/docs

How UpProfiler works

What we read, how we score it, and what the number means.
Docs
/docs/what
What is the conviction score?

Your conviction score is a single number from 0–100 that answers one question: does your LinkedIn profile and your recent posts tell the same, coherent story? A high score means someone landing on your profile gets one clear, convincing signal of who you are and who you help. A low score means the two halves are pulling against each other — and the algorithm (and your audience) can't tell where to place you.

It's the same diagnostic Jasmin Alić runs by eye on his 1:1 clients, turned into something you can get for free in about 90 seconds.

The flow

1
Paste your LinkedIn URL
We read only the public profile data LinkedIn shows without logging in — your headline, About, experience, skills, and up to your 15 most recent posts from the last 30 days. No login, no scraping behind a wall.
2
We read the profile and the posts
Two surfaces, one question: do they tell the same story? We extract the themes you claim and the themes you actually post about, then measure the overlap.
3
AI scores the text; the server verifies public evidence
The model applies five text dimensions worth 80 points. A separate banner pass and bounded public-evidence checks add up to 20 points for the banner, offer clarity and social proof.
4
You get a number, the gaps, and the fix
One conviction score, an eight-part breakdown, the themes you claim-but-don't-post (and vice versa), plus an AI rewrite of your headline and About to close the gap.

How the 100 points break down

The current v5-jasmin-banner rubric has five AI-scored text dimensions worth 80 points and three bounded evidence dimensions worth 20. Alignment remains the largest component, while the banner, offer and proof now affect the same number a visitor sees.

30
15
12
12
11
10
6
4
AlignmentTopical authority
0–30 pts
Do the themes your profile claims — in your headline, About, skills and experience — actually show up in what you post? One clear lane across both surfaces ranks. Scatter your topics and LinkedIn doesn't know where to place you.
ConsistencySustainable rhythm
0–15 pts
A weekly cadence you can hold all year beats a one-off burst — and staying in one lane matters as much as raw frequency. High volume scattered across unrelated topics does not earn a top score.
SpecificityConcrete > generic
0–12 pts
Named work, real numbers and niche language vs. generic platitudes. "Drove churn from 8% to 3% for 40 SaaS teams" beats "passionate, results-driven professional" every time.
VoiceCraft & thesis
0–12 pts
Do your posts read like one human with a point of view and real craft — value-up-front hooks, a hook that pays off, authority signposted naturally — or scattered engagement bait and curiosity gaps that hide the point?
CompletenessSurfaces doing their job
0–11 pts
Is your headline front-loading a specific hook so compact LinkedIn surfaces still carry meaning? Are your About and Experience substantive? Is there a clear next step?
BannerPositioning at a glance
0–10 pts
Does the public cover image communicate one clear position, credible proof and a next step? A separate vision pass grades the banner A-F, then the server converts that grade into a bounded 0-10 contribution.
Offer clarityA visible route forward
0–6 pts
Can a visitor identify the service, profile action, Featured destination or explicit call to action? The evidence is reduced to bounded yes/no signals so missing or noisy provider fields cannot invent extra points.
Social proofEvidence, not adjectives
0–4 pts
Does the public profile show recommendation evidence, skill endorsements or a concrete proof claim? These are server-verified signals, capped at four points.
/docs/banner
What about my banner?

In the current v5 rubric, the banner contributes up to 10 points to the conviction score. A separate vision pass grades positioning, proof and the next step, then the server maps the grade to a fixed point value. If a present banner cannot be analysed reliably, UpProfiler fails the score instead of silently treating it as blank. Historical v4 results and reports keep their original five-dimension methodology label.

/docs/bands
What's a good score?
85–100
Convincing
Profile and posts tell one story. Keep the rhythm.
70–84
Mostly aligned
A few claims are silent in your posts. Small fixes, big lift.
55–69
Mixed
The two halves disagree. Pick one lane and commit.
0–54
Drifting
Your profile and posts read like strangers. Start with alignment.
/docs/data
What we read — and what we don't
  • We read only public profile data LinkedIn shows without authentication. We are not affiliated with or endorsed by LinkedIn.
  • Posts: scoring looks at up to your 15 most recent posts from the last 30 days. Consistency is judged on that 30-day window, so cadence reflects what you're doing now — not your all-time best month.
  • Reposts: we score your own words. Original posts — and reposts where you added your own commentary — count. A bare repost with no commentary doesn't; that's amplifying someone else, not telling your story.
  • We don't ask for your LinkedIn password and we never log in as you.
  • Your score history is saved to your account only when you sign in. See the Privacy Policy and Cookie Policy.
Methodology FAQ

Questions about the LinkedIn profile score

Short answers to the questions people ask before they run or interpret a profile audit.

What is a LinkedIn profile score?

A LinkedIn profile score is a structured assessment of how clearly your public profile communicates who you help, what you know and what a visitor should do next. UpProfiler calls its 0-100 result a conviction score because it also checks whether your recent posts support the positioning in your profile.

How does the UpProfiler LinkedIn profile checker work?

Paste a public linkedin.com/in profile URL. UpProfiler reads public profile fields and up to 15 recent original posts from the last 30 days. The current rubric combines five text dimensions with bounded banner, offer-clarity and social-proof evidence. The result includes the score, breakdown, positioning gaps and practical fixes.

What is a good LinkedIn profile score?

On the current UpProfiler rubric, 85-100 is Convincing, 70-84 is Mostly aligned, 55-69 is Mixed and 0-54 is Drifting. Treat the band and dimension breakdown as a diagnostic, not a universal ranking of a person or their career.

What LinkedIn data does UpProfiler read?

UpProfiler reads only profile information that is publicly visible without signing in, plus recent public posts available for analysis. It does not ask for your LinkedIn password, log in as you, read private messages or post to your account.

Is UpProfiler affiliated with LinkedIn?

No. UpProfiler is an independent tool and is not affiliated with or endorsed by LinkedIn Corporation. LinkedIn members control which parts of their profiles can appear publicly through LinkedIn privacy settings.

Is the LinkedIn profile audit free?

Yes. You can run the profile check and read the result without paying or creating an account. Email delivery, monthly re-scores and product updates are separate choices shown at the result step.

Does UpProfiler score my LinkedIn banner?

Yes. In the current v5 rubric, UpProfiler grades the public banner for positioning, proof and a clear next step, and the grade contributes up to 10 points. Historical v4 reports keep the earlier text-only score and show the banner separately.

Can I get a new score later?

Yes. You can paste the profile URL again whenever the public profile or recent posts change. If you explicitly opt in to monthly re-scores, UpProfiler can check again on roughly a 28-day cycle and email the updated result.

Ready to see your number?
Free, ~90 seconds, no login required.
Score my profile →Got feedback?