AI voice agents must check Do Not Call lists before dialling, yet most teams still handle this manually or not at all. The Federal Trade Commission reported in 2023 that TCPA violations cost businesses an average of $43,000 per case, and a single campaign can trigger dozens of lawsuits if compliance fails. AI voice agent DNC compliance is not optional; it is the difference between a scalable outbound operation and a legal liability.
This article explains how automated DNC enforcement works inside AI voice systems, what breaks when it is set up wrong, and how to implement it without stopping your outbound calling dead. We will walk through the technical mechanisms, real costs, and the specific platforms and practices that separate compliant operations from the ones regulators target.
What Is DNC Compliance for AI Voice Agents
The Do Not Call Registry is a federal database maintained by the FTC where consumers can register phone numbers they do not want called. Telemarketers, debt collectors, and survey companies must scrub their calling lists against this registry before dialling. The Telephone Consumer Protection Act sets the rules, and violations trigger civil penalties of $500 to $43,000 per call, with class action lawsuits common when violations cluster. AI voice agents that ignore this rule face the same penalties as human callers.
But DNC compliance extends beyond the federal registry. Most industries maintain internal DNC lists: customers who asked not to be called, numbers that belong to employees, hospitals, or government offices. Many states and countries have their own DNC registers separate from the federal one. TCPA compliance AI calling means checking all three layers before the agent picks up the phone to make the first call.
The compliance obligation falls on the caller, not the platform. If you hire a third-party voice provider, you remain liable for calls made on your behalf. The FTC does not care that you used automation; it cares that a phone number on the federal DNC list received a call. Your vendor's automation is your defence, not your exemption.
In practice, this means your AI voice agent system must have three capabilities: access to the current DNC database, real-time lookup before dialling, and documented proof that the lookup happened. Without these, your outbound calling is not scaling; it is accumulating legal exposure.
How Automated Do Not Call Checking Works
The federal Do Not Call Registry publishes its full list monthly as a downloadable file. A compliant AI voice system ingests this file, stores it in a queryable database, and checks every phone number against it before the agent dials. This process takes milliseconds. The agent has the intended outcome (book a follow-up appointment, qualify a lead) and a phone number; the compliance engine runs a lookup; the system either proceeds or skips and moves to the next contact.
The lookup is exact. A number either appears on the DNC list or it does not. But the process has real friction points. Phone numbers change format between systems: 15551234567, (555) 123-4567, and +1 (555) 123-4567 are the same number, but a naive lookup fails if the formats do not match. The database must normalize formats before comparison. State DNC lists use the same phone but different databases; a system checking only the federal list misses state-level protections in California, Texas, and others.
Internal DNC lists add another layer. Your sales team marks numbers "do not call again." Those marks must feed into the same pre-dial check, or the agent calls the same person twice and violates the customer's stated preference. When your system is built in silos, the voice agent uses one list while your CRM maintains another. That mismatch is where most unintentional violations happen.
Real-time checking is also essential. If you batch-check your list once a month and then dial against the result, any number added to the federal DNC list between your check and your dial is a violation. Compliant systems check on every call attempt. The infrastructure cost is low (a millisecond database lookup per call), but it requires the system to be designed around it from the start. Bolting DNC checking onto an existing dialer after the fact usually breaks other features or introduces lag.
Where Automated Dialer Compliance Fails in Practice
Many businesses implement DNC checking and still get sued. The failure is usually not technical but operational. A company adds DNC checking to its outbound dialer, runs a campaign, and gets flagged because the compliance layer was turned off by mistake, or a manual override was used without logging, or the DNC list was three months out of date. The system was compliant in design but not in execution.
One concrete example: a mortgage broker integrated a voice AI system with built-in DNC checking. During a refi campaign, the sales manager needed to call a specific person urgently and manually added the number to the dial queue, bypassing the automated check. That call landed on someone who was on the federal DNC list. The FTC later found records showing dozens of similar manual overrides. The automated system was blameless; the business lost $67,000 in settlements.
