Type your TrustScore into a plain average calculator and the numbers won't reconcile. That's because TrustScore is not an average — it's a Bayesian weighted score with time decay, and the differences decide how fast a profile can recover.
The three mechanics that matter
- A Bayesian prior. Every profile starts from a neutral anchor (historically seven 3.5-star 'phantom' reviews). With few reviews, the anchor dominates — which is why a brand-new profile with three 5-star reviews doesn't show 5.0.
- Time decay. Older reviews progressively lose weight. A 1-star from two years ago drags far less than one from last week.
- Recency emphasis. Fresh reviews carry the most weight, so the displayed score responds fastest to what's happening now.
Why bursts backfire anyway
If recency is weighted, why not dump 50 reviews in a week? Because the same recency machinery makes anomalies obvious. Trustpilot's fraud systems model what normal inflow looks like for a business your size; a spike disconnected from real customer activity is the classic trigger for investigation and removal. Pacing is not a compliance nicety — it's what makes the gains stick.
Estimating your own target
Start with the straight-average math to get a conservative baseline: our free TrustScore Calculator converts your current average, review count and target into a review volume. On Trustpilot the true requirement is usually lower thanks to recency weighting — treat the calculator's output as your ceiling, then let the Review Velocity Planner spread it across a natural window.
What a healthy target looks like
Aiming for 5.0 is a mistake twice over: it's mathematically unreachable once any lower review exists, and buyers distrust perfect-but-thin profiles. The conversion sweet spot across industries sits around 4.4–4.7 with visible volume — credible, excellent, human.