January 31, 2026 · charmbox team
VOIP vs carrier numbers: how phone verification actually works
You try to verify an account with a Twilio number. The platform rejects it. Google Voice, same. TextNow, rejected. The reason is straightforward: platforms query carrier lookup APIs that return the line type of any phone number. If it comes back as VOIP, verification is denied before a code is ever sent. This isn't heuristic or guessable -- it's a direct database lookup against telecom infrastructure records.
Every US phone number is registered in databases tracking its carrier and line type: mobile (T-Mobile, AT&T, Verizon), landline, fixed VOIP, or non-fixed VOIP (Twilio, Google Voice, TextNow). Platforms send numbers to lookup services like Twilio Lookup, IPQS Phone Validation, Sinch Number Lookup, or Telnyx Number Lookup. These APIs query Local Number Portability (LNP) databases maintained by the NPAC -- every US number has a Location Routing Number pointing to its current serving carrier. Even ported numbers reflect the current owner. If your number belongs to Twilio or Bandwidth in the carrier registry, that's what the API returns. No configuration trick changes it. Sinch now uses machine learning on top, scoring carriers on VOIP likelihood to catch edge cases that simple carrier-name matching would miss.
Most major platforms enforce VOIP blocking. WhatsApp explicitly blocks VOIP for registration. Instagram rejects VOIP at verification. Telegram, TikTok, and Snapchat all apply filtering. Discord requires mobile numbers explicitly. Financial platforms like Coinbase and Binance block VOIP for fraud prevention. Steam detects and rejects VOIP numbers too. The success rate gap is stark: VOIP numbers see roughly 20-40% success on major platforms, while real carrier numbers hit 95-99%.
At the user level, a VOIP number and a carrier number look identical -- same 10-digit format, both receive texts. The differences are entirely at the network layer:
| Feature | VOIP Number | Carrier Number (SIM/eSIM) |
|---|---|---|
| Provisioned by | Internet telephony provider (Twilio, Bandwidth) | Mobile carrier (T-Mobile, AT&T, Verizon) |
| IMSI | None | Yes -- unique subscriber identity tied to SIM |
| HLR authentication | None | Yes -- authenticates via Home Location Register |
| Signaling network | Routes over internet (SIP/RTP) | Routes over SS7/cellular infrastructure |
| Carrier lookup result | Returns VOIP provider name | Returns mobile carrier name |
| Line type | nonFixedVoip or fixedVoip | mobile |
| Passes platform verification | Unreliable -- blocked on most platforms | Yes -- consistently passes |
The key distinction is whether a number exists on a mobile network with a real Mobile Network Code. Carrier numbers authenticate against the carrier's HLR -- the SIM presents its IMSI, the network sends a challenge, the SIM encrypts it with its secret Ki key. VOIP numbers skip this entirely. They have no IMSI, no Ki key, no HLR record. That's the fundamental reason they're detectable: they don't exist in mobile network infrastructure at all.
This is where Charmbox fits. Instead of finding ways to sneak VOIP numbers past detection, Charmbox provisions actual carrier-network eSIMs on real physical phones. Each device has a real baseband processor connected to a real cellular network, so carrier lookups return a legitimate mobile carrier name -- not Twilio, not Bandwidth, not Google Voice. The eSIM authenticates through the same HLR process as any other mobile subscriber. This matters because phone verification is a binary gate: the carrier lookup either returns mobile or it doesn't, and no configuration changes the answer. Pricing scales with volume -- see charmbox.ai for details.
Whatever provider you use, verify that numbers show up as mobile in a carrier lookup before committing. You can check with the Twilio Lookup API or the IPQS free carrier lookup tool. If the line type comes back as anything other than mobile or wireless, it will likely be rejected. The platforms aren't wrong to block VOIP -- cheap, disposable VOIP numbers are the primary tool for spam and fraud. But understanding the detection mechanism is the first step toward solving the problem correctly.