandatory wording, prohibited claims
Preferences     : freedom level (high/medium/low) + what stays the creator's call

B2B and B2C

• Same for both (published guidance never segments these by business model - a weak universality signal, itself an inference):
  • the brief structure
  • the non-negotiables/preferences split
  • deliverable precision
  • the disclosure obligation
  • bounded revisions
  • the do-not-script principle
• Honest limit: most published brief guidance is B2C-shaped, though one B2B-specific source now exists (ContentGrip's briefing playbook, Ariana Sorensen): for regulated categories (fintech, healthcare, cybersecurity, enterprise software), add required reviewer names or teams, substantiation requirements, prohibited words, mandatory disclaimers, and an approval SLA, with review compartmentalized - legal reviews claims and disclaimers, product reviews accuracy, brand reviews fit. This confirms and sharpens the approval-chain claim below.
• The one sourced B2B-specific mechanical constraint (single-vendor): in thought-leader ad formats, the ad creative is the creator's own post, and edits must come from the original author. Settle paid-readiness before the organic post goes live - the post cannot be promoted without a re-post otherwise.
• Sourced (ContentGrip, regulated B2B categories): approval chains add named reviewers per function (legal, product, brand) plus an approval SLA, lengthening the review window - compartmentalized by what each reviewer checks, not stacked as one generic pass.
• Deliverable-mix effectiveness (survey - Content Marketing Institute / MarketingProfs 2025 B2B benchmark study, 980 B2B respondents of 1,186 total, fielded mid-2024): B2B marketers rate webinars (51%) and email newsletters (37%) among their most effective distribution channels, while reported podcast use fell year over year. A channel-effectiveness signal, not a pricing benchmark - the study prices nothing.
• Deliverable-mix pricing and measurability (single-vendor account - Ahrefs' Igor Gorbenko, on over $1M of creator sponsorship spend): LinkedIn and YouTube tracked cleanly; newsletter performance depended on partner-supplied data with limited visibility; podcast sponsorships were hardest to attribute to brand awareness for a name the audience did not already know, and an initial $14,000 across five podcast campaigns produced what the account calls "horrendous" results. The same account prices sponsored posts in three bands: $500-2,000 baseline, $3,000-9,000 a middle tier called not worth the cost, $10,000+ for creators with a track record and a fixed, trackable audience - its own examples being YouTube and newsletters.
• Labeled inference, not documented practice: B2B deliverables skew to newsletter slots, podcast reads, webinars, and long-form posts.
• Corroborated across three independent B2B creator-marketing publishers (Directive Consulting's Casie Akins, Walker Sands' Payal Pathak, ContentGrip's SaaS-creator coverage): personal, hands-on operating experience is the credibility layer a B2B audience actually trusts - real execution, not personality or reach - while business-outcome metrics (pipeline influence, demo requests, deal velocity) are the separate layer that measures and justifies the program afterward. Treat the two as complementary proof layers inside the same piece of content, not a strict B2B-versus-B2C swap of one proof type for the other.
• B2C: short-form video, story-format, and in-feed placements are well documented, as are regulated-category guardrails (health, beauty, food) - see references/freedom-and-guardrails.md.

Quality gate

Score the drafted brief against all ten items. Pass threshold: 10/10. Iterate - redraft and re-validate - until nothing fails.

1. Deal-consistency test: every deliverable, review-round count, and rights mention matches the signed deal exactly; any contradiction was flagged to the user, never silently rewritten.
2. Interchangeability test (single-vendor test, adopted here as the gate): if any creator could have produced this content from this brief, the brief has destroyed the audience-trust the brand paid for. It must fail for everyone but this creator.
3. Every item is explicitly non-negotiable or preference; nothing sits unlabeled between the two.
4. No locked voice: no scripted sentences outside mandatory legal wording, no camera directions, no music or cut choices, no dictated caption phrasing.
5. Every deliverable carries count, duration or length, aspect ratio (visual formats), link or code mechanism, draft deadline, and posting window.
6. Disclosure has its own section - wording, placement, platform tool, verifier - never a footnote.
7. Paid-media intent is declared if amplification is even possible; in B2B thought-leader formats, paid-readiness is settled pre-publication.
8. The freedom level is stated, starts from the high default, and any demotion names the condition that caused it.
9. Key messages number three or fewer, each with proof the brand can back; prohibited claims are listed.
10. The brief fits one to two pages and carries only the 5-10 brand rules that matter for this content type, not the whole style guide.

KPIs and measurement

• No trustworthy benchmark exists for brief quality or revision counts.
  • The largest annual industry benchmark study (315 marketers, 100 experts) measures neither.
  • A circulating "37% of creators" figure fails source verification - never quote it.
  • Any revision-count number found elsewhere is a vendor assertion.
• The brief states the campaign goal and the tracking mechanics the creator must use. Designing KPIs and attribution is the measurement skill's job - route it there.
• Benchmark against your own campaign log instead. Track per campaign:
  • revisions used vs rounds included
  • on-time draft rate
  • spec-compliance rate on first submission
  • disclosure compliance at publication
  • whether the creator re-signed
  • the campaign metric vs its goal
• If your harness has persistent memory, keep that log across sessions; otherwise hand the table to the user to maintain.

Common failure modes

Reference

• See references/brief-template.md for the complete fill-in-ready brief document.
• See references/brief-anatomy.md for the section-by-section spec with provenance labels, the approval workflow model, and revision-round and timeline guidance.
• See references/deliverable-specs.md for per-deliverable precision fields, the format spec table, and the paid-amplification audio trap.
• See references/freedom-and-guardrails.md for the lock/leave-open lists, freedom levels by goal, and legitimate vs over-restrictive guardrails.
• See references/examples.md for a complete B2C brief, a complete B2B brief, and an annotated over-scripted negative example.
• See mbfinotti/partnerships-skills@influencer-outreach for the first-touch pitch - also done by the time the deal is signed.
• See mbfinotti/partnerships-skills@influencer-negotiation-playbook for the deal terms this brief reflects and must never renegotiate.
• See mbfinotti/partnerships-skills@affiliate-disclosure-compliance for the pre-publication disclosure check on the content this brief produces.
• See mbfinotti/partnerships-skills@influencer-measurement-framework for designing the KPIs and attribution the brief's tracking section feeds.