Another common failure point is list staleness. The federal DNC Registry updates continuously; a compliant system must refresh its local copy at least weekly. Some platforms claim to offer DNC checking but actually use a static list downloaded once and never updated again. That looks compliant on paper but is illegal in practice. Ask your vendor when they last pulled the registry file and how often they refresh it.
Integration breaks are also common. An AI voice agent may have DNC built in, but if your CRM does not sync its internal DNC list to the dialer, you are checking a partial list. The agent sees the federal registry and your company's DNC list, but not the customer who asked your sales team not to call them last week. That customer gets called again, feels harassed, and you face both a complaint and a repeat-violation penalty.
Building DNC Compliance Into Your Calling Architecture
Implement DNC checking at the source. Before the AI agent attempts a call, the system should query the compliance database. If the number is flagged, the system logs the skipped attempt, moves to the next contact, and provides your team with a report. This flow is so simple it is easy to overlook its importance: every call attempt must be logged, not just completed calls. If your system skips a number but does not record that it tried, you have no proof of compliance if a regulator asks.
Use a third-party DNC service if you are not equipped to manage federal and state registries yourself. Services like TrueCaller, Marchex, and LexisNexis maintain updated databases and offer API access. Cost ranges from $0.01 to $0.05 per lookup, so a 10,000-contact campaign costs $100 to $500 in compliance overhead. That is not insignificant, but it is far cheaper than a single TCPA settlement. A built-in CRM that integrates with these lookup services avoids having to manually pipe data between tools.
Document every lookup. Your compliance logs should show: phone number, lookup timestamp, database queried, result (found or not found), and whether the call proceeded. If the FTC asks for proof that you checked before dialling, these logs are your evidence. Without them, the call looks intentional, not accidental. Systems that perform the lookup but do not log it are still in violation; you just cannot prove otherwise.
Set up monitoring and alerts. If your DNC lookup service goes down, your dialer should pause rather than skip the compliance check and proceed anyway. Too many operations run a backup: if the DNC lookup fails, call anyway. That is worse than no checking at all because it is negligence, not an outage. A circuit-breaker pattern is safer: if DNC checking fails, log the failure, alert your team, and pause the campaign until the issue is fixed.
State and Regional DNC Lists Beyond Federal Registry
The federal Do Not Call Registry covers most of the US, but California, Texas, and several other states maintain their own lists with separate rules. California's list is older and predates the federal registry; some numbers are on California's list but not the federal one. Texas allows businesses to keep internal DNC lists and sue telemarketers who violate them. Checking only the federal list in these states leaves you exposed.
International calling adds another layer. Canada has the National Do Not Call List, equivalent to the US federal registry. The UK has the Telephone Preference Service. If your AI voice agent operates across borders, each region has its own compliance framework. A system that works for US calls may be illegal in Canada because it does not check the DNCL or respect Preference for Contact rules.
Many platforms claim to handle "state compliance" but actually check only the federal list. Ask your vendor explicitly: do you check California's DNC list, Texas's internal list, Canada's DNCL, or any other regional registry? If they are vague, they do not. The safest approach is to use a compliance service that covers all jurisdictions you operate in, even if you only expect to call a few numbers in each state. The per-lookup cost is the same; the risk reduction is significant.
Some businesses operate in verticals with sector-specific rules. Debt collectors must comply with both the DNC list and the Fair Debt Collection Practices Act, which has its own calling restrictions. Healthcare providers calling for appointment reminders must avoid calling patients who have asked not to be contacted. Nonprofits soliciting donations have different obligations than for-profit telemarketers. Your compliance system must reflect your specific sector, not a generic one-size approach.
TCPA Compliance AI Calling and Consent Frameworks
DNC list checking is one layer of TCPA compliance, but it is not the whole picture. The Telephone Consumer Protection Act also requires prior express written consent for calls to cell phones and auto-dialled calls to any number. An AI voice agent that complies with DNC but lacks consent logging is still breaking the law.
