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.
Support Operations
AI Support Copilot for Intercom
Teammates in Intercom get a drafted answer and the matching help center article while the customer is still in the chat.
Live chat customers expect quick answers, and your teammates feel that pressure in every conversation. This copilot drafts a suggested reply in each Intercom conversation a teammate picks up, using your help center articles, policies and the customer's contact attributes. The draft and its source appear as a note, your tags choose the right playbook, and the teammate edits and sends every message.
- Drafted replies ready while the customer is still in chat
- Help center source and customer context shown alongside
- Teammates edit and send every message themselves
$5,900
20–30 working days
Optional maintenance $295/mo
What it is
AI Support Copilot for Intercom drafts a suggested reply in each Intercom conversation a teammate picks up, using your help center articles, policies and the customer's contact attributes. The draft and its source article appear as a note, and the teammate edits and sends. Your tags choose the playbook.
Why businesses buy it
In live chat, every minute a customer waits feels longer, and teammates juggling several conversations end up searching for articles while the customer drifts away. Quality varies with whoever is online. After launch, teammates open each conversation to a grounded draft, the right article and the customer's context, so replies are faster and consistent without losing the human voice. The Scope Card lists the Intercom fields, sources and playbooks before you order, so you know exactly what the copilot suggests and that customers only hear from your team.
Problems this solves
Teammates search the help center while the customer waits
When a teammate picks up an Intercom conversation, a suggested reply is already waiting as a note. It draws on your help center articles and policies, with the source article shown alongside.
Answers ignore who the customer is or which plan they hold
Drafts use the customer's Intercom contact attributes through the agreed field mapping. A reply can reflect the person asking, not only the question, and the teammate edits before sending.
Different topics need different handling but replies all sound alike
Your Intercom tags choose the playbook each draft follows, across up to six workflows with your approval rules. A billing question and a bug report can be drafted in quite different ways.
A good fit if
Live chats in Intercom wait while teammates hunt for the right help center article to answer from.
Contact attributes such as plan or company change the right answer, and teammates forget to check them.
You want quicker chat replies, with a person still deciding every message the customer sees.
You can create an Intercom app access token, and an admin can join a thirty-minute setup call.
Not the right fit if
You want routine conversations resolved without a teammate. See AI Customer Support Agent for Intercom.
Conversations land with the wrong team inbox before anyone replies. See AI Ticket Routing for Intercom.
New customers need structured onboarding conversations. See AI Customer Onboarding Agent for Intercom.
Help center gaps keep forcing teammates to write answers from scratch. See AI Support Knowledge Agent for Intercom.
How it works
- Your Intercom admin and we agree the mapping on a thirty-minute call: conversations, contacts, tags and help center articles to read, and notes to write.
- When a teammate opens a conversation, the copilot reads the messages so far and the contact's attributes, such as plan or company.
- Tags on the conversation decide which of your playbooks applies, so a billing chat and a bug report get different guidance.
- A reply is drafted only from your help center articles, policies and approved documents, and posted as a note with a link to the article it used.
- The teammate edits, sends or ignores the draft. The customer never receives anything the teammate has not sent.
- Account actions are prepared for the teammate to confirm under your approval rules, after fifty test scenarios on test conversations.
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 Intercom app access token, created by your admin.
- Your Intercom admin for the thirty-minute mapping call.
- Your help center articles, policies and the tags you use to sort conversations.
- Playbooks for your main conversation types, and approval limits for account actions.
- Ten real conversations with the replies your team sent.
- Escalation rules, support hours and the approver for exceptions.
- A decision-maker for draft tone and playbook scope.
- 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; Intercom admin call booked.
Weeks 1–2
Mapping agreed; tag-based playbooks and drafting built.
Week 3
Intercom and the second system connected; notes tested.
Weeks 3–4
Fifty conversation scenarios run, drafts shared.
Week 4
Teammate feedback; two rounds of changes.
Go-live
Live in your inbox; recorded 60-minute handoff; support begins.
| Option | Delivery time | Price |
|---|---|---|
| Standard | 20–30 working days | $5,900 |
| Express | 15–25 working days | $6,800 |
| Rush | 10–20 working days | $7,650 |
What you receive at handover
- AI Support Copilot configured and working inside your Intercom workspace
- A documented field mapping for the two-way Intercom connection, plus up to one more system
- Access through user invites or keys you control, never sent by email
- A test report from 50 documented scenarios on sandbox or test records
- An admin guide plus the recorded 60-minute handoff session
- 30 days of post-launch support
Expected benefits and how to measure them
- Teammates can reply faster without leaving the conversation to look things up.
