Questions, answered

Typically 2-3 weeks from kickoff to live calls. This includes script writing, voice selection, CRM integration, calendar connection, testing, and a quality review round. Complex multi-language or multi-location builds may take 4 weeks. You'll know the exact timeline before we start.

No. We use natural-sounding neural voices and conversational scripts - not rigid IVR menus. Most callers don't realize they're talking to AI. You can listen to a real call recording before committing to anything.

The agent does a warm transfer to a live team member and passes along a full call summary so the customer doesn't have to repeat themselves. You control the transfer rules - by call type, time of day, caller intent, or custom triggers you define.

Yes - that's a core feature. The agent checks real-time availability and books directly into your calendar system during the call. It works with Google Calendar, Calendly, Acuity, ServiceTitan, and most scheduling tools via API. No follow-up call needed, no double-booking.

Yes. Outbound campaigns are a separate use case we build for: quote follow-up sequences, appointment reminders, recall campaigns for existing customers, and lead reactivation. These run on a schedule you define and can be paused or triggered manually.

Yes - 24/7/365. The agent handles every call with the same quality whether it's 2pm on a Tuesday or 11pm on Christmas Eve. You can configure different behaviors by time of day - for example, offering emergency-only booking after hours vs. full scheduling during business hours.

Yes - unlimited simultaneous calls with no degradation in performance. Every caller gets answered on the first ring regardless of how many calls are coming in at once. This is one of the biggest advantages over a human receptionist or answering service.

Yes. We support Spanish, French, and other languages. The agent can detect the caller's language preference and switch automatically, or you can set up dedicated Spanish-language agents on a separate number. This is especially popular with HVAC and home services businesses in bilingual markets.

Setup is a one-time project fee of $2,500 to $5,000 depending on complexity (number of call flows, integrations, outbound sequences). Monthly platform and maintenance runs $500 to $1,500/mo based on call volume. There are no per-minute charges on our end - we absorb usage within the monthly fee.

Most clients see positive ROI within 18-30 days of go-live. A plumbing company answering 5 additional after-hours calls per week at $350/job average captures $7,000/month in previously lost revenue - against a $1,000/mo platform fee. The math typically works well within the first month.

We don't do free trials of production systems, but we do offer a risk-limited engagement structure: you pay the setup fee, and if you're not satisfied with the agent after a 30-day live period, we'll rebuild the affected flows at no additional charge. We also show you a working demo of a similar deployment before you commit.

The setup fee is a one-time charge. Monthly platform fees are month-to-month with 30 days' notice to cancel. We don't lock you into annual contracts. Most clients stay because the system is performing, not because they're contractually obligated.

Monthly fee covers: platform hosting, call handling, AI model usage (no per-call charges), minor script updates, integration monitoring, and access to the analytics dashboard. Major new flow builds or additional integrations are scoped separately.

After you sign off on scope, we do a 60-90 minute kickoff call to capture your call flows, business rules, common questions, and tone preferences. We then build the first version of the agent, do internal testing, and send you a demo recording to review. After your feedback round, we go live. You don't need to write scripts or technical specs - we handle all of that.

The kickoff call is the most time-intensive part - about 60-90 minutes. After that, you'll spend maybe 30 minutes reviewing the demo recording and giving feedback. Post-launch, it's essentially zero - the system runs itself. Most clients spend less than 30 minutes per month reviewing analytics dashboards.

Yes - minor updates like adding new services, changing pricing info, or adjusting transfer rules are included in the monthly fee. We typically turn these around within 48-72 hours. Major flow additions (new call types, outbound campaigns) are scoped as small add-on projects.

Our primary focus is US-based service businesses, but we've worked with clients in Canada and Australia. International deployments may have different telephony setup costs and timelines. Reach out and we'll assess fit for your specific location.

We build the full stack: voice agents, workflow automation (Make.com, n8n, custom), chatbots, AI-powered data pipelines, and custom AI integrations. Voice AI tends to be the highest-ROI starting point for most service businesses, but many clients add automation layers after the voice system is running. See our full services page.

HubSpot, Salesforce, ServiceTitan, Jobber, Google Calendar, Calendly, Acuity, and most major CRM and scheduling platforms via direct API. We also support Zapier and webhook-based integrations for custom systems. If you're unsure whether your stack is compatible, mention it on our discovery call and we'll check before you commit.

Yes - we can forward calls from your existing number to the AI agent, or we can provision a new number if you prefer. Most clients use call forwarding so there's no need to publish a new number to customers. The transition is invisible to callers.

Call recordings and transcripts are encrypted at rest and in transit. For healthcare clients, we build within HIPAA-aware frameworks: call summaries avoid storing PHI, transfers to human staff handle sensitive clinical conversations, and we can execute BAAs as needed. We don't store PHI in our platform logs.

We build on top of Retell AI's voice platform, which uses real-time speech-to-text, large language models for response generation, and text-to-speech neural voices. The specific models are configurable and we select the best combination for your latency and quality requirements. We are model-agnostic and update as better options become available.

Response latency runs 500-900ms end-to-end in good conditions - similar to a slight satellite delay. Callers rarely notice or comment on it. We optimize this at the infrastructure level and test under realistic call conditions before launch. You'll hear the actual latency on the demo call we do before you commit.

We configure fallback routing so calls forward to your business number or voicemail if the AI platform is unreachable. You're never left with dead phones. We monitor uptime via our dashboard and alert you within 15 minutes of any degraded service.

Every client gets a shared Slack channel with direct access to the team that built their agent. We monitor performance dashboards weekly and proactively flag issues. Response time is same-business-day for non-urgent requests and within 2 hours for anything affecting live calls.

Yes - you get a live dashboard showing call volume by day/hour, booking conversion rate, transfer rate, top call reasons, and agent performance metrics. Call recordings and transcripts are available for any call. We also send a weekly summary email with trend data.

Yes - every call is recorded and transcribed (with appropriate disclosure to callers). You can search, filter, and playback any call from the dashboard. This is useful for quality review, training your team on common questions, and spotting issues with the agent early.

30 days' written notice to cancel the monthly platform fee. Setup fees are non-refundable (we've delivered the build). We'll give you full exports of your call data, transcripts, and configuration. No hostage tactics - you own your phone numbers and customer data.