Consent must be documented before the call. If you call someone without proof that they agreed to receive calls from you, the call is a violation even if their number is not on the DNC list. Your compliance system must track when and how the person consented, what channel (email signup, form submission, phone call), and what service they consented to be contacted about. Some people consent to sales calls but not promotional calls; others want updates only via email.
Automated consent retrieval is risky. Some platforms claim to check "consent databases," but most consent is scattered across your email marketing system, your website sign-up forms, and handwritten notes in your CRM. A robust system pulls consent from all these sources and makes it available to the voice agent at dial time. If the agent finds no consent record for a contact, the agent should not call, or it should call with a soft introduction and allow the person to confirm consent verbally before proceeding.
Managing consent revocation is equally critical. When someone says "remove me from your list" or clicks an unsubscribe link, that instruction must flow back into the voice system immediately. A caller who unsubscribed yesterday should not get a call today. Real-time integration between your email, web forms, and voice system is necessary. Many businesses rely on batch updates that run overnight, creating a window where calls are made to numbers that have already requested to be removed.
Logging and Documentation for Regulatory Audit
If the FTC opens an investigation into your calling practices, you will need to produce records showing compliance. For every call attempted, you need to show: the phone number called, the date and time, the reason for the call, evidence that the number was checked against the DNC list before the call, and the result of that check. If you called someone who was on the DNC list, you need evidence of an exception (such as established business relationship) or consent.
Most businesses underestimate the documentation burden. An audit of a year-long campaign with 100,000 calls means producing 100,000 DNC check records. If your system does not log these automatically, you cannot produce them, and the FTC assumes non-compliance. The compliance layer in your voice system must treat logging as a first-class requirement, not an afterthought.
Retention policies matter too. The FTC expects records to be kept for at least 18 months. Some industries require longer. Your system should retain logs indefinitely or at minimum for two years. Deleting logs to save storage is a compliance violation if those logs would have shown lawful calling. Set up archival storage for old call records so you can produce them if asked without deleting them from your active system.
Your logs should also include exceptions and overrides. If a manager overrode an automated skip to call a specific number, that override should be logged with the reason and the approver's name. If a call was made under the "established business relationship" exception to the DNC list, the log should state why that exception applied. These details turn a compliance log from a list of calls into a legal defence.
Integrating DNC With Your Existing Call Workflows
Adding DNC compliance to an existing outbound operation is disruptive. The most common mistake is bolting DNC checking onto a dialer that was not built for it. The agent starts with a list of 1,000 numbers, the system checks them all at the start of the campaign, and then dials away. That checks each number exactly once, six weeks ago. New numbers added to the DNC list after that initial check are called illegally.
Real-time per-call checking is necessary. Every time the agent attempts a call, the system checks that specific number against the current DNC database. This introduces a small delay, usually 50 to 200 milliseconds, between the decision to dial and the actual call attempt. For human agents this is imperceptible; for high-volume automated systems, it adds measurable latency. Modern systems handle this with parallel processing, checking while the call is connecting, but older dialer architectures struggle.
The integration also affects list management. If your sales team uses spreadsheets to manage contact lists, and those lists are uploaded to the dialer weekly, you have a window where DNC numbers slip through. A better flow uses your CRM as the source of truth, maintaining a single contact record that includes DNC status, consent status, and call history. The dialer reads from the CRM, not from a detached spreadsheet. When the CRM marks a contact as "do not call," the dialer respects that immediately.
Some businesses need to move away from batch calling entirely. If you upload 10,000 numbers at 9 AM and dial through the day, any number added to the federal DNC list by 2 PM is illegally called at 3 PM. More sophisticated operations use dynamic lists, where the dialer requests the next set of numbers to call from the CRM in real time, checking each batch against current DNC status immediately before dialling. This is slower but compliant.
The Cost and Performance Trade-offs of Compliance
Implementing automated DNC compliance has real costs and introduces real friction. Per-call DNC lookups cost $0.01 to $0.05 each when outsourced to a third-party service. A business making 100,000 calls per month faces $1,000 to $5,000 in monthly compliance costs. For a small operation, this is significant. Some platforms promise "free" DNC checking and deliver a static list that is never updated, which is not compliance, it is fraud.
