We collect evidence and prepare detailed reports of deceptive review activity and false advertising for submission to online platforms and to federal, state, and local government agencies responsible for enforcement.
Our work focuses on identifying coordinated fake review schemes and misleading advertising practices, documenting them with concrete data, patterns, and screenshots so authorities and platforms can investigate, enforce their policies, and protect consumers.
From a plumber's detection system to the industry standard for review fraud accountability
The foundation is live. Seven detection engines running nightly on real competitors.
Click items to mark them complete. Everything tracked, nothing dropped.
Multiple revenue streams from a single detection engine
Based on organic growth with minimal ad spend. Assumes 5% free-to-paid conversion.
| Metric | Month 3 | Month 6 | Month 12 | Month 24 |
|---|---|---|---|---|
| Free Scans (cumulative) | 100 | 500 | 2,000 | 8,000 |
| Monitor Subscribers ($99/mo) | 5 | 20 | 60 | 200 |
| Enterprise Subscribers ($299/mo) | 1 | 5 | 15 | 50 |
| Consulting Reports ($500-5K) | 1 | 3 | 8 | 20 |
| Monthly Recurring Revenue | $794 | $3,475 | $10,425 | $34,700 |
| Annualized Revenue | $9,528 | $41,700 | $124,900 | $416,400 |
The regulatory environment is shifting. Enforcement is coming. CRFFA is first to market.
High-intent keywords with low competition -- CRFFA can own these.
Google is the platform. CRFFA is the watchdog FOR businesses ON that platform.
Google catches obvious spam. CRFFA catches sophisticated fraud -- real names, real photos, coordinated agencies -- that Google's automated system misses.
Two companies tried this before us and both are gone. That's not a red flag -- it's a lesson. They failed for specific, avoidable reasons that don't apply to CRFFA. Understanding exactly why they failed is what makes our approach different. Here's what killed them and what we're doing to make sure it doesn't happen to us.
One sentence: Fakespot gave it away to shoppers who didn't care enough to pay. We sell it to business owners who can't afford NOT to pay.
Federal, state, and platform policies all support what CRFFA does
The FTC finalized its rule on the use of fake reviews and testimonials, making it illegal to: buy or sell fake reviews, use AI-generated reviews, suppress negative reviews through threats, or misrepresent insider reviews as independent. Penalties: up to $50,000 per fake review. CRFFA is positioned as the enforcement tool that makes this rule actionable.
Section 43(a) of the Lanham Act lets businesses sue competitors for fake reviews as false advertising. Courts have already ruled on this. CRFFA builds the case.
CRFFA becomes the evidence engine + attorney referral marketplace for Lanham Act fake review lawsuits. We detect the fraud, build the case, and connect victims with attorneys who specialize in this exact law. Revenue from every referral.
Why it matters: Most businesses don\'t know they can SUE over fake reviews. The Lanham Act gives them the right. But they need two things: evidence (CRFFA provides this) and an attorney (CRFFA connects them). We sit in the middle of every case.
The Lanham Act is a B2B weapon. Consumers can\'t bring Lanham Act claims -- only businesses harmed by a competitor\'s false advertising. This is perfect for CRFFA\'s model: we serve businesses who are being hurt by competitors posting fake reviews. Every CRFFA customer is a potential Lanham Act plaintiff.
SaaS subscriptions + evidence packages + attorney referral fees. The marketplace model that scales.
