Vetting Creators: Brand Fit, Content Quality, and Fraud Checks

Why vetting matters (and what “good” looks like)

Vetting is the quality-control step between a promising shortlist and a paid partnership. The goal is to confirm three things: brand fit (their style and values align with yours), content quality (they can reliably produce the format you need), and audience integrity (their reach and engagement are real and relevant). A creator can look perfect on paper but still be a poor match if their tone clashes with your brand voice, their community is hostile, or their metrics show signs of manipulation.

A structured vetting workflow (repeatable for every creator)

Use the same process each time so decisions are consistent across team members and campaigns. The workflow below is designed to be completed in 20–45 minutes per creator for an initial pass, then deepened for finalists.

Step 1: Review recent posts for brand fit (last 30–90 days)

Scan a creator’s most recent content to answer: “Would our brand feel natural here?” Focus on what they post now, not their best-performing post from two years ago.

• Tone and voice: Is it educational, comedic, edgy, sarcastic, luxury, minimalist, chaotic? Flag anything that conflicts with your brand safety standards (e.g., aggressive language, polarizing takes, excessive profanity if that’s a risk for your category).
• Values and themes: Look for recurring topics (body image, dieting, politics, gambling, alcohol, medical claims, etc.). Even if not “wrong,” some themes may be incompatible with your brand positioning.
• Visual identity: Lighting, framing, editing style, color palette, and on-screen behavior. Ask: can your product be integrated without looking out of place?
• Past sponsorship behavior: Are ads clearly disclosed? Are they over-saturated with back-to-back sponsorships? Do they promote direct competitors frequently?

Practical scoring tip: Rate brand fit on a 1–5 scale across tone, values, visual style, and sponsorship compatibility. Require a minimum average score (e.g., 3.5/5) to move forward.

Step 2: Evaluate content quality and consistency

Quality isn’t just “pretty.” It’s whether the creator can reliably deliver the specific assets you need (UGC-style demos, tutorials, reviews, lifestyle integration, etc.).

Continue in our app.

You can listen to the audiobook with the screen off, receive a free certificate for this course, and also have access to 5,000 other free online courses.

Or continue reading below...

Download the app

• Hook and retention signals: Do videos start strong? Are intros tight? Do they use clear structure (problem → demo → result)?
• Audio and clarity: Clean audio, readable captions (if used), stable framing, and understandable pacing.
• Storytelling and credibility: Do they explain why something matters, show proof, and avoid exaggerated claims?
• Consistency: Look for steady posting cadence and stable quality. One viral post doesn’t equal reliable performance.

Quick check: Pick 10 recent posts and note format mix (tutorial, vlog, review), average engagement, and whether the creator repeats the same concept too often.

Step 3: Community interactions and audience sentiment

A creator’s comments section is a live focus group. You’re checking whether the audience trusts them and whether the community culture is healthy.

• Creator behavior: Do they respond respectfully? Do they answer questions? Do they pin helpful comments?
• Audience sentiment: Are comments supportive, curious, and specific? Or are they hostile, spammy, or filled with skepticism?
• Trust indicators: Look for comments like “I tried this because of you” or “Your recs always work.” These suggest influence beyond vanity metrics.
• Ad reaction: On sponsored posts, do followers complain about “too many ads” or accuse the creator of being inauthentic?

Practical method: For 3–5 recent posts (including at least one sponsored post), categorize the top 50 comments into: positive, neutral, negative, spam/bot. If spam/bot is high or negativity dominates, flag for deeper review.

Step 4: Comment authenticity checks (spotting bots and pods)

Authentic comments tend to reference specifics from the content. Inauthentic engagement often looks repetitive, generic, or oddly timed.

• Bot-like patterns: Repeated short phrases (“So good!”, “Nice video!”, “Love it!”) across many posts, often from accounts with no profile photo, no posts, or random usernames.
• Engagement pod signals: The same small group of accounts commenting immediately on every post with overly enthusiastic, generic praise. Pods can inflate perceived engagement without real audience interest.
• Mismatch between comments and content: Comments that don’t match what was shown (e.g., praising a “recipe” on a non-food video).

Practical check: Click 10 commenters on a post. If many profiles look empty, newly created, or follow thousands while having few followers, treat it as a risk signal (not automatic disqualification, but escalate).

Basic fraud detection: what to look for in performance and growth

Fraud detection is about pattern recognition. You’re not proving intent; you’re identifying risk. Use multiple signals before making a decision.

1) Suspicious follower growth spikes

Sudden jumps can happen (viral content, press, platform features), but suspicious spikes often lack a clear content reason.

• Red flag: Large follower increase with no corresponding high-performing posts.
• What to do: Ask what drove the growth (viral post link, feature, collaboration). If they can’t explain, treat as higher risk.

2) Low view-to-follower ratios (or inconsistent reach)

A large follower count with consistently low views can indicate inactive followers, purchased followers, or a content-audience mismatch.

• Red flag: Very low average views relative to follower count across many posts, especially if engagement looks artificially “okay.”
• What to do: Compare the last 10–20 posts. If most posts underperform heavily and only occasional spikes occur without clear cause, flag for deeper review.

3) Repetitive bot comments and unnatural engagement

Bot engagement often appears as repetitive, generic comments and suspiciously uniform likes/comments across posts.