The latency cost is less obvious but equally real. A per-call DNC lookup adds 50 to 200 milliseconds of delay before each call connects. For a predictive dialer optimized to connect calls as fast as possible, this delay reduces call concurrency and lowers productivity. A team of five agents expecting to make 150 calls per day might hit 140 after latency is added. That is a 6% productivity loss, which compounds across hundreds of agents.
Some types of calling are more expensive to comply with. Outbound appointment reminders, where you call thousands of patients the day before a doctor's visit, have lower TCPA risk because the consent is usually documented by appointment booking. Adding per-call DNC lookups to a reminder campaign is expensive compliance overkill; many compliance experts recommend a lighter touch. Cold sales calling, by contrast, has high compliance risk and requires the full apparatus.
The operational overhead is significant. Compliance systems fail. A DNC lookup service goes down, or your local copy of the federal registry becomes corrupted, or an employee misconfigures a rule. Your team needs people trained to spot these issues, respond quickly, and understand why compliance failures happen. Many small businesses do not have this capacity and should avoid high-volume outbound calling until they do.
Choosing a Voice Platform With Built-in Compliance
Not all AI voice platforms take compliance seriously. Some treat it as a checkbox feature and deliver something that looks compliant without actually being so. Look for specific signals: does the platform list which DNC registries it checks? Can you see the logging and audit trail for a test call? Do they document their refresh rate for the federal registry? Are they transparent about costs?
Platforms designed for compliance-heavy verticals like healthcare and debt collection tend to have stronger DNC infrastructure than general-purpose systems. A platform built for appointment reminders may have weak sales-call compliance because their customer base does not need it. Ask for a case study from someone in your industry, not just a generic customer list. The person who successfully used this platform for mortgage outbound will tell you things a general customer will not.
Integration capability matters more than the feature itself. You want a voice platform that connects to your CRM, pulls contact and consent data from it, and writes call outcomes back to it. Sysevo and similar platforms with built-in CRM functionality reduce the number of integration points where compliance breaks. When the voice system and CRM are the same product, data flows cleanly.
Ask about their compliance framework during a demo. What happens if a call reaches someone on the DNC list? Is it logged? Can you pull a report of all DNC skips for a campaign? What is their refresh rate for the federal registry? If they cannot answer these questions with confidence and specificity, walk away. The wrong voice platform will create legal liability faster than no platform at all.
Manual Overrides and Exceptions in AI Calling
Legitimate exceptions to the DNC list exist. If you have an "established business relationship" with a customer (they bought from you in the past 18 months, or they contacted you first), you can call them even if they are on the DNC list. Debt collectors, creditors, and certain nonprofits have sector-specific exceptions. Your AI system must be able to recognize and log these exceptions.
But exceptions are where compliance often breaks. A sales manager sees a high-value prospect on the DNC list and says "call them anyway, we have a relationship." The manager is wrong about the relationship, or the relationship was too long ago, or the relationship does not meet the legal threshold. Without oversight, every exception claim is acted on, and violations accumulate. Compliant systems require someone to verify the exception before allowing the override.
The logging of exceptions is critical. If an override is used, that call must be flagged in the audit trail with the reason and approver. "Established business relationship" is not enough; the log should show the date of the last interaction, the nature of the relationship, and who verified it. When regulators audit you, these details separate intentional commerce from negligent calling.
Some platforms allow blanket exception classes: "do not skip anyone in the sales department's spreadsheet." This is usually a violation waiting to happen. Better practice is to require individual verification for each exception. It is slower, but it creates the audit trail you need and reduces the risk that exceptions are claimed for bogus reasons.
Monitoring and Alerting for Compliance Drift
Compliance is not a set-it-and-forget-it feature. Once your DNC checking is live, you need continuous monitoring to catch drift. This means alerts for: DNC lookup failures or latency spikes, unusual skip rates (if 40% of your calls are being skipped for DNC when the normal rate is 2%, something is broken), calls placed to numbers that should have been skipped, and changes to your DNC configuration.
