Methodology

How we score tools.

The Bluebench Safety Score is a 0–100 rating across five weighted dimensions — disclosed and reproducible. We don't take money from vendors for better scores.

/ 100

Compliance

Does the tool comply with LinkedIn ToS, GDPR and platform policies? Documented evidence outweighs marketing claims.

Weight · 25%

Data privacy

How securely are user data and LinkedIn cookies handled? Hosting region, certifications, data deletion.

Weight · 20%

Sender reputation

How high is the ban risk for your account? From architecture, documented limits and publicly reported suspensions.

Weight · 25%

Transparency

Clearly documented pricing, limits and roadmap. No hidden data flows.

Weight · 15%

Support

Response time, documentation and account-recovery help when things go wrong.

Weight · 15%

The formula — open and reproducible

score = compliance · 0.25 + dataPrivacy · 0.20 + sender · 0.25 + transparency · 0.15 + support · 0.15

Each dimension is scored 0–100, then weighted, summed and rounded. Example, Expandi:

65·0.25 + 78·0.20 + 70·0.25 + 80·0.15 + 68·0.15 = 71,55 → 72

Score bands

What a number means

Solid — Low ban risk, well documented.

Usable with care — Works, but mind limits & setup.

Elevated risk — Only with clear guardrails.

High risk — Not recommended for most teams.

Evidence

What goes into each dimension

Compliance · 25%

ToS-compliant usage (no headless/scraping violation)
Documented GDPR compliance (DPA available)
Public policy evidence instead of ad claims

Data privacy · 20%

Hosting region (EU preferred)
Certifications (SOC-2, ISO-27001)
Session-cookie handling & data deletion

Sender reputation · 25%

Cloud vs. local / dedicated IP
Sending limits & warm-up mechanics
Publicly reported suspensions & community reports

Transparency · 15%

Clear, public pricing
Documented limit caps
Changelog/roadmap, no hidden data flows

Support · 15%

Response time & reachable channels
Depth of documentation
Account-recovery assistance

When does a score change?

Continuously — when new compliance evidence surfaces, suspensions are reported, or vendors change pricing/policies. Bigger movements are logged in the news feed so every change stays traceable.

Native LinkedIn products

Sales Navigator and Recruiter receive 100/100 — compliant by definition because LinkedIn operates them itself. That's a rule of the methodology, not a placeholder.

Independence & funding

No vendor can buy a better score. Affiliate links are labelled as such (rel="sponsored") and have zero influence on score or ranking. We're funded by the buyer side — audits and consulting for teams that use tools, never by the vendors we rate.

Limits of the score

The Safety Score is a best-effort editorial risk assessment — not legal advice, not a compliance statement and no guarantee against an account ban. A score of 72 does not mean a tool is "72% permitted": whether a tool is authorised by LinkedIn is shown by the separate policy status on every tool page. Platform rules and tool behaviour change — verify safety-critical decisions with current sources (or have your stack audited directly).

Data sources & corrections

Pricing, features and certifications come from vendor sources (website, docs, public reports); the assessment of architecture, risk and transparency is our own editorial desk research. Hands-on tests will be explicitly labelled as such. Every tool page shows review status, review date and the sources used. Vendors can submit corrections with evidence to info@bluebench.io — we verify and update with a new review date. Paid partnerships (affiliate) influence neither assessment nor score.

This methodology applied to your stack

We audit the tools you use by exactly this score — with a written risk sheet.