Provider provenance
Performance snapshots retain the provider, channel, provider post ID and capture time so observed metrics have an identifiable source.
BoosterStar's analytics layer deliberately separates publishing cadence from performance. Reach, views, reactions, comments, shares, saves and clicks only enter performance analysis when a connected provider returns those values.
A social dashboard can look impressive while quietly converting activity into pseudo-performance. Number of posts published is useful operational data, but it is not reach, engagement or proof that an algorithm rewarded a creative choice.
Built for execution that remains understandable when the team, campaign volume or number of connected channels grows.
Performance snapshots retain the provider, channel, provider post ID and capture time so observed metrics have an identifiable source.
If a provider does not return a metric, BoosterStar keeps it null. It is not replaced with zero and is never estimated from publication counts.
Recommendations state how many distinct posts support the selected metric and whether that sample is too thin for a stable pattern claim.
The output is a concrete next test—such as varying a hook, format or CTA while keeping the comparison metric consistent—not a generic growth score.
A practical operating sequence you can test with a real campaign rather than a feature demo.
Authorise the social accounts for which BoosterStar has real provider integration and required permissions.
Successful sends retain provider post IDs so later telemetry can be associated with the correct internal post and channel.
Collect the metrics each provider actually exposes. Different networks can provide different metric sets and permissions.
Use a metric with enough distinct-post evidence, inspect the strongest observation and run a controlled content variation.
Identify which observed posts deserve a closer look without pretending that correlation proves why they performed differently.
Explain the source and evidence behind a recommendation instead of relying on opaque composite scores.
Keep one comparison metric stable while testing a hook, format or CTA, then add new provider observations to the evidence set.
No. Missing provider metrics remain missing. Publishing volume is never converted into reach or engagement.
Confidence reflects the number of distinct posts carrying the selected provider-observed metric. It describes sample strength, not causal certainty.
Networks expose different analytics fields, scopes and eligibility rules. BoosterStar preserves those differences rather than forcing every provider into a fabricated common metric.
Start with a real campaign, keep the final editorial decision human, and let provider evidence make the next brief smarter.