Standard CTIA keyword matching takes care of the opt-outs that use the most obvious language (“cancel,” “unsubscribe,””quit,” etc.), but people don’t always communicate in the most obvious ways. Yet they do still expect you to honor their “take me off your list” or “please don’t text me again” messages. STOP-word filters are table stakes. TrueOptOut goes further by honoring the meaning of opt-out texts, even if they don’t exactly match common opt-out keywords.
TCPA Protection Beyond Keywords
Honor Every Opt-Out, However It’s Worded
TrueOptOut® is TrueDialog’s AI-powered opt-out intent detection engine, meaning it recognizes contact opt-out requests that standard keyword matching misses. TrueOptOut understands what people mean, keeping your lists clean and protecting you from TCPA (Telephone Consumer Protection Act) exposure before it becomes liability.
Tour TrueOptOut
See how TrueDialog catches the opt-out requests keyword matching misses — using AI to honor intent, not just the word “STOP,” and building a compliance-grade audit trail on every decision.
Keyword Matching Sees Words. TrueOptOut Understands Intent.
Your contacts don’t always use the right words when asking to opt out. When someone texts “please take me off your list,” CTIA keyword matching doesn’t recognize it as an opt-out.
That puts your lists at a disadvantage and your brand at risk of TCPA violations that carry hefty fines. On top of that, every message sent to someone who has already asked to stop is wasted spend, and it chips away at the trust and sender reputation you rely on.
TrueOptOut uses machine learning to evaluate inbound messages in under 10ms with >=90% recall. Conservative by design, the system honors ambiguous signals as opt-outs by default. That makes the system highly unlikely to miss any opt-out requests, protecting your reputation and reducing your risk of having to defend a missed request.
TrueOptOut captures the vast majority of opt-outs, sending questionable messages first to an LLM (large language model) adjudication layer, then to a human review queue. Your team has the chance to confirm whether the message counts as an opt-out before final action is taken, or opt them back in. TrueOptOut then applies that decision to ongoing model training to improve future results.
The details of every opt-out decision TrueOptOut makes are automatically logged, as are human review decisions. That creates a compliance-grade paper trail that’s ready for your team, regulators, auditors, or legal review. For organizations operating in regulated industries such as financial services, healthcare, and higher education, TrueOptOut helps you cover all your bases.
TrueOptOut doesn’t replace CTIA keyword matching — you can count on all the opt-out messages that worked before to still work. It simply augments it, helping you make sure you capture the kind of opt-out messages that were previously missed.
TrueOptOut’s AI runs entirely within TrueDialog’s private infrastructure. That means your message content is private and secure. It will never be shared with third-party or open-source models. TrueOptOut, like everything else TrueDialog offers, is consistent with SOC 2 Type II, TX-RAMP, and ISO 9001 certifications.
At large organizations with big contact lists, the number of opt-out messages — even just those that get past the keyword matching layer — can be too much for your team to handle manually. Even if they could, your employees have better things to do than sort through potential opt-outs. TrueOptOut takes that work off their plate, ensuring you capture all the opt-outs, without wasting anyone’s time, no matter how high the numbers are.
Frequently Asked Questions
About TrueOptOut
TrueOptOut is a new TrueDialog feature that automatically reviews inbound text messages that pass the initial CTIA keyword-matching screen, then flags any that still count as opt-out requests.
The main messages CTIA keyword matching catches are: STOP, QUIT, END, REVOKE, OPT OUT, CANCEL, and UNSUBSCRIBE.
Standard keyword matching overlooks a few different types of opt-out messages, such as:
- Implicit requests like “take me off your list,” or “don’t text me anymore”
- Contextual frustration like “I already told you to stop,” and “why are you still texting me”
- Messages with misspellings and abbreviations like “stopp,” “plz stop,” and “no more txts”
- Hostile, non-keyword expressions such as “this is harassment,” or “I never signed up for this”
No, TrueOptOut augments your existing process. Keyword matching runs first and applies the opt-out on obvious matches. TrueOptOut then evaluates what gets past it.
TrueOptOut adds no perceptible latency. On the high-confidence path, TrueOptOut may add extremely brief latency (≤10ms). But mid- and low-confidence messages will still deliver immediately, with adjudication happening in the background.
Yes. You’ll be able to see all relevant details directly in the TrueDialog dashboard: message, confidence score, action, and timestamp.
TrueOptOut has ≥90% recall on the opt-out class. It’s built to catch at least 9 in 10 opt-outs that keyword matching would miss. Human review and conservative LLM adjudication backstop the ambiguous cases.
Ongoing retraining from human review decisions and new opt-out phrasing ensure accuracy will only improve over time.
Yes, TrueOptOut can even infer opt-out from negative emojis.
Yes, contact TrueDialog customer support if you want to modify your thresholds.
No. Classification runs on TrueDialog’s own model, while mid-confidence adjudication uses Claude Haiku under TrueDialog’s private infrastructure. All message content is private and secure.
