A composite 0–100 risk rating, built from four independent dimensions, so brand and marketing teams can stop arguing about which metric matters and start making decisions.
The headline score sits on top of four sub-scores. Each is calculated independently, then weighted into the composite.
Real-vs-bot follower ratio, profile quality, engagement authenticity, geographic alignment.
Last 12–24 months scanned for controversies, FTC issues, brand conflicts, and tone drift.
Comment authenticity, engagement-to-reach ratio, save and share rates, anomaly signals.
Niche alignment, tone alignment, audience overlap with target customer, competitor exposure.
Lower is safer. Use these bands as a default decision framework — adjust thresholds based on your team's risk appetite and deal size.
Authenticity, content history, and brand-fit all healthy. Standard contract terms are appropriate.
One sub-score is below band. Investigate before signing deals over your standard size threshold.
Multiple sub-scores are stressed. Consider performance-based terms, smaller deal size, or pre-approval gates.
Compounding issues across multiple dimensions. Most brands should pass — the downside risk is too concentrated.
A defensible risk score combines four dimensions: audience authenticity (real vs. bot followers), content history (controversies, FTC issues, brand conflicts), engagement quality (genuine vs. inflated interaction), and brand-fit (niche, tone, audience overlap). Each dimension is scored independently, then weighted into a single 0–100 number. Perkifi exposes both the final score and the sub-scores so you can see where risk concentrates.
Lower is safer. Scores under 20 indicate low risk and are usually safe to proceed with. 20–40 is moderate — investigate before signing larger deals. 40–60 means significant risk in at least one dimension. Above 60, the creator likely has multiple compounding issues and most brands should pass.
A single number makes creators directly comparable. Without one, every pitch turns into a debate about which signal matters more — engagement rate, follower count, audience size, content tone. A composite score creates a shared decision currency the team can align on.
Default weighting prioritizes audience authenticity and content history because those are the dimensions that have caused the most damage in real-world brand-creator deals. Enterprise customers can adjust the weights to match their own risk tolerance.
Any composite score is probabilistic. The methodology is multi-signal so single-point failures rarely change the final score, but every score is shown alongside its underlying evidence so a human can sanity-check edge cases.
No — the score sits on top of a full breakdown. Audience demographics, fake-follower percentage, content scan results, brand-conflict flags, and engagement quality are all surfaced separately so reviewers can read past the headline.