The story is moving fast.
You still need the full picture.
Compare claims, explore sources, and understand where coverage differs. Newjee’s Reality Card helps your newsroom see what needs checking before publication.
Run a Reality Card
The problem
Ten tabs open. A deadline approaching. And the facts still don’t line up.
One report says the decision is final. Another says talks are ongoing. Several repeat the same claim — but you still need to find where it came from.
Before you can write the story, you have to untangle the coverage.
- Conflicting accounts. Different figures, timelines, and details to reconcile.
- Unclear sourcing. A claim appears everywhere, but repetition isn’t independent confirmation.
- Framing mixed with facts. Someone else’s interpretation can slip into your reporting.
Every hour spent piecing this together leaves less time for interviews and investigation.
The solution
One link. Every claim checked.
Paste the article. About fifteen seconds later the Reality Card hands back the work you were about to do by hand — and every answer in it is traceable to the source it came from.
Every claim, matched against the rest of the coverage.
Each checkable assertion is pulled out of the article and tagged verified, contradicted or unverifiable, with the count of sources behind it. Where outlets disagree, the card names both accounts instead of averaging them into one.
The chain, not the echo.
Repetition is counted as repetition. Deep source traces follow a claim back towards who reported it first, and every source the analysis consulted is linked, so you can read the primary material yourself.
The language, marked where it leans.
Bias direction and intensity, sensationalism, and six manipulation indicators — loaded language, omission, false equivalence and the rest — surfaced as verbatim spans you can see before they carry into your copy.
A real card
Click through a finished one.
The actual viewer, not a screenshot. A story where the sources genuinely disagree, so you can see the conflict handling rather than take it on faith.
Iran Protests: Multi-Source Analysis
Coverage of the Iran protests shows moderate reliability with significant variance across sources. While major facts about protest locations and government responses are consistently reported, casualty figures and protest scale estimates vary considerably. Western and Iranian state media show predictable framing differences that readers should account for.
Protests have occurred in multiple major Iranian cities including Tehran, Isfahan, and Shiraz
Iranian government has implemented internet restrictions during protest periods
Security forces have used crowd control measures including tear gas
Exact number of protesters exceeds one million nationwide
Foreign governments have directly organized and funded the protests
What comes back
Eight tabs,
eight questions.
It opens on Facts — whether what the article asserts is holding up elsewhere. The other seven answer what follows from that.
Run a cardFactsEvery claim, checked
Each claim is tagged verified, contradicted or unverifiable, with a bar showing how many sources agree and which ones conflict. Claims carrying a date sort into a timeline.
SourceThe publisher, decomposed
Trust axes from the publisher’s Media Trust profile, weighted and plotted. Only measured axes are drawn — below three you get bars, because a radar through missing data reads as a finding.
CitationsEverything it consulted
Every source behind the analysis, linked out so you can read the primary material yourself. None found says so rather than padding the list.
VerdictOne score, and its weak points
A reliability score with its grade band, a written summary, and an explicit list of what could not be verified.
FramingWhere the language leans
Bias direction and intensity, sensationalism, emotional appeal and political framing — then six manipulation indicators: cherry picking, false equivalence, appeal to authority, loaded language, omission, framing bias.
PropagationHow it travelled
Cross-platform spread with a severity tier — observed, notable, anomaly — and the likely intent behind the amplification.
ActionsWhat to do next
Ranked follow-ups: which figure to confirm independently, which omission to chase, which competing account to read first.
EntitiesWhose account it rests on
People, places and organisations pulled from the text — and, by their absence, who was left out of it.
How it is computed
Six passes, about fifteen seconds.
- 01
Read the article
Navigation rails and boilerplate stripped first.
- 02
Extract the claims
Only the sentences that assert something checkable.
- 03
Check them
Against other coverage. Conflicts named, not averaged.
- 04
Score the publisher
Domain-decomposed against Media Trust.
- 05
Measure the framing
Indicators surfaced as verbatim spans.
- 06
Attach confidence
Unmeasured stays unmeasured.
A high score means the claims converge and the publisher has a record — convergent coverage can still be wrong. Nothing is gap-filled: unmeasured reads “Not measured”, and conflicting sources are named rather than averaged. The methodology.
Pricing
Every plan includes the full analysis.
What changes is how far back you can look, and what you can hand to someone else.
Free
Try it
€0
Free forever. No card.
- 5 analyses a month
- 30-day window
- Three editorial blocs
- Chrome extension
- 13 languages
Reporter
One journalist or researcher
€49
per year — about €4 a month
- Everything in Free
- 12-month archive
- Unlimited analyses
- 20 deep source traces a month
- Five blocs, framing comparison
- Scholarly consensus checking
- Video and audio from a URL
Desk
Newsroom or research group
€3,900
per year — up to 5 people
- Everything in Reporter
- Full archive, no limit
- Exportable evidence dossier
- Unlimited deep traces
- Shared workspace and alerts
- Statistical significance on every figure
- Extra seats €390/year
- Italian public bodies: no tender required
Institutional
Companies and institutions
from€24,000
per year — unlimited people
- Everything in Desk
- API and MCP server
- Audit-ready retention with full trail
- Your own blocs and languages
- Right to republish commercially
- SSO, DPA, named contact
Prices exclude VAT. Annual billing. Academic and non-profit rates available on request.