• Red flag: Many comments that are short, generic, and posted within minutes of publishing.
• What to do: Sample commenter profiles and look for patterns (empty accounts, same commenters across posts, identical phrasing).

4) Engagement pods and “too-perfect” early engagement

Pods can create a burst of early engagement that looks strong but doesn’t translate into real conversions.

• Red flag: The same accounts appear immediately in the first 10–20 comments on every post.
• What to do: Check whether those accounts are creators themselves and whether they comment similarly on each other’s posts.

5) Recycled content and repost networks

Some creators rely heavily on reposting others’ clips or recycling the same content with minor edits. This can create copyright risk and reduce brand differentiation.

• Red flag: Frequent “borrowed” clips without clear credit, watermarks from other accounts, or repeated reposts across platforms.
• What to do: Ask for confirmation they own the content and can grant usage rights for your campaign (especially if you plan to whitelist or run paid amplification).

6) Inconsistent performance with no narrative explanation

All creators have variance, but extreme swings can signal manipulation or unstable audience interest.

• Red flag: Alternating very high and very low performance with no change in content quality, topic, or format.
• What to do: Ask for platform insights for the last 30–90 days to confirm reach sources and audience composition.

Verifying audience demographics (when available)

Demographics are often self-reported by platforms and can be incomplete, but they’re still useful for screening. Your goal is to confirm the creator’s audience matches your target market by country/region, age range, and gender split (as relevant), plus any category-specific needs (e.g., language, household income proxies, interests).

What to request (simple, standard, and non-invasive)

• Media kit (PDF or deck) with audience breakdown, top locations, age, gender, and past brand work.
• Platform insights screenshots for the last 30 days (or 90 days for more stability), including:
  • Audience top countries/cities
  • Age ranges and gender split
  • Reach and impressions (where available)
  • Content performance overview (views, watch time/retention if available)
  • Traffic sources (For You vs Following, Explore, etc., if the platform provides it)

How to ask (copy/paste template):

To confirm fit, could you share either your media kit or screenshots of your platform insights for the last 30–90 days? Specifically: audience top locations, age ranges, gender split, and a recent content performance overview. Screenshots are perfect—no need to share any personal info.

How to sanity-check demographics

• Language and location alignment: If content is in English but top audience is non-English regions, ask why (international reach is fine, but it should match your distribution needs).
• Audience mismatch vs comments: If insights claim one region but comments show another language heavily, investigate further.
• Too-perfect targeting: Extremely “ideal” demographics (e.g., 95% in one premium market) can be real, but if paired with other fraud signals, escalate.

Media kits and insights: what “good” documentation includes

A strong media kit or insights package should make it easy to predict collaboration fit and reduce surprises.

Approval checklist (use this before contracting)

Use this checklist to approve, reject, or escalate. Keep it as a shared internal doc so decisions are auditable.

Brand fit

• Recent content tone aligns with brand voice
• No recurring themes that conflict with brand safety standards
• Sponsored content style feels authentic (not overly scripted or misleading)
• Competitor conflicts are acceptable or manageable

Content quality

• Can produce the required format (demo, tutorial, review, lifestyle integration)
• Consistent quality across recent posts
• Clear communication style; credible explanations; no exaggerated claims

Community and sentiment

• Comment sections show trust and genuine conversation
• Creator engages respectfully and handles criticism well
• Sponsored posts do not trigger strong negative backlash

Fraud and integrity checks

• No unexplained follower growth spikes
• View-to-follower ratio and engagement patterns appear plausible
• No heavy bot-comment patterns or obvious engagement pod behavior
• No signs of recycled/copyright-risk content that would affect usage rights
• Performance variability is explainable (format changes, seasonality, platform shifts)

Demographics verification (if required for the campaign)

• Media kit or platform insights screenshots received
• Top locations match campaign targeting
• Age/gender split aligns with product audience (where relevant)
• Data timeframe is recent (30–90 days)

Escalation rules (sensitive categories, compliance, and reputation risk)

Some creators should not be approved by a single reviewer. Set escalation rules so higher-risk decisions get additional scrutiny from legal, compliance, or brand leadership.

Escalate to legal/compliance when:

• Category involves regulated claims (e.g., health, wellness, supplements, finance, insurance, medical devices, alcohol, gambling, or products marketed to minors)
• Creator has a pattern of making strong claims without evidence (before/after promises, “guaranteed results,” medical or financial advice)
• Disclosure practices appear inconsistent or unclear on sponsored posts
• You plan to reuse content in paid ads (usage rights, music licensing, third-party footage concerns)

Escalate to brand/reputation stakeholders when:

• Creator posts polarizing content that could trigger backlash (even if unrelated to your product)
• Comment sections show frequent controversy, harassment, or hate speech
• Creator has recent public disputes, callouts, or negative press that could resurface
• Audience sentiment is mixed and the partnership could appear inauthentic

Escalate to analytics/performance review when:

• Multiple fraud signals appear (growth spikes + bot comments + inconsistent reach)
• Demographics data is missing or conflicts with observed audience behavior
• Creator cannot provide basic insights documentation for a campaign where targeting is critical

Decision rule: If you see one mild risk signal, request clarification. If you see two or more meaningful risk signals across different areas (e.g., authenticity + compliance), pause approval and escalate.