- Makes it easier to give each customer an answer that fits their account.
- Helps your tags and playbooks shape replies consistently across the team.
- Median first response time
- Pull Intercom's conversation reports and compare four weeks before launch with four weeks after.
- Conversations closed per teammate
- Use Intercom's teammate performance report for the month before and the month after go-live.
- Conversation ratings
- Compare the share of good ratings in Intercom's customer satisfaction report across the same two periods.
Before you start: Save a month of Intercom reports on response time, closed conversations and ratings. List the tags that drive most of your volume.
These are things to track, not promised results. Outcomes depend on your data, your team and your market.
Running costs
You pay for the build once. Running it costs roughly $50 to $250 a month for AI model usage and hosting in your own accounts, on top of your Intercom subscription. Heavy live-chat volume pushes usage toward the higher figure.
Usage during testing is billed to you directly.
Ask about running costsCustom requirements
Several Intercom workspaces, a third connected system, or more than six playbook workflows need a written Change Quote. Leave a note at checkout and the quote arrives.
Request a custom versionKeep it running
Monthly maintenance covers monitoring, fixes and change hours for work such as a new tag playbook or updated policies after a release. It starts after the post-launch support period, bills monthly and can be cancelled any month.
$295/mo (5% of the build price a month, includes 2 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 | Built-in assist features bill per seat or resolution | An Intercom partner quotes per project | Ours is one fixed price, shown on this page before you buy |
| Customer context | Drafts from the message alone | Depends on the brief | Ours reads plan, company and tags before drafting |
| Control | May send low-risk replies automatically | Agreed during the project | Ours drafts notes; a teammate sends every reply |
| Delivery | On in minutes, tuned over weeks | Set per project | Ours has a fixed delivery time shown before you buy, with faster options |
Questions about this product
Is AI Support Copilot for Intercom fast enough for live chat?
It is designed for it. Drafting starts the moment a teammate opens the conversation, so the suggestion arrives while the chat is still live. How quickly depends on message length and your AI provider, which is why the fifty test scenarios run on your own conversations, letting you judge the pace before launch.
How do Intercom tags change what the copilot drafts?
Each tag you use to sort conversations can point to a playbook, such as billing, bug report or account access. The copilot reads the tags on the conversation and applies the matching playbook, so drafts follow the right rules. Untagged conversations use your general guidance.
Can the Intercom copilot see a customer's plan or company?
Yes, from the contact attributes named in the mapping. That lets a draft reflect what the customer's plan includes, for example. If billing details live in a separate tool such as Stripe, that tool can be the second connected system, so drafts quote real figures.
Will customers ever receive a message straight from the Intercom copilot?
No. Drafts are posted as notes inside the conversation, which customers cannot see. A teammate copies, edits or ignores each one and sends their own reply. If you want routine conversations answered with no teammate involved, AI Customer Support Agent for Intercom is the build for that.
What does the Intercom copilot cost with express or rush delivery?
One fixed price covers the whole build and is shown on this page before you buy, with express and rush delivery listed beside it at a set extra cost. Each option includes the documented mapping, fifty conversation scenarios, two rounds of changes and a recorded 60-minute handoff for your support lead.
What happens to the copilot after we change a feature or price?
Update the help center article or policy, and the next drafts use the new version, because the copilot only drafts from approved sources. Bigger changes, such as a new playbook for a new product area, fit inside the monthly change hours that come with maintenance.
Guides on this topic
All guides Answer live chats faster without letting go of the reply.
Order the Intercom edition and teammates see a drafted reply, the right article and the customer's context in every conversation, and still send each message themselves.
Works well with
Support Operations
AI Ticket Routing
Routes tickets to the right team or agent based on content and skills.
Support Operations
AI Support Knowledge Agent
Finds gaps in your help center and drafts new articles from resolved tickets.
Support Operations
AI Support QA
Scores conversations against your quality rubric and highlights coaching points.