Analytics: reading attribution, relevance, and account health
Every metric SkyFollowing tracks and how to read it: follow-back attribution, relevance buckets, source breakdowns, and printable reports.
SkyFollowing measures growth by cohort and by campaign, so every number answers a decision: which targeting deserves budget, which account needs attention, and whether pace increases are earning their keep. This page defines each metric the way the dashboard computes it.
The core metrics
| Metric | Definition |
|---|---|
| Queued | Candidates that passed filters and scoring, waiting to be followed |
| Followed | Follows executed, including those that later followed back |
| Followed back | Followed accounts that followed you back |
| Follow-back rate | Followed back divided by followed, per campaign, as a percentage |
| Unfollowed | Cleaned up after the wait window with no follow-back and no engagement |
Relevance buckets
Each campaign's audience view groups followed accounts by their AI relevance score and shows the follow-back rate per bucket. This is the fastest way to tune your minimum score: when Fair converts far below Strong, raising the threshold buys quality at little volume cost.
| Bucket | AI score | How to read it |
|---|---|---|
| Excellent | 85-100 | Near-certain niche fit; protect the sources producing these |
| Strong | 70-84 | The workhorse band for most campaigns |
| Fair | 50-69 | Acceptable when volume matters; watch its conversion closely |
| Weak | 0-49 | Only present if your minimum score allows it; usually should not |
Sources and the queue
The audience view also breaks follows down by source (which keyword or seed handle produced them) with a follow-back rate per source, and previews the highest-scoring profiles waiting in the queue. A source converting well below the campaign average is the first thing to cut; the queue preview tells you whether the research engine still has quality left to work with.
Timelines and the live feed
Campaigns chart daily follows against follow-backs over the last 30 days, so you can watch a cohort convert through its wait window. The activity feed streams every event as it happens: follows, follow-backs, unfollows, research runs, pauses, and errors. The same event stream powers webhooks, so anything you can see in the feed can land in Slack.
Account-level health
Beyond campaigns, every connected account tracks follower count, health score (100 minus its ban-risk score), warmup progress, and daily budget usage. Workspace analytics roll all of it up with 30-day series and a per-campaign performance board ranked by follows and conversion.
Before you promise a number (to a client or to yourself), model it: pace times follow-back rate over a period. Then let attribution tell you the real figure.
Reports
The Reports page assembles workspace results into a clean, printable summary: growth deltas, campaign performance, and account health. Agencies use it as the client-facing artifact, with the analyst role as the view-first seat for exactly this.
Frequently asked questions
Why does my follow-back rate differ from followers gained?
Follow-back rate only counts people your campaigns followed. Organic followers, mentions, and discovery all add followers without touching campaign attribution, which is exactly why the metric stays trustworthy.
How fresh is the activity feed?
It streams in real time. Follows, follow-backs, and safety events appear as the workers record them.
Can clients see analytics without touching settings?
Yes. The analyst role gives clients dashboards, reports, and the live feed without workspace-management rights, and by convention campaign controls stay with your team.
Every setting on this page ships with safe defaults. Free for 7 days, no card required.