Too many operations notice compliance problems only when the FTC calls. A compliance-aware team receives a daily or weekly report showing DNC activity: calls placed, calls skipped due to DNC, exception overrides, and lookup latency. Unusual patterns trigger investigation before they become legal exposure. This monitoring layer is less glamorous than the dialer itself, but it is where compliance actually lives.
Audit your logs yourself before a regulator does. Once a quarter, pull a random sample of 100 calls from your recent campaigns and verify that each one was checked against the DNC list before dialling. Look for patterns: are certain numbers routinely called despite being flagged? Are exceptions being claimed for weak reasons? This self-audit takes a few hours and can catch systematic compliance gaps before they turn into violations.
Changes to your calling rules are also compliance events. If you start calling a new state, you need to add that state's DNC list to your checks. If you change your consent requirements or add a new consent channel, you need to verify that the voice system and your CRM both reflect the change. A single misconfigured rule change can create months of illegal calling; monitor changes as carefully as you monitor call volume.
When AI Voice Calling Is Not the Right Choice Yet
Despite all the automation available, some businesses should not use AI voice calling yet. If your team lacks compliance expertise and your budget does not allow hiring someone or contracting with a compliance consultant, you are not ready. Compliance failures happen when no one understands the rules. An AI platform makes the technical part effortless, but humans still need to understand what compliance means and why it matters.
If your calling volumes are tiny (fewer than 1,000 calls per month), the compliance overhead makes the operation uneconomical. The per-call costs of DNC checking and the operational overhead of monitoring are fixed regardless of scale. At 1,000 calls per month with 3% of those skipped due to DNC, you are spending $100 to avoid 30 illegal calls. That math improves quickly as volume grows, but at low volumes, it is not compelling.
If your business model relies on calling people who have not consented and are not past customers, you are in an especially difficult compliance position. Cold calling with prior express written consent is possible but requires meticulous record-keeping and integration work. Some businesses decide the compliance burden is not worth the expected revenue and pivot to inbound or consent-based outbound instead. That is often the right call.
Similarly, if your contact data quality is poor (phone numbers are stale, names are wrong, addresses do not match), adding a compliant voice system will not fix that. You will end up calling the wrong person at the right number, triggering complaints, and creating the same legal exposure. Clean your data first, then automate the calling.
Building Your AI Voice Compliance Program From Scratch
If you are starting an outbound calling operation for the first time, build compliance in from day one. Start by defining your calling profile: what you are calling about, what consent you have, what states and countries you are calling into. Then map that to the specific compliance requirements. Debt collection calls have different requirements than appointment reminders; international calls have different rules than domestic.
Choose your platform based on your calling profile. Do not pick based on price or brand name; pick based on demonstrated compliance for your specific use case. If you are doing healthcare appointment reminders, find a platform used successfully by clinics. If you are doing B2B sales, ask platforms to show you compliance examples from B2B customers. Review platform pricing in light of the compliance costs you will incur, not in isolation.
Hire or contract for compliance expertise. If you do not have in-house knowledge of the TCPA, state DNC laws, and sector-specific calling rules, you need someone who does. This person does not need to be full-time, but they need to be involved in platform selection, campaign setup, and ongoing audit. The cost of one compliance consultant is less than the cost of one TCPA settlement.
Integrate your platform with your CRM and data systems from the start. Do not set up the voice system and the CRM as separate, disconnected tools. They must share contact data, consent status, and DNC flags in real time. The cost of integration work up front is less than the cost of compliance failures later. If you are evaluating platforms, integration capability should be weighted heavily in your decision.
Technology Infrastructure for Compliance at Scale
As you scale outbound calling, your compliance infrastructure must scale with it. At 10,000 calls per month, a few missed DNC checks are annoying. At 10 million calls per month, a 0.1% failure rate is 10,000 violations. Your system architecture must be built for reliability, not just functionality.
