Guides
Phone screening intercepts outbound calls before they connect. This guide shows you how to configure Thoughtly's Call Screening Bypass to respond to automated gates, pair it with branded calling, and stop losing leads to screening filters.
Last updated
Every outbound team knows the problem: you dial a prospect within minutes of their form submission, and instead of a ring, you get a screening prompt. "Who is calling? What company are you with? What is the reason for your call?" If your AI agent can't answer those questions clearly and quickly, the call never connects — and the lead goes cold.
Phone screening is everywhere now. iOS Call Identification, Google Pixel Call Assist, Samsung Bixby screening, Truecaller, Hiya, and carrier-level call protection all intercept unfamiliar numbers before they reach the person. For high-volume outbound teams in insurance, mortgage, home services, education, and healthcare, screening gates are one of the biggest silent killers of connect rates.
Thoughtly's Call Screening Bypass feature gives your AI agent a screening stage at the start of every outbound call. When an automated gate asks who's calling, the agent responds with the name, company, and reason you configure — then continues the normal conversation flow once the recipient picks up.
This guide walks through the setup, shows you how to write screening responses that actually get through, and covers the mistakes that quietly tank connect rates on outbound campaigns.
Before you configure anything, it helps to understand what your agent is up against. Call screening comes in several forms, and each one behaves slightly differently.
| Screening type | How it works | What the agent hears |
|---|---|---|
| iOS Call Identification | Prompts the caller to state their name; plays it back to the recipient who decides whether to accept | "Please state your name after the tone" |
| Google Pixel Call Assist | Google Assistant answers, asks the caller's purpose, and transcribes the response for the recipient in real time | "Hi, the person you're calling is using a screening service. Please state your name and why you're calling" |
| Samsung Bixby | Bixby answers, asks why the caller is reaching out, and shows the transcript to the recipient | "Please tell me who you are and why you're calling" |
| Carrier-level filtering (Verizon, T-Mobile, AT&T) | Flags or blocks calls that look like spam based on calling patterns, reputation scores, and STIR/SHAKEN attestation | Silence, fast disconnect, or "This call has been flagged as potential spam" |
| Third-party apps (Truecaller, Hiya, RoboKiller) | Intercept unknown calls, ask for a reason, or send straight to voicemail based on crowdsourced spam databases | Varies — may ask a screening question or silently block |
The key insight: most screening prompts ask the same three questions in slightly different phrasing. Who are you? What company? Why are you calling? If your agent can answer those clearly, it passes the gate.
In Thoughtly, Call Screening Bypass is configured per agent or per phone number. When enabled, the agent begins with a short screening stage before the main conversation flow. During this stage, it listens for screening prompts and answers with the values you provide.
You'll configure three fields:
| Field | What to enter | Example |
|---|---|---|
| Agent name | The name the agent should give when asked who is calling | Ava |
| Company name | Your business name — what the prospect would recognize from the form they filled out | Acme Home Services |
| Reason for calling | A short, specific, truthful statement about why you're calling | following up on your quote request from this morning |
After the screening interaction completes and the recipient answers, the agent transitions into the normal start node and continues the conversation as designed.
The reason-for-calling field is the one that makes or breaks your connect rate. Screening services display or read back this text to the prospect, and that single sentence determines whether they tap "Accept" or "Decline." Here's how to get it right.
The prospect took an action — they filled out a form, requested a quote, started an application, booked a consultation. Reference that action directly.
Good examples:
Bad examples:
If the prospect filled out a form on "QuickQuote Insurance," your company name field should say "QuickQuote Insurance" — not your parent company name, not your holding company, not your agency group. The recipient is deciding in two seconds whether to pick up. Name recognition is everything.
Screening services truncate long responses or make them sound unnatural when read aloud. A short, clear reason wins every time. If you can't explain why you're calling in one sentence, the screening response isn't the problem — your outbound trigger is.
Call Screening Bypass and Branded Calling solve different parts of the same problem. Screening bypass handles the automated gate. Branded calling handles the caller ID display. When both are active, the prospect sees your verified business name on their screen and hears a clear, specific reason for the call if their device screens first.
Together, they form a two-layer trust stack:
| Layer | What it does | Where it helps |
|---|---|---|
| Branded Calling | Displays your verified business name and identity on the recipient's caller ID | Before the prospect decides to answer — reduces "Unknown caller" rejection |
| Call Screening Bypass | Responds to automated screening prompts with your agent name, company, and reason for calling | During the screening gate — gives the prospect enough context to accept the call |
If you're running high-volume outbound campaigns — speed-to-lead follow-up, re-engagement sequences, appointment confirmations — enable both. See Branded Calling and Spam Label Prevention for AI VoiceAI voiceAn artificially generated, natural-sounding voice produced by a TTS model. Thoughtly supports a library of AI voices and brand-specific cloning. Agents for the full branded calling setup guide.