| Input (From Kalen) | Output (Robert Builds) | Timeline | Status |
|---|---|---|---|
| Research 5 Lanham Act attorneys in KC metro | Attorney directory page on CRFFA site | By Apr 21 | PLANNED |
| Decide referral fee structure (% or flat) | Attorney partnership agreement template | By Apr 28 | PLANNED |
| Draft "attorney pitch" talking points | Attorney onboarding landing page | By May 5 | PLANNED |
| Make first 3 attorney outreach calls | Automated case referral email system | By May 12 | PLANNED |
| Get one signed partnership agreement | Referral tracking dashboard | By May 19 | PLANNED |
Full fraud detection + evidence generation for less than a Netflix subscription
| Service | Monthly Cost | What It Does | Status |
|---|---|---|---|
| 🔎 Outscraper | FREE (500/mo) or ~$10 | Full competitor review sets -- unlimited reviews, not just 5 | READY |
| 🕵️ SerpAPI | $25/mo | Contributor profile lookups -- up to 200 reviews per reviewer. The "catalog" detection Kalen wants. | READY |
| 🤖 OpenAI GPT-4o-mini | ~$2-5/mo | AI authenticity scoring. Batch API = $0.65/10K reviews | READY |
| 💬 Google Cloud NL | FREE (5K/mo) | Structured sentiment analysis | FREE TIER |
| 🛡️ Perspective API | FREE | Spam/toxicity detection (sunsetting Dec 2026) | FREE TIER |
| 👁️ Google Vision | FREE (1K/mo) | Profile photo analysis -- detect stock/AI-generated photos | FREE TIER |
| 📚 Wayback Machine | FREE | Evidence archiving with timestamps | FREE TIER |
| 📄 WeasyPrint | FREE | PDF evidence package generation | LIVE |
| 📸 ScreenshotAPI.net | FREE (100/mo) | Visual screenshots for evidence packages | FREE TIER |
| 🚀 TOTAL | ~$27-40/mo | Full fraud detection + evidence stack. Near-infinite SaaS margins. | |
Google Places API only gives 5 reviews per business. Outscraper gives ALL of them. This is the difference between a toy and a weapon.
SerpAPI Contributor Reviews shows a reviewer's FULL history (200 reviews) -- this is how we detect the "catalog" pattern Kalen described: reviewer reviews 15 unrelated businesses across 3 states.
Both major competitors are gone -- Fakespot shut down Jul 2025, ReviewMeta shut down 2023. Total market vacuum. Nobody is doing this.
FTC Consumer Review Rule: $53K per violation (effective Oct 2024). Creates massive demand for evidence-grade fraud detection.
Operating cost is ~$27-40/month. Even at $49/scan, one customer/month covers the entire stack. At scale with SaaS pricing ($99-299/mo per client), margins approach 95%+.
DOVE Dictionary + CompTIA Data+ certification = credibility no competitor can match
Robert Dove\'s thesis-level NLP project that powers Engine 2 (Linguistic Analysis) of the 7-engine detection system. Academic-grade natural language processing for identifying fake review patterns.
Robert Dove is pursuing CompTIA Data+ certification via NW Missouri State / NCLab. The same statistical rigor that backs the CRFFA detection system.
Four scenarios, four plays. CRFFA wins in every one.
Review platforms continue slow, reactive enforcement. Industry-wide false negative rates of 70-90% persist across Google, Yelp, Facebook, BBB, and Angi.
Play: Build market share fast. Become the de facto standard for review integrity.
Major platforms create partner/API programs for third-party review auditors.
Play: Position CRFFA as the certified audit partner. Early mover advantage is massive.
Platforms or regulators mandate third-party review verification for advertising programs.
Play: CRFFA becomes compliance infrastructure. Recurring revenue from every home services company.
One or more platforms dramatically improve internal detection, closing their gap to under 20% false negatives.
Play: Pivot to legal evidence packaging. Google catching fakes does not equal courtroom-ready evidence. CRFFA becomes the forensic layer.
CRFFA is not betting against any platform. We are the independent auditor that works across Google, Yelp, Facebook, BBB, Angi, and HomeAdvisor. The same way companies have internal accounting AND external auditors, businesses need both platform review systems AND independent verification. The FTC Rule (Oct 2024) created $51,744/violation penalties that did not exist when Fakespot or ReviewMeta launched.
Five gates between AI detection and public reporting. Zero false accusations.
Pattern analysis (timing clusters, reviewer history, linguistic signals) produces a confidence score 0-100.
Only scores above 85% confidence advance to human review. Below 85% stays flagged internally but never public.
Trained reviewer confirms or rejects within 48 hours. AI recommends, humans decide.
Flagged business gets 14 days to respond before any public report is issued.
Any report naming a specific business gets attorney sign-off. Never say "fake" -- say "anomalous review patterns detected."
CRFFA makes professional judgments about business legitimacy. Budget $3K-8K/year for a $1M E&O policy. Get it before the first public report ships.
Start at $0-2/month. Scale only when revenue justifies it.
Heuristics catch 70-80% of clear cases. GPT-4o-mini handles the ambiguous 20-30%.
