Customer Support and Voice AI
AI receptionist for small business: calls, bookings and handoffs
What an AI receptionist does on a real call, the rules that apply, a sample greeting and a pass/fail checklist to launch one safely.
Inbound
AI Inbound Call Agent for Aircall
Inbound calls on your Aircall line worked through to an outcome, then tagged and filed against the right Aircall contact.
Your team lives in Aircall, but the phones still ring when everyone is busy. This agent answers your Aircall number, recognizes callers from your Aircall contacts and works up to five defined call types through to an outcome. Each call is tagged with what happened and keeps its recording, so it sits in the same call history your team already checks.
- Recognizes callers from your existing Aircall contacts
- Tags every AI-handled call by outcome in Aircall
- Works defined call types through to a recorded outcome
$6,400
20–30 working days
Optional maintenance $320/mo
What it is
AI Inbound Call Agent for Aircall answers a number on your Aircall account and handles up to five defined call types from greeting to outcome. It looks up callers among your Aircall contacts, applies call tags for what happened, and keeps recordings with the call, on a mapping agreed with your admin.
Why businesses buy it
Calls that ring out or wait in a queue cost you customers, and when overflow goes to a separate bot, its calls vanish from Aircall reporting. This edition keeps everything in the tool your team trusts. Repeatable call types are handled end to end, outcomes appear as the call tags you already report on, and anything unusual is passed to a colleague. The Scope Card lists the Aircall data used, the flows and the second system before you order, so the build is a known quantity, not an open project.
Problems this solves
Repetitive Aircall calls follow the same steps but still need an agent
The agent answers a number on your Aircall account and handles up to five defined call types. Each call runs from greeting to outcome without an agent having to pick up.
Agents ask returning callers for details Aircall already holds
The agent looks up each caller among your Aircall contacts before it starts. It can handle the call with that context instead of asking for the same details again.
Aircall reports cannot show what each call actually achieved
Call tags are applied for what happened on every call the agent handles. Your Aircall analytics can then show how many calls ended in each outcome.
Admins hesitate to let a new tool change Aircall data
The field mapping is agreed with your Aircall admin and documented in full. Access comes through user invites or keys you control, and testing runs on sandbox or test records.
A good fit if
Your team already works in Aircall, and you want AI-handled calls to show up in the same call history they check every day.
You sort calls with Aircall call tags, and you want the AI to tag its own calls by outcome so your reporting stays consistent.
Many of your callers are already Aircall contacts, so the agent can recognize them before asking a single question.
A team member with Aircall admin rights can create an API ID and token and join a 30-minute call.
Not the right fit if
Callers mainly want your hours, a booking or a transfer. See AI Voice Receptionist for Aircall, which is built around front-desk work.
Your numbers sit in a programmable telephony account rather than Aircall. See AI Inbound Call Agent for Twilio.
You want your own team to keep answering, with calls summarized afterwards. See AI Call Summarization.
Your callers need more than one language. This build covers one; add the Additional Language Pack, for a second.
How it works
- The mapping is agreed first. Your Aircall admin joins a 30-minute call where we settle which number the agent answers, which call tags it may apply and which contact fields it can update.
- The caller is looked up. The incoming number is checked against your Aircall contacts, so a known customer moves into the right flow without repeating basic details.
- One of up to five flows takes over. Each follows the questions and answers you approved, drawing on up to 50 documents or an FAQ set and the one other system you connected.
- People step in when needed. A caller who asks for someone, or a question outside every flow, is transferred live to the number you set for that call type.
- Aircall gets the outcome. The call is tagged with its result, the recording stays attached, and a first-time caller can be added as a contact where the mapping allows.
- Your team reviews AI-handled calls in the same Aircall call history as their own, sorted by the tags agreed at setup.
What you provide within 72 hours
Your portal lists these the moment you order. Delivery starts the working day after everything is in and your access is checked.
- An Aircall API ID and token, created by your Aircall admin.
- That admin's time for a 30-minute setup call.
- The Aircall number the agent should answer.
- The call tags you use today, or the ones you want the agent to apply.
- Scripts, FAQs and business hours for each call type, with a transfer number for each.
- Access to the one other system, usually your calendar or CRM.
- Your chosen voice, your recording disclosure wording, and a single decision-maker.
- An API key for your own AI model account (for example OpenAI or Google Gemini), created for this build with a spending limit.
The 72-hour rule. Send everything on your checklist, including access, within 72 hours of ordering so your build starts on time.
Timeline
Start
Intake checked, Aircall admin call held, tags and flows listed.
Weeks 1–2
Flows, voice and FAQ answers built; Aircall mapping written up.
Week 3
Contact lookup, call tags and the second system connected.
Weeks 3–4
Fifty test calls run, with transcripts and tag results shared.
Week 4
Two rounds of changes after your review.
