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 Knowledge Agent
Gaps in your help center found from real resolved tickets, and draft articles written to fill them, ready for your review.
Your agents answer the same questions every day because your help center does not. This agent scans resolved tickets to find the questions your articles miss, then drafts new articles in your brand voice from the answers your team already gave. Nothing is published without a person approving it, so your help center grows with content you trust.
- A gap report built from real resolved tickets
- Article drafts written from answers your team already gave
- Every draft reviewed by a person before publishing
$3,400
20–30 working days
Optional maintenance $249/mo
What it is
AI Support Knowledge Agent scans your resolved tickets to find questions your help center does not yet answer, then drafts new articles in your brand voice, ready for a person to approve. It delivers to one destination such as your CMS or drive.
Why businesses buy it
Help centers fall behind because writing articles always loses to answering the next ticket. The result is a cycle: missing articles create tickets, and tickets leave no time to write articles. After launch, you get a clear list of what customers ask that your help center does not answer, plus drafts ready for review. The Scope Card sets the destination, output formats and review step before you order, so you know exactly where drafts land and who approves them.
Problems this solves
Your help center misses questions customers ask every week
The agent scans your resolved tickets to find questions your help center does not yet answer. You get a clear view of the gaps, drawn from real customer conversations.
Nobody has time to write new help articles
For each gap, the agent drafts a new article in your brand voice. It follows your brand profile and up to 20 reference examples. Your team edits rather than starting from a blank page.
You worry automated articles will go live without review
Drafts are delivered to one destination, such as your CMS or drive, ready for a person to approve. Nothing is published until someone on your team says so.
A good fit if
Your agents keep answering the same question over email or chat because no help center article exists for it yet.
You have a backlog of resolved tickets that, read together, would make a very useful set of articles, if anyone had time to write them.
You already have a brand voice and some existing articles to draw a style from, and want new ones to match.
You are comfortable reviewing and approving drafts rather than needing every article to be perfect from AI on the first pass.
Not the right fit if
You need the agent to answer customer questions live, not draft articles about them. See AI FAQ Agent or AI Customer Support Agent for that job.
Your real gap is agents drafting replies inside live tickets, not building the help center. See AI Support Copilot.
You want AI to judge how well past conversations were handled, not generate new content. See AI Support QA.
You need more than three output formats or a destination system beyond one CMS, drive or store. Talk to us through our Custom AI enquiry form.
How it works
- The agent reads a batch of your resolved tickets and looks for questions that come up repeatedly without a matching help center article.
- It checks each candidate gap against your existing help center, up to twenty reference examples, so it never proposes an article that already exists.
- For each real gap, it drafts an article in your brand voice, following the tone and structure of the examples you approved.
- Drafts are delivered to the one destination you chose, such as your CMS, a shared drive, your store, or an inbox, in up to three output formats.
- Nothing publishes itself. Every draft sits with the approver you named until they review and either publish it or send it back.
- We review twenty-five sample outputs with you before launch, so the tone is dialled in before the agent runs on its own.
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.
- Your brand guidelines and a few tone-of-voice examples so drafts sound like you, not like a generic template.
- Ten to twenty existing articles or pieces of content you like, to set the style the agent should follow.
- Access to the one destination system where drafts should land, such as your CMS, store or drive.
- A starting list of products or topics you want covered first.
- The name of the person who reviews and approves every draft before it goes live.
- Admin access and API keys for every connected system, shared by adding our team as a user or creating a key you control, never by email.
- Your business information (company basics, hours and brand voice), shared during intake.
- One decision-maker who can confirm scope questions during the build.
- 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 reviewed, brand voice and topic list confirmed.
Weeks 1–2
Gap-finding and drafting pipeline built from your examples.
Week 3
Destination system connected and delivery format tested.
Weeks 3–4
Twenty-five sample outputs produced, with results sent to you.
Week 4
Your review, plus two rounds of requested changes.
Go-live
Live drafting from resolved tickets, recorded handoff session, support begins.
| Option | Delivery time | Price |
|---|---|---|
| Standard | 20–30 working days | $3,400 |
| Express | 15–25 working days | $4,000 |
| Rush | 10–20 working days | $4,600 |
What you receive at handover
- The knowledge agent configured with one generation pipeline and up to three output formats
- Brand profile and up to 20 reference examples set as the drafting guide
- Delivery connection to one destination, such as your CMS, drive, store or inbox
- Test report with 25 sample outputs reviewed and approved by you
- Admin guide, recorded 45-minute handoff session and 30 days of post-launch support
Expected benefits and how to measure them
- Your help center can keep pace with what customers actually ask.