This means redundancy. If your DNC lookup service goes down, you need a backup. If your local copy of the DNC list becomes corrupted, you need a cached copy. If your connection to the third-party compliance service times out, you need a circuit-breaker that pauses dialling rather than proceeding without compliance. These redundancies add cost, but they are the difference between "we had an outage" and "we had 50,000 illegal calls during an outage."
Data pipeline integrity is also critical. If you are pulling contact lists from multiple sources (a database, a CRM, a spreadsheet), ensuring that DNC flags and consent status are consistent across all sources is complex. A single source of truth, ideally your CRM, reduces the chance that one system has stale data while another has current data. When sources conflict, compliance breaks.
At scale, you also need dedicated compliance monitoring infrastructure. Real-time dashboards showing DNC skip rates, lookup latency, error rates, and call outcomes. Alerts for unusual patterns. Regular automated audits of sample calls to verify compliance. The voice agent handles the call; the compliance infrastructure handles the proof that it was made legally. Too many operations build the dialer and forget to build the monitoring.
Common Compliance Mistakes and How to Avoid Them
The most common mistake is static list checking: uploading 10,000 numbers at the start of a campaign, checking them all against the DNC list once, then dialling for two weeks. By week two, hundreds of those numbers have been added to the federal DNC list. You are calling them illegally and have no defence. Check every number immediately before each call attempt, not once at the start.
The second mistake is checking only the federal list. Forty percent of the US population has opted into state-specific DNC lists before the federal registry existed. Checking only the federal list in California misses 15% of the protections consumers expect. Use a compliance service that covers all jurisdictions you operate in, or manually add state lists to your checking process.
The third is poor consent documentation. You think someone consented to be called because they filled out a form on your website three years ago. But they clicked a box that said they were interested in information, not that they consented to receive marketing calls. When the FTC asks to see proof of consent, that three-year-old form does not cut it. Document what each person consented to, when, and on what channel.
The fourth is no monitoring. You set up DNC checking, run a few test calls, and consider it done. No one reviews the actual compliance logs. Three months later, you discover that your DNC lookup service has been failing silently for six weeks, and you have made 50,000 unchecked calls. By then, complaints have already been filed. Monitor continuously, not once at launch.
The Relationship Between AI Voice Quality and Compliance
A high-quality AI voice agent that sounds natural and handles complex conversations is one part of a compliant operation. The other part is the invisible infrastructure that prevents illegal calls from happening in the first place. Some platforms invest heavily in voice quality and skimp on compliance. They deliver beautiful calls to DNC-protected numbers.
When evaluating a voice platform, separate voice quality from compliance infrastructure. Ask to see logs of a test campaign. Ask what their per-call compliance costs are and whether they are transparent about them. Can they show you a sample DNC audit trail? Many platforms gloss over this because it is unglamorous, but it is where the legal safety lives.
The best platforms invest equally in both. Voice quality gets the call answered; compliance infrastructure keeps the company out of court. Sysevo's approach of pairing voice AI with a built-in CRM is one example, but the market has others. The point is to find a vendor that takes compliance as seriously as they take voice quality, not as an afterthought.
Your team will also need to train on the platform's compliance features. A platform with excellent compliance architecture is worthless if your team does not know how to use it or misunderstands what it does. Ensure your vendor provides compliance training, not just voice agent training. The people running campaigns need to understand what they can and cannot do.
Future Regulatory Trends and Compliance Preparation
The FTC has signalled that TCPA enforcement will increase. Agency budgets have grown, and the number of TCPA settlements has more than doubled in five years. If you are operating on the edge of compliance, regulators are increasingly likely to find you. The cost of being first to be audited in your space is high.
State-level enforcement is also increasing. States like California and New York have their own attorneys general offices focusing on TCPA violations and unfair calling practices. A call that is legal under federal rules but violates state rules can trigger both federal and state enforcement. Your compliance infrastructure needs to be at least as strict as the strictest jurisdiction you call.