After 5-10K labeled reviews, fine-tune DistilBERT. Runs on current VM.
Multiple customers. Ensemble model + optional local LLM for explanations.
| Approach | Monthly Cost | Speed | Accuracy | Setup |
|---|---|---|---|---|
| Heuristics Only | $0 | 100+ rev/s | 70-80% | Done |
| Heuristics + GPT-4o-mini | $0.30-2 | 5-10 rev/s | 85-92% | 1 day |
| Heuristics + DistilBERT | $0-15 | 20-50 rev/s | 85-90% | 5-10K labels |
| Local 7B LLM (quantized) | $30-55 | 0.5-2 rev/s | 90-95% | GPU needed |
| Full GPU Instance | $200+ | 10-30 rev/s | 95%+ | Revenue first |
LoRA freezes the base model and trains tiny adapter matrices (0.1-1% of parameters). Result: custom fraud detection model for $1-4 total training cost.
Fakespot and ReviewMeta are dead. Here is why we survive.
10M free users, zero revenue, Mozilla killed it
One-man project, creator became paramedic
B2B from day one. Honest businesses pay for certification badges and competitor audits. No free consumer tool as primary product.
Monthly monitoring subscriptions + Lanham Act evidence packages + "Verified Clean" certification badges. If one dips, two others carry.
No browser extension. No mobile app. Backend-only architecture. Apple and Google cannot remove what does not exist in their stores.
Outscraper + SerpAPI + Google Vision + Perspective API. If one blocks us, three others still work. No single dependency.
We help every review platform ecosystem. We are not scraping or intercepting traffic. We report fraud that helps platform integrity across the board.
FTC Rule ($51K/violation) + Lanham Act Section 43(a) create paying customers (law firms, AGs) that Fakespot never had.
Clients build legal cases on our evidence. Switching means starting evidence collection over. Monthly monitoring creates lock-in.
State AGs, FTC, consumer protection agencies. SAM.gov registration + SOC 2. One state AG pilot proves the model, then scale to all 50.
LLC taxed as S-Corp. E&O insurance before first report.
Home services alone is $374.6M. Multi-vertical expansion into legal, automotive, healthcare, and restaurants takes it to $1.9B.
| Segment | Businesses | Price | TAM |
|---|---|---|---|
| Small business | 1.8M | $199/mo | $214.9M |
| Mid-market | 400K | $299/mo | $114.8M |
| Enterprise | 50K | $499/mo | $44.9M |
| Vertical | TAM |
|---|---|
| Legal services | $134M |
| Automotive | $83.6M |
| Healthcare | $239M |
| Restaurants | $299M |
| Real estate | $448M |
No dedicated fake review detection with court-ready evidence exists. CRFFA creates a new category: Review Fraud Intelligence.
| Company | What They Do | Fraud Detection? | Price |
|---|---|---|---|
| BirdEye | Review management | Basic sentiment only | $299-499/mo |
| Podium | Review solicitation | None | $289-449/mo |
| Reputation.com | Enterprise reputation | Some anomaly | $500-2K+/mo |
| Fakespot | Consumer analysis | DEAD Jul 2025 | N/A |
| ReviewMeta | Amazon only | DEAD ~2024 | N/A |
80-85% gross margin. Break-even at $2-3M ARR. 30%+ operating margins at scale.
| Tier | Monthly | LTV | LTV:CAC |
|---|---|---|---|
| Shield | $199 | $3,383 | 11.3x |
| Sentinel | $399 | $11,192 | 22.4x |
| Fortress | $2,450 | $98,000 | 19.6x |
| Investigator | $599 | $12,729 | 25.5x |
Likely acquirers: Reputation.com, BirdEye, LexisNexis, TransUnion, Thoma Bravo, Vista Equity.
Neighborly (5,000+ locations, 30+ brands) at $49/location/month = $2.94M ARR from one contract.
| Franchise | Locations | Annual Deal |
|---|---|---|
| Neighborly | 5,000+ | $1M-2.5M |
| ServiceMaster | 3,000+ | $500K-1M |
| Roto-Rooter | 600+ | $180-360K |
| Benjamin Franklin | 250+ | $75-150K |
From action plan to live platform in 30 days
Nobody else is doing this at this level. The technology is built. The legal framework supports it. The market is waiting. Kalen -- this is your lane. Own it.