Call screening behavior varies by device, carrierCarrierA telecommunications provider that routes phone calls and SMS over its network. Twilio, Telnyx, and Bandwidth are the three most common in the AI voice space., and app. Before you push a campaign to thousands of contacts, verify that your screening responses actually work.
If the screening response sounds unnatural or gets cut off, shorten it. If the agent skips straight to the main flow without responding to the gate, confirm screening bypass is enabled on the correct agent or number.
Different outbound motions call for different screening responses. The reason-for-calling field should match the campaign context, not a generic default.
| Campaign type | Company name example | Reason for calling example |
|---|---|---|
| Speed-to-lead (form fill) | Acme Home Services | following up on your quote request from a few minutes ago |
| Appointment confirmation | Bright Smile Dental | confirming your appointment for tomorrow morning |
| Re-engagement (aged lead) | Pacific Mortgage Group | checking in about the pre-approval you started last month |
| Enrollment follow-up | Westfield University Admissions | following up on your application for fall enrollment |
| Insurance quote follow-up | Shield Insurance | calling about the auto insurance quote you requested |
| Post-service follow-up | Summit HVAC | following up after your furnace repair last week |
If you're running multiple campaign types from the same workspace, you may want separate agents — each with its own screening configuration — rather than a single agent with a generic reason. Thoughtly's Agent Builder makes it straightforward to clone and customize agents for different outbound motions.
"Following up on your recent inquiry" sounds fine until you realize it's the same phrase telemarketers use. Prospects screening calls have heard it hundreds of times. Reference the specific action — the form, the quote, the application — so your agent sounds like it belongs.
You clone your insurance agent to build a mortgage agent, launch a campaign, and your screening response still says "following up on your home insurance quote." Every cloned agent needs its screening fields reviewed before the first call.
Screening bypass only helps once the screening gate activates. If the prospect sees "Spam Likely" on their caller ID and declines before screening even starts, your bypass configuration never fires. Branded calling addresses the caller ID layer. Use both.
If your screening reason says "following up on your roof estimate" but your agent's start node opens with "Hi, I'm calling about an exciting opportunity," the disconnect erodes trust immediately. Align the screening reason with the first thing the agent says after connection.
iOS, Pixel, and Samsung handle screening differently. A response that sounds natural on iPhone may get truncated on Pixel Call Assist. Test across at least two device types before scaling.
Call screening bypass doesn't have its own dashboard — you measure it through the downstream metrics it affects.
| Metric | What to track | Where to find it |
|---|---|---|
| Connect rate | Percentage of outbound calls that reach a live person (not voicemail, not blocked, not screened out) | History — filter by status: Completed vs. No Answer / Left Voicemail |
| Screening pass-through rate | Of calls that hit a screening gate, how many got accepted by the prospect | History transcripts — look for screening prompt text followed by normal conversation |
| Answer-to-conversation ratio | Percentage of answered calls where the agent completed at least one meaningful exchange | History — filter Completed calls, check transcript length |
| Campaign-level contact rate | Contacts reached as a share of total contacts dialed in a batch campaign | Analytics + History export — compare dialed vs. Completed |
Run a before/after comparison. If you're enabling screening bypass on an existing outbound campaign, pull one week of connect-rate data before and one week after. The difference isolates the screening bypass impact from other variables.
For broader performance tracking, see How to Use Thoughtly Analytics to Optimize Agent Performance.
No. Call screening bypass helps your agent respond to automated screening prompts, but the prospect still decides whether to accept the call. If they see an unfamiliar number and decline before the screening service even asks, the bypass doesn't activate. That's why pairing it with branded calling — which addresses the caller ID display — is the recommended approach.
Yes. Screening bypass can be configured per agent or per phone number. If you're running separate campaigns from different numbers — for example, one number for insurance follow-up and another for mortgage leads — each can have its own screening configuration.
The agent skips the screening stage and goes straight to the normal conversation flow. Screening bypass is additive — it only activates when the agent detects a screening prompt. Calls that connect directly are unaffected.
Screening bypass responds to the prompts that Google Pixel Call Assist asks, just like it responds to iOS screening or Samsung Bixby. The agent provides its name, company, and reason. The Pixel transcribes this in real time for the recipient, who then decides whether to accept.
Check your History transcripts. If you see calls ending after a brief exchange that looks like a screening prompt — or calls with no transcriptTranscriptThe text record of a voice conversation, used for review, training, compliance audit, and search. at all that show "No Answer" — screening and spam filtering are likely contributors. Compare connect rates on numbers with and without branded calling enabled. If connect rates are low even on branded numbers, the issue may be timing, consent freshness, or list quality rather than screening.
