Example Prompts
- “Find top 20 AI agent influencers on Twitter”
- “Find micro-influencers for sustainable fashion”
- “Who are the most influential voices in cybersecurity on Twitter?”
- “Discover crypto influencers with high engagement and low bot probability”
Scoring Formula
| Factor | Weight | Description |
|---|---|---|
| Relevance | 30% | How closely the influencer’s content matches the target niche |
| Engagement | 30% | Like, retweet, and reply rates relative to follower count |
| Reach | 20% | Follower count and impression volume |
| Authenticity | 10% | Bot probability score (lower is better, via isInauthenticProbScore) |
| Consistency | 10% | Regular posting cadence in the target niche |
Influencer Tiers
| Tier | Follower Range |
|---|---|
| Mega | 1M+ |
| Macro | 100K - 1M |
| Micro | 10K - 100K |
| Nano | 1K - 10K |
Voice Types
Each influencer is classified into a voice type based on their content style:- Analyst: Data-driven commentary and research
- Builder: Hands-on creators sharing their work
- Educator: Teaching and explaining concepts
- News: Breaking news and industry updates
- Commentator: Opinion and thought leadership
- Community: Community building and curation
Xpoz Tools Used
| Tool | Purpose |
|---|---|
getTwitterUsersByKeywords | Find users relevant to the niche (returns relevantTweetsCount, relevantTweetsLikesSum, relevantTweetsImpressionsSum, isInauthenticProbScore) |
getTwitterPostsByAuthor | Analyze recent posts from candidate influencers |