Influencer Discovery Brief

Influencer Discovery Brief

Define the sourcing criteria and produce the discovery brief a partnership manager uses to shortlist candidate creators - who qualifies, by what weighted criteria, and with what evidence.

The "3 Rs" (Reach, Relevance, Resonance) is a genuinely recurring industry convention for judging influence, but research could not trace it to a canonical originator - treat it as a floating mnemonic, not a citable method. A "4 Rs" variant adding "Reference" is one influencer-analytics vendor's framework, not an industry standard. This skill therefore works from explicit weighted criteria the manager approves, never from a named model presented as authoritative.

Interview

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

• Is this a B2C brand campaign or a B2B creator/thought-leader program? (Or both - then weight criteria separately per track.)
• What is the brand/product, and what is the value prop a creator's _audience_ would actually care about?
• What is the campaign objective: awareness, conversion, content-for-paid, or trust-building? The objective changes the criteria weights, so answer before any scoring.
• Who is the target _customer_ - demographics, role, buying context? (Not the target creator; the creator is a channel to this person.)
• Which platform(s) does the campaign run on?
• Which geographies/markets actually convert for this product? (Audience anywhere else is reach you pay for and can't use.)
• What is the budget tier, and is compensation cash, product, or a mix?
• How many creators will actually be contracted from the shortlist?
• What are the brand-safety redlines, and is the product in a regulated category (alcohol, health, finance, gambling, kids)?
• Any competitor or category-exclusivity conflicts to enforce? Who is already on the roster? (Needed for audience-overlap checks.)
• What audience data can you actually obtain: creator discovery platform exports, audience-insights exports, media kits, or public profiles only? (A licensed discovery platform moves its export pass from the slowest sourcing option to the fastest. Public profiles only deletes the two-stage weighting gate from the menu.)
• By what date must the shortlist land, and what is waiting on it? (A date inside a week promotes the roster pull, the customer-mention pass, and the floor-gate weighting. A quarter makes social listening and full objective-led scoring affordable.)
• One-off campaign or a creator program you will run again? (A one-off stops at a shortlist. A repeating program justifies the full objective-led scoring pass, whose approved weights are the only output the post-campaign fit-rate KPI can tune.)
• What is the effort ceiling - your own hours, whose data you can request, and whether you can commit to a standing monitoring job? (A solo ceiling deletes social listening. No budget owner for a platform licence deletes discovery-platform exports.)

Carry these last three answers into every ranking below. They are the only inputs that reorder the defaults for this manager.

Every ordering in this skill is a default, not a law. It ranks by qualified candidates returned per unit of effort, which is never the same ordering as cheapest-first, and it shifts with context and with who runs the pass.

Re-rank against what you already know about this manager before recommending anything. Say which answer moved which pass, so the reordering is auditable rather than a preference you asserted.

• A discovery platform already licensed moves its export pass to the front.
• An existing creator roster with performance history makes the roster pull the strongest pass rather than merely the cheapest.
• A niche too small for platform databases deletes the export pass outright and promotes customer mentions and social listening in its place.

Workflow

1. Run the Interview; collect every answer before looking at a single candidate.
2. Translate the objective into a draft weighted-criteria set - awareness, conversion, content-for-paid, and trust-building each weight differently. See references/sourcing-criteria-and-weighting.md.
3. Enter brainstorming mode: present the weighting approaches in ranked order, lead with the recommendation, and give each one's trade-off in the manager's own vocabulary (criteria, floors, knockouts, shortlist). Full trade-offs in references/sourcing-criteria-and-weighting.md.

• efficiency: two-stage audience-fit gate > objective-led fixed weights > floor-gate plus equal weights, wherever one audience-fit figure can be obtained per candidate; floor-gate leads and the other two are unaffordable where the only data is public profiles.

Default to the two-stage gate. Drop to the floor-gate when the data answer was public profiles only, or the date answer leaves under a week. Pay the full objective-led scoring pass at least once when the answer was a repeating program - it is the only approach that produces tunable weights.

1. Ask brainstorming questions one at a time; score nothing until the manager explicitly picks a weighting approach.
2. Build the raw candidate pool through sourcing passes. Six passes exist; treat them as a budget, not a checklist. Run them in this order and stop once the pool is deep enough to survive the knockouts.

• Existing roster and CRM - an hour, no tool access. The only pass returning performance history with _your_ brand: who already posted, what it did, who the manager would sign again. It returns no net-new names, and step 9's overlap check needs it regardless.
• Customer and community mentions - a day pulling support and community threads, review text, and a "which creators do you follow" question into a survey already going out. Names arrive with fit evidence attached, because your buyers demonstrably watch them, and this is the only pass reaching podcasts, newsletters, and closed communities no platform indexes.
• Discovery-platform exports - near-zero with a licence in hand: one filtered query returns breadth plus the audience-demographic export step 8 scores against. Without one, a quarter and a procurement conversation. Covers only what the vendor indexed - thin in narrow B2B niches and on non-social formats.
• Competitor collaborations - an hour browsing competitors' tagged and disclosed posts. Pre-qualified: they take sponsorships, they are in category, and their sponsored-versus-organic engagement is visible before you make contact. Every competitor works the same list, and the exclusivity knockout removes part of it.
• Niche keyword and hashtag search - an hour that keeps being an hour, no tool access. It mostly re-derives by hand what the export returns, at a low qualified-per-hour rate; it earns its place only where no platform covers the niche.
• Social listening - a week to design queries and a standing job to maintain. It surfaces creators already talking about the category unprompted, the strongest unique-name yield of any pass, but it is a monitoring system whose discovery output is a by-product.

• effort: social listening > discovery-platform exports without a licence > customer and community mentions > competitor collaborations == hashtag search (tied: both an hour of manual browsing, no tool access, no one else's permission) > roster and CRM == discovery-platform exports with a licence (tied: both an hour spent querying data you already hold)