Go-live
Answering your Aircall number, recorded handoff, support starts.
| Option | Delivery time | Price |
|---|---|---|
| Standard | 20–30 working days | $6,400 |
| Express | 15–25 working days | $7,350 |
| Rush | 10–20 working days | $8,300 |
What you receive at handover
- The AI Inbound Call Agent live on a number in your Aircall account
- An Aircall connection covering contacts, call tags and recordings, with a documented field mapping
- Up to five configured call types and a connection to one more system
- Access set up through user invites or keys you control, with none sent by email
- A test report on sandbox or test records, an admin guide and a recorded handoff session
- 30 days of post-launch support
Expected benefits and how to measure them
- Can free Aircall agents from calls that follow a fixed pattern.
- Helps Aircall analytics show outcomes, not just call counts.
- Makes returning callers feel recognized.
- Calls completed by the agent
- Filter Aircall analytics by the agent's tags. Compare with agent-handled volumes for the same call types before launch.
- Calls tagged with an outcome
- Count tagged calls in Aircall before launch. Recount monthly after launch.
- Average agent handling time
- Use Aircall analytics. Compare the four weeks before launch with the four weeks after.
Before you start: Export four weeks of Aircall analytics with handling times and tags. Note which call types follow the same steps each time.
These are things to track, not promised results. Outcomes depend on your data, your team and your market.
Running costs
You pay once for the build. Call minutes are billed in your own accounts at roughly $0.13 to $0.31 per minute all in, covering the voice platform, AI model and telephony, plus number rental. Your Aircall subscription carries on as it is; this build does not include Aircall licenses or plan changes.
AI usage while we build and test also runs through your own accounts, never ours.
Ask about running costsCustom requirements
Need a second Aircall number, more than five flows, or Aircall objects beyond the agreed mapping? Add a checkout note and a written Change Quote follows.
Request a custom versionKeep it running
Monthly maintenance is recommended, with monitoring, fixes and monthly change hours for things such as a new call tag or a reworded flow.
$320/mo (5% of the build price a month, includes 3 change hours)
Optional. Starts after the post-launch support period. Cancel any month.
How maintenance worksHow this compares
| Do it yourself with a SaaS tool | Hire a custom agency | Upmark Labs | |
|---|---|---|---|
| Price | An AI add-on for a phone system is often billed per seat or per minute | A phone-system consultant quotes per project | Ours is one fixed price, shown on this page before you buy |
| Call tags | An add-on may use its own labels instead of yours | Set during the project | Ours applies the Aircall tags you agree at setup |
| Contacts | Lookups depend on the add-on's own sync | Depends on the brief | Ours reads and updates Aircall contacts on a documented mapping |
| Delivery | Switching on is quick, shaping the flows takes longer | Set per project | Ours has a fixed delivery time shown before you buy, with faster options |
Questions about this product
Which Aircall data does the inbound call agent use?
Calls, recordings, contacts and call tags. It reads contacts to recognize callers, handles the call in one of your flows, then tags the call with its outcome and keeps the recording attached. Which contact fields it may update is set in the mapping your Aircall admin approves. Custom Aircall objects are not included.
Will AI-handled calls appear in our Aircall call history?
Yes. Calls the agent answers sit in Aircall next to the calls your team takes, with recordings attached and the agreed call tags applied, so your team reviews them without opening another tool. A transcript and summary can also go to the one other system you connect, such as your CRM.
Can the inbound agent use the call tags we already have in Aircall?
Yes. During the 30-minute setup call we list the tags you already use and agree which ones each flow may apply, such as a tag for a completed order check. New tags can be added at the same time. Changing tags after launch fits inside the monthly change hours that come with maintenance.
What Aircall access is needed for the inbound agent?
An Aircall API ID and token, created by your Aircall admin, plus that admin for a 30-minute setup call. Credentials are never collected by email. Testing runs on sandbox or test records first, so nothing touches your live Aircall data until the test calls have been reviewed with you.
How does this differ from AI Voice Receptionist for Aircall?
The receptionist is built around front-desk work: routine questions, bookings and transfers. This edition is for defined call types handled end to end, such as an order check or a dispatch request, where each flow has its own questions, lookups and tagged outcome.
What do the faster delivery options cost for the Aircall edition?
Each faster option is priced in the delivery panel on this page, next to the standard timeline, so you choose before you pay. Every option starts the working day after your Aircall API ID, token, call scripts and transfer numbers are verified, so finishing intake within 72 hours keeps the date. The scope, tags and testing stay the same at every speed.
Do we keep paying Aircall as normal?
Yes. Your Aircall subscription stays with Aircall and is not part of this build, and neither are license or plan upgrades. Call minutes for the AI run roughly $0.13 to $0.31 per minute all in, plus number rental, billed to your own accounts. Upmark charges nothing per call.
Guides on this topic
All guides Let Aircall show what happened on every call.
Order AI Inbound Call Agent for Aircall and on a fixed delivery date your Aircall number handles defined call types end to end, with each call tagged, recorded and matched to its contact.
Works well with
Call Intelligence
AI Call Summarization
Summarizes calls into notes, outcomes and next steps in your CRM.
Inbound
AI Voice Receptionist
Answers calls, handles common questions, takes messages and transfers callers.
Inbound
AI Voice Customer Support
Resolves support calls using your knowledge base with human transfer.