The Problem: Guilty Until Proven Innocent
You've spent months building your outbound sales operation. Your team is trained, your CRM is dialed in, and your conversion rates are climbing. Then, overnight, your answer rates plummet from 18% to 3%.
Your numbers have been tagged as "Scam Likely".
This isn't a hypothetical. We've seen legitimate enterprises—insurance companies, healthcare providers, debt collectors—lose millions in revenue because carrier algorithms flagged their numbers as spam. And once you're in the black hole, getting out is nearly impossible.
How Carrier Reputation Works
Carriers (AT&T, Verizon, T-Mobile) use a combination of signals to score your outbound numbers. The exact algorithms are proprietary, but we've reverse-engineered the key factors:
1. Call Volume Variance
If your daily call volume spikes by more than 300%, you're flagged. This is why "blitz campaigns" (calling 10,000 leads in one day after weeks of silence) are suicide.
Red Flag Pattern
Monday: 50 calls → Tuesday: 50 calls → Wednesday: 5,000 calls
Result: Instant "Scam Likely" tag within 4-6 hours.
2. Answer Rate
If fewer than 8% of your calls are answered, carriers assume you're robocalling. This creates a vicious cycle: low answer rates → scam tag → even lower answer rates.
3. Call Duration
Calls under 10 seconds are treated as "abandoned" or "test calls." If more than 15% of your calls are sub-10s, you're flagged as a predictive dialer (which carriers hate).
4. Complaint Velocity
Carriers track how many users report your number via their "Report Spam" feature. Just 5-10 complaints per 1,000 calls can trigger a scam tag.
5. STIR/SHAKEN Attestation
If your calls aren't signed with STIR/SHAKEN "A" attestation (full verification), you're penalized. Most CPaaS providers (Twilio, Bandwidth) only offer "B" or "C" attestation, which means you're starting with a handicap.
The Analytics Engines
Carriers don't do this alone. They outsource reputation scoring to third-party analytics engines:
- First Orion (INFORM): Powers AT&T's Call Protect and T-Mobile's Scam Shield.
- TNS (Transaction Network Services): Provides data to Verizon and regional carriers.
- Hiya: Consumer-facing app that feeds data back to carriers.
These engines aggregate data from millions of calls and use machine learning to predict spam likelihood. The problem? Their models are trained on consumer behavior, not B2B patterns. If you're calling businesses during work hours with legitimate offers, you still look like a robocaller to the algorithm.
Case Study: The Healthcare Disaster
In 2023, a major Australian healthcare provider was tagged as "Scam Likely" while calling patients to confirm surgery appointments. Why?
- They ramped from 200 to 2,000 calls/day in one week (variance flag).
- Many patients didn't answer unknown numbers (low answer rate).
- Calls under 30 seconds were common (duration flag).
The result? 40% of patients missed their appointments, leading to a $2.3M revenue loss and a PR nightmare. It took 6 weeks and a direct escalation to Telstra's enterprise team to remediate.
Remediation Strategies
If you've been tagged, here's how to escape:
1. Register with Free Call Registry (Australia)
If you're calling Australian numbers, register your DIDs with the Free Call Registry. This signals to carriers that you're a legitimate business.
2. Implement STIR/SHAKEN "A" Attestation
Work with your carrier to get full attestation. This requires proving you own the number and have the right to use it. If you're using a CPaaS provider, switch to one that supports "A" attestation (e.g., Bandwidth, Telnyx).
3. Use Branded Calling
Register your numbers with Google Verified Calls and Apple Business Connect. When you call, the recipient sees your company name and logo instead of just a number. This dramatically improves answer rates and reduces spam reports.
Success Metric
One of our clients saw answer rates increase from 4% to 22% within 2 weeks of implementing branded calling.
4. Gradual Volume Ramping
Never spike your call volume. Instead, ramp gradually:
- Week 1: 100 calls/day
- Week 2: 200 calls/day
- Week 3: 400 calls/day
- Week 4: 800 calls/day
This trains the carrier algorithms to recognize your pattern as legitimate.
5. Monitor Your Reputation Score
Use tools like Free Carrier Lookup to check your number's reputation across carriers. Check daily and react immediately if you see a drop.
6. Carrier Escalation
If you're tagged unfairly, escalate directly to the carrier's enterprise support team. Provide evidence:
- CDRs showing legitimate call patterns
- Proof of consent (CRM exports, opt-in records)
- Business registration documents
This process can take 2-6 weeks, so start immediately.
Prevention: The Dreamtel Approach
At Dreamtel, we've built a governance layer that prevents scam tagging before it happens:
- Pre-Call Reputation Checks: Before dialing, we check the number's current reputation score. If it's degraded, we rotate to a clean number.
- Adaptive Throttling: Our system automatically adjusts call velocity based on real-time answer rates and complaint signals.
- Consent Verification: We cryptographically link every call to a consent record in your CRM, providing audit-ready proof for carrier disputes.
- Branded Call Orchestration: We handle the entire Google/Apple registration process and inject branding metadata into every SIP INVITE.
The Future: Regulatory Intervention
The current system is broken. Legitimate businesses are being punished while sophisticated scammers evade detection by constantly rotating numbers. We're advocating for:
- Transparent Scoring: Carriers should be required to disclose why a number was tagged.
- Appeals Process: A standardized, fast-track appeals process for businesses.
- Consent Registries: A national registry where businesses can register consent records, making it easy to prove legitimacy.
Until then, you need to be proactive. Talk to us about protecting your outbound reputation.