How to Build and Document a Creator Shortlist

What matters most

Use one canonical creator identity connected to every known platform account, representative, and past brand relationship. This prevents double-counting and fragmented communication.

Compare candidates by campaign role rather than one universal score. A reach creator, technical explainer, community host, conversion partner, and reusable-content producer contribute different value.

Show evidence and uncertainty beside the recommendation. Decision-makers should be able to understand why a candidate is present, which assumptions remain unverified, and what must happen next.

A practical workflow

1. 01

   Deduplicate candidate identities and attach current handles and representatives.

2. 02

   Assign each candidate a campaign role and supporting evidence.

3. 03

   Summarize relevance, audience, performance, safety, conflict, cost, and uncertainty.

4. 04

   Create primary, alternate, hold, and reject states with reason codes.

5. 05

   Assign owners and next actions, then freeze the reviewed version for auditability.

A shortlist is a recommendation, not a menu

A list of twenty creators with evidence attached, handed to a stakeholder to choose from, moves the decision to the person with the least context. It also invites selection on the one dimension anybody can read without doing the work, which is follower count.

Write the recommendation. State which creators you propose, in what roles, at what budget split, and why that combination rather than another. Put the alternatives underneath as alternatives, with the reason each is second choice. The evidence is there to support the recommendation and to let somebody disagree with it specifically.

This also changes what the review conversation is about. A menu produces a discussion of individual names driven by whoever recognises one. A recommendation produces a discussion of the reasoning, which is the conversation worth having and the one that improves the next shortlist.

Size it from the funnel, backwards

Shortlist length is usually chosen by convention and should be derived from the funnel. Work back from the number of creators you need signed, through the proportion that will decline or be unavailable, the proportion priced beyond the budget, and the proportion that will fail vetting or contracting.

Those rates are specific to your team, your category and your budget, and after two or three campaigns you will know them. Until then, assume attrition at every stage is larger than it feels: availability in the window is the constraint that most often removes candidates, and it is invisible until you ask.

The reason this matters is that an undersized shortlist fails late. Discovering three weeks before launch that you have two creators and needed six means either restarting the search under time pressure or accepting whoever is available, which is how campaigns end up with creators nobody chose.

The rejected column is what you will reuse

Most teams delete the creators they declined, which means the next campaign repeats the same research and reaches the same conclusions. Keep them, with a reason code and a date, and the second campaign starts from a materially better position than the first.

Reason codes matter more than free text because they are what makes the record queryable later. Distinguish not relevant, audience mismatch, unavailable in window, over budget, conflict, failed vetting, no response and declined. Several of those are temporary: unavailable and over budget both change, and a creator who was wrong for a launch campaign may be right for an always-on program.

Freeze the version that was reviewed and approved. Shortlists get edited after decisions are taken, and a document that has quietly changed since sign-off cannot answer the question of what was actually agreed. A dated snapshot costs nothing and is the only thing that can.

Common mistakes

• Ranking everyone by one opaque composite score.
• Presenting conclusions without source content or dates.
• Deleting rejected candidates and losing the reason.
• Starting outreach before ownership and priority are agreed.

Working checklist

• Each creator has one canonical profile.
• Campaign role and evidence are explicit.
• Fit, risk, commercial status, and uncertainty are visible.
• Alternates and rejection reasons are retained.
• Every active candidate has an owner and next action.

Questions and answers

How many creators belong on a shortlist?

Derive it from the funnel rather than convention: the number you need signed, divided by the proportion that survives outreach, availability, pricing, vetting and contracting. Until you know your own rates, plan for substantially more attrition than feels necessary, because availability in a specific window removes more candidates than anything else and does so late.

Should stakeholders see the rejected candidates?

Show the reason codes, not the findings. Knowing that eleven creators were assessed and declined for stated reasons demonstrates the search was thorough and pre-empts the question about a specific name. Circulating detailed vetting conclusions about individuals more widely than the decision requires creates exposure and helps nobody.

What format should a shortlist take?

Whatever holds one row per creator, one column per criterion, and a coverage indicator showing which fields were measured. The requirements are that it is comparable across rows, that every conclusion carries a source and a date, and that a version can be frozen at sign-off. A spreadsheet meets all three.

How long does a shortlist stay valid?

Weeks rather than months for availability and pricing, which change constantly, and longer for relevance and audience fit, which move slowly. Reusing a six-month-old shortlist without re-checking availability and conflicts is the common failure. Keep the research, re-verify the perishable fields.