- Writers can spend time editing drafts rather than finding topics.
- Customers can find more answers without contacting you.
- Articles published from drafts
- Count drafts approved and published each month from your CMS or help center history.
- Tickets on topics now covered
- Track ticket volume on each new article's topic. Compare four weeks before and after it goes live.
- Editing time per draft
- Ask editors to log time spent per draft for the first month, and compare with writing an article from scratch.
Before you start: Record your current article count, monthly articles published and the average time it takes to write one.
These are things to track, not promised results. Outcomes depend on your data, your team and your market.
Running costs
The build price is one-time. Running AI Support Knowledge Agent costs roughly $20 to $150 a month in AI model or media-generation usage, depending on how many articles you generate each month. Occasional use sits at the low end, weekly drafting runs sit toward the top.
AI usage during the build and afterwards runs through your own accounts, so the cost and the provider relationship both stay visible to you.
Ask about running costsCustom requirements
Need more than three output formats, a second destination system, or a broader reference set than twenty examples? Add a note at checkout for a written Change Quote, with work starting only once you accept it. Larger or bespoke projects go through the Custom AI request form instead.
Request a custom versionKeep it running
Monthly maintenance covers monitoring, fixes and change hours for work such as retuning the brand voice or adding a new topic source. It starts after the post-launch support period, bills monthly and can be cancelled any month.
$249/mo (the minimum fee, 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 | A generic AI writing tool bills per seat or per word generated | A custom agency quotes after discovery calls | Ours is one fixed price, shown on this page before you buy |
| Scope | A writing tool needs your brand voice and topics fed in by hand each time | An agency agrees scope during the project | Ours is written on the Scope Card before purchase |
| Delivery | A writing tool is instant but has no idea what your help center is missing | An agency build usually takes several weeks, set per project | Ours has a fixed delivery time shown before you buy, with faster options |
| Where topics come from | A generic tool only writes what you type in as a prompt | An agency build depends on what was agreed | Ours finds real gaps from your own resolved tickets first |
Questions about this product
How does AI Support Knowledge Agent decide what to write about?
It reads a batch of your resolved tickets and looks for questions that come up repeatedly without a matching article in your help center. Each candidate is checked against your existing content and the reference examples you approved, so it only proposes genuine gaps, not topics you have already covered.
What does AI Support Knowledge Agent cost to run each month?
Roughly $20 to $150 a month in AI model or media-generation usage, depending on how many articles you generate. Occasional, light use sits at the low end of that range, while a weekly drafting schedule sits toward the top. You pay this directly in your own accounts.
How long does AI Support Knowledge Agent take to launch?
You see the fixed delivery time before you buy, counted from a verified intake, and you can choose express or rush delivery if your launch date is close. All three include the same twenty-five sample outputs, reviewed with you, before the agent starts drafting from your live resolved tickets.
Does AI Support Knowledge Agent publish articles automatically?
No. Every draft is delivered to the person you named as approver and sits there until reviewed. Nothing publishes itself. This keeps a human check on tone, accuracy and brand fit before anything reaches customers, which matters most for help center content.
What access does AI Support Knowledge Agent need?
Access to the one destination system you choose, such as your CMS, drive or store, plus whatever holds your resolved tickets so it can find real gaps. Credentials are shared by adding our team as a user or creating a key you control, never by email, and can be revoked at any time you choose.
Can AI Support Knowledge Agent match our existing writing style?
Yes. You provide ten to twenty reference examples during intake, and we review twenty-five sample outputs with you before launch specifically to dial in tone and structure. Two rounds of changes are included, so the voice is close before the agent starts running on its own.
Can I add new topics or a different brand voice later?
Yes, through monthly maintenance. Its change hours cover work such as retuning the voice or adding a new topic source, and you can cancel any month if your needs get lighter. For a voice change, share a few fresh tone-of-voice examples. Maintenance also covers monitoring and fixes, and starts after the post-launch support period.
Guides on this topic
All guides Turn resolved tickets into the help center articles you never had time to write.
Order AI Support Knowledge Agent and the answers your team already wrote become help center drafts, ready for review, so customers find answers themselves.
Works well with
Support Operations
AI Support Knowledge Agent for Freshdesk
Finds gaps in your help center and drafts new articles from resolved tickets, connected natively to Freshdesk so it reads and updates…
Support Operations
AI Support Knowledge Agent for Intercom
Finds gaps in your help center and drafts new articles from resolved tickets, connected natively to Intercom so it reads and updates…
Support Operations
AI Support Knowledge Agent for Zendesk
Finds gaps in your help center and drafts new articles from resolved tickets, connected natively to Zendesk so it reads and updates…