NoticeThis is a concept mockup, not a functioning prototype
RigorAppdigital credibility Index
draft concept · v1.0
Which brand watchdogs can you actually trust?
Terrorizing people online for clicks has gone too far. It's confusing people, driving up anxiety, and earns undue personal and financial gain for the so-called "watchdog."
Trusting brands is key — and influencer reviewers are an important part of that discernment. But not all those with digital reach are equally rigorous or knowledgeable.
RigorApp scores brand-critic creators on accuracy of their content, context, and the asymmetry between how far their false claims travel vs their (infrequent) corrections.
⌘K
Featured creator
Updated 2 days ago
w
WaterApp
@waterapp · TikTok · X · IG
8.4M reach (90d)
34
/ 100
Grade: D — Low credibility
Harm amplifierRepeat corrections
Reaches millions with brand accusations that frequently cite thresholds well below regulatory limits. Strong production, weak primary sourcing.
"Major bottled water brand X is 4× over the safe PFAS limit."
Threshold cited is 0.4 ppt — roughly 10× stricter than the current EPA enforceable maximum (4 ppt for PFOA/PFOS). Brand X tested within EPA limits in the same lab report.
False thresholdNo right-of-replyFear language ×7High amplification
"This protein brand is hiding heavy metals from you."
Tested values fall within FDA guidance. Framing implies concealment without evidence. Only one cited figure is accurate; the rest are wrong.
Misleading framingOne verified figureNo methodology link
"Our prior post about Brand Y overstated Heavy Metals — here's the corrected lab range."
Self-issued correction with linked lab PDF — but the correction reached only 2.3% of the original false post (88K vs 1.8M views). The asymmetry penalty applies in full: the audience that saw the false claim did not see the fix.
Reach: false post vs. correction−95% gap
False post1.8M
Correction88K
Correction issuedPrimary source linkedSevere reach gapNet harm not offset
What does a high-credibility brand watchdog look like? Same axes, opposite pattern: primary sources, disclosed methodology, and a willingness to publish results that hurt the virality of the story.
S
SupplementApp
@supplementapp · IG · X
1.2M reach (90d)
89
/ 100
Grade: A — High credibility
Trusted authorityScope-limited
Buys off the shelf anonymously, tests at an ISO 17025–accredited lab, publishes failures alongside passes. Disciplined within its category.
"NAD+ testing finds widespread failures: 7 of 8 popular brands missed label claims by >20%."
Products purchased anonymously from each brand's site. Tested at Eurofins (ISO 17025). All raw data and brand responses published per product page. Brands given 14-day right-of-reply window.
Primary source linkedRight-of-reply offeredMethodology disclosedRe-tested for confirmation
"Creatine gummy products tested: 0 of 6 met label claims for active ingredient."
Lab-confirmed at ISO 17025–accredited facility. Failed products re-tested with fresh batches; results held. Two brands have since reformulated and will be re-tested in 60 days.
ISO 17025 labRe-tested for confirmationReformulation tracked
What most credible creators actually look like — strong in their lane, with conflicts to manage. The index isn't binary; it scores creators on the dimensions they own and the ones they overreach in.
T
Influencer
@influencer · YouTube · IG · X
3.4M reach (90d)
79
/ 100
Grade: B+ — Credible with focus
Cites primary literatureSelf-taught
Cites peer-reviewed studies, generally measured, reliably accurate within nutrition. Affiliate disclosures; not a clinician.
"This grass-fed whey isn't worth the premium — here's what the studies actually show."
Cites Sharp et al. (n=42) and Devries et al. (n=86) showing marginal-to-zero performance differential vs conventional whey isolate. Acknowledges the brand's third-party testing. Concludes the product is fine; the pricing premium is not.
"These fasting products will keep you in ketosis."
Cites two mechanism studies on sodium during ketosis but only one direct RCT (n=14). Affiliate link present in description, not visually disclosed in-content. Thumbnail and title overstate the magnitude of the cited effect.
Mechanism studies citedSingle small RCTAffiliate buriedThumbnail overstates
Content that goes far and is false does the most damage. RigorApp plots every scored post on two axes: Reach (views × shares) and Accuracy (claims verified against primary sources). The combination — not either alone — drives the Harm Index.
WaterApp
SupplementApp
Influencer
Other scored creators
The editorial standard
Substanceoversensation
Rigoroverrumor
Dataoverdrama
Contextoverclickbait
Manifesto
Operating principles on why this index exists
"Reach and rigor are different skills, and the algorithm only rewards one of them."
How we score, in full
Open rubric
Creator score — 100 pts
Weighted sum, recalculated each Week.
25%
Claim accuracy
% of testable claims that match primary sources (lab data, peer-reviewed literature, regulator filings).
/25
15%
Primary sourcing
Direct links to lab PDFs, regulator pages, or named experts — not screenshots of screenshots.
/15
15%
Correction integrity
When wrong, does the creator issue a clear correction — and does it reach the same audience as the original?
/15
15%
Sensationalism
Frequency of fear-priming language ("toxic," "hiding," "poisoning") vs neutral framing, normalised to topic.
/15
10%
Domain focus
Track record per topic. A nutritionist gets credit on supplements, less on toxicology.
/10
10%
Conflict disclosure
Affiliate revenue, paid app subscriptions, competing products — disclosed in the post?
/10
10%
Reach-vs-retraction
The asymmetry penalty. If a wrong claim reaches 200k and the correction reaches 2k, that gap costs points.
/10
A watchdog who never misses the alarm but can't tell smoke from steam isn't keeping you safe — they're keeping you scared.
Manifesto
Why this matters
01
Trust the evidence, not the engagement.
Influencer reviews have become a real input in how people choose what to eat and drink — and that's a good thing. The problem is that reach and rigor are different skills, and the algorithm only rewards one of them.
02
The algorithm rewards alarm.
Digital algorithms are pro-inflammatory engines — they distribute the scariest version of a claim, not the most accurate — with zero accountability.
03
Accountability is a foundational pillar of a functioning society.
You learn early on in school to double-check your sources before turning in your paper. The justice system is a critical part of accountability — but it also is an expensive and slow mover. Accountability in the digital landscape is nearly zero these days.
04
Free speech and defamation for personal gain are not the same thing.
When you are sharing a free-speech opinion, that's one thing. But when you are running a business of criticising other people's business, that's trade speech. And it comes with financial liability.
05
Let's get something straight about the "watchdogs" — it's a business too.
We're big believers in questioning what you're told — including what you're told by the accounts whose brand is based on questioning others. A lot of them have a real following and with it, real revenues. Be it advertising, affiliate links, or app revenues… or the hope of earning those in the future.
06
There's no one path to health — make your own (informed) call.
We're not here to tell you what to think. We're here to give a way to understand who's earned trust through rigor — and who hasn't. Sourcing accuracy, lab accreditation, whether the methodology even matches what's being tested. Context over clickbait, evidence over outrage, empowerment versus fear. You know... the things you would expect of those that you place your trust in.
You're someone looking out for your health — not someone to be conned over clickbait. This initiative aims to help keep a record of which watchdogs have earned the right to consumer trust. Ultimately, you make the call.