New technologies for voice calling, like voice cloning and synthetic voice agents, are attracting regulatory scrutiny. Some states are considering rules that require explicit disclosure that a caller is an AI. Your compliance roadmap should account for the possibility that states will start requiring voice disclaimers. Building in the ability to prepend a compliance statement to every AI call is easier if you plan for it now.
The rise of class action litigation means that a single compliance failure by one person can turn into a lawsuit representing thousands. Consumer advocacy groups are increasingly using social media and public pressure alongside legal action. Public-facing compliance, like transparency about who is calling and why, is becoming competitive advantage, not just legal necessity. Platforms that make it easy to comply transparently will become more valuable as litigation risk rises.
Getting Started With Compliance Automation
If you have an existing outbound calling operation and compliance is an afterthought, you have two paths: shut down the operation and clean it up, or audit what you are doing and fix the gaps. Neither is comfortable, but doing nothing is worse. Start with an audit: pull 100 random calls from the past three months and verify that each one was checked against the DNC list before dialling. If you cannot produce that evidence, your operation is in violation.
If you are starting a new operation, use a consultation with a platform vendor to understand what compliance looks like for your specific use case. Do not assume a platform has compliance built in; verify it. Ask for references from customers in your industry. Ask to see logs from a test campaign. Take compliance seriously from day one, not as an afterthought.
Budget for compliance. Account for the per-call costs of DNC checking, the infrastructure to maintain logs, the monitoring system to catch drift, and the personnel to understand and manage it. If these costs surprise you, they should. Compliance is not free, but the cost is trivial compared to the cost of even one TCPA settlement. Build it into your calling cost per hour and price your services accordingly.
Train your team. Make sure everyone involved in outbound calling understands what the rules are, why they exist, and what happens when they are broken. A single employee who does not understand consent or DNC rules can create massive liability for your company. Compliance culture starts with education. Check in with your team quarterly about compliance; make it part of your regular operations review, not something you think about once a year.
Frequently Asked Questions
Can I use an AI voice agent to call numbers on the Do Not Call list if I have a prior business relationship?
Yes, but only if the relationship genuinely exists. The TCPA defines "established business relationship" as a transaction, communication, or inquiry from the person in the prior 18 months. A prior purchase, customer service call, or website inquiry counts. A stale lead from five years ago does not. Your system must verify and log the relationship before the override is allowed.
How often should I update my DNC list data?
The federal registry updates continuously; check it in real-time before each call, not in batch. If real-time checking is not feasible, refresh your local copy at least weekly, ideally daily. Some state lists update monthly. The closer to real-time you can get, the safer you are. Static lists older than a few days are indefensible if the FTC audits you.
What records do I need to keep for TCPA compliance?
For every call, keep: the phone number dialled, the date and time, the DNC lookup result, whether the call was placed, and the outcome. If the call was not placed due to DNC, log that decision. Keep these records for at least 18 months; longer is safer. Make the logs machine-readable so you can run reports and respond quickly to audits.
Does an AI voice agent that handles DNC checking in-house cost more than one that outsources checking?
In-house checking requires you to maintain the federal and state DNC databases and handle failures yourself. Outsourced checking costs $0.01 to $0.05 per lookup but includes infrastructure, failover, and updates. For most operations, outsourcing is cheaper and safer than attempting to manage this in-house unless your call volumes exceed 10 million per month.
What happens if my DNC lookup service fails while I'm running a campaign?
Your system should immediately pause outbound calling and alert your team. Continuing to dial without DNC checks is worse than stopping. Have a runbook for this scenario: who gets notified, what channels pause, how long you wait before resuming. Log the outage and the time the calls paused. This becomes part of your compliance record if the FTC asks.
Can I use AI voice agents for cold calling without prior consent?
Cold calling to cell phones without prior express written consent violates the TCPA. You can call landlines without consent if you comply with DNC checks and maintain an internal DNC list, but cell phones are off limits unless the person has explicitly consented in writing. Many compliance experts recommend getting written consent for all outbound calling, not just cell phones, to avoid disputes.
