Knowledge Assistants
AI Company Knowledge Hub for Agencies
Account teams answer what is in scope, what the client approved and how onboarding works from one cited source.
Agency account teams lose hours confirming what a retainer covers, what the client signed off and how onboarding works, because the answers sit in statements of work, email threads and project tool comments. This hub answers those questions from your own project tool, Google Workspace, CRM and ad platforms, with the source attached. Every account lead gives the same answer, and client data stays in your accounts.
- Scope questions answered from the actual statement of work
- Client approvals found in seconds, not weeks of scrolling
- Every account lead runs onboarding the same way
$12,400
20–30 working days
Optional maintenance $620/mo
What it is
AI Company Knowledge Hub for Agencies is a cited Q&A assistant across your project tool, Google Workspace, CRM and ad and analytics platforms, built for client onboarding, reporting, approvals and scope tracking questions. Account and delivery teams get answers with the source attached. Client files stay with the right account team.
Why businesses buy it
Scope creep and inconsistent answers quietly eat agency margin. When an account manager cannot quickly check what a retainer covers, extra work slips through unbilled, and when approvals are buried in email, disputes take days to settle. This hub gives account and delivery teams cited answers from your own documents, so scope conversations happen with the source in hand and new staff onboard clients the way your best lead would. The Scope Card fixes the sources, users, permission groups and testing before you order, so the build itself never creeps.
Problems this solves
Account teams dig through folders for client brand guidelines and briefs
The hub answers questions across your project tool, Google Workspace, CRM and ad and analytics platforms. Account and delivery teams get the brief or guideline with its source attached.
Scope creep goes unnoticed because nobody can check the original agreement
Workflows for scope tracking and approvals let staff ask what was agreed for a client. The answer cites the source document, so scope conversations start from the record.
Client files must stay with the account team that owns them
Client files stay with the right account team through up to three permission groups. The Agencies content source map records which team can see which client sources.
A good fit if
Account managers spend billable hours answering what a retainer covers, and the answer is in a statement of work nobody can find quickly.
Client approvals land in email threads and project tool comments, so confirming what was signed off means scrolling back through weeks.
Every new client onboarding repeats the same questions about access, assets and timelines, answered differently by each account lead.
Reporting questions, such as which metrics a client receives and when, depend on documents spread across Google Workspace and your CRM.
Not the right fit if
You mainly want brand guidelines, campaign learnings and creative templates made searchable. See AI Agency Knowledge Base.
You need client access, assets and briefs collected during onboarding, not just onboarding questions answered. See AI Agency Client Onboarding.
You want white-label client performance reports built for you. See AI Agency Reporting.
You want sales, proposals, onboarding, reporting and project management run as one system. See the AI Agency OS bundle, with one combined intake.
How it works
- Up to 3 sources connect, picked from your project tool, Google Workspace, your CRM, or an ad or analytics platform, with up to 2,000 documents and records indexed.
- The Agencies intent library recognizes client onboarding, reporting, approvals and scope tracking questions, in the way clients and account teams phrase them.
- Ask whether a request sits inside a client's retainer and the hub returns the relevant clause from the statement of work, cited, so the account manager can check the wording.
- Approval questions return the email, comment or document where sign-off was recorded, with its citation, rather than someone's memory of it.
- Up to 3 permission groups keep commercial terms with leadership and account leads, while delivery teams see briefs and process documents.
- 60 benchmark questions from agency scenarios run before launch, and the monthly refresh picks up new statements of work and onboarding material.
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.
- Access to up to three sources, such as your project tool, Google Workspace and CRM, or an ad or analytics platform.
- Your team list by group, for example leadership, account management and delivery, with the client material each may see.
- 30 sample questions with expected answers, covering onboarding, reporting, approvals and scope.
- Your agency policies and service information: retainer structures, onboarding checklists, reporting schedules and approval rules.
- Admin access to each source and a workspace admin who can set up sign-in.
- One agency lead to approve the groups and the benchmark answers.
- 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 verified; sources, team groups and sample questions agreed.
Weeks 1–2
Statements of work, onboarding material, approvals and reporting documents indexed by group.
Week 3
Agencies intents, citations and your team's interface connected.
Weeks 3–4
60 benchmark questions from agency scenarios run, accuracy report delivered.
Week 4
Your feedback and two rounds of changes.
Go-live
Live with a recorded handoff; 30 days of support follow.
| Option | Delivery time | Price |
|---|---|---|
| Standard | 20–30 working days | $12,400 |
| Express | 15–25 working days | $13,650 |
| Rush | 10–20 working days | $14,900 |
What you receive at handover
- The AI company knowledge hub for Agencies, live in a web app, Slack or Microsoft Teams
- An Agencies content source map and test question set
- Connections to up to three sources, such as your project tool, Google Workspace, CRM or ad and analytics platforms
- Cited Q&A with up to three permission groups and a monthly content refresh
- A test report using agency scenarios, with sixty benchmark questions and an accuracy report
- An admin guide, the recorded 60-minute handoff session and 30 days of post-launch support
Expected benefits and how to measure them
- Can make client onboarding smoother for new team members
- Helps account managers spot out-of-scope requests early
- Keeps client information separated by account
- Time to brief a new team member
- Ask account leads to estimate onboarding time for new joiners before launch and after.
- Change requests raised against scope
- Count change requests raised against agreed scope each month in your CRM or project tool.
- Internal 'where is' questions
- Count requests for briefs and guidelines in team chat for two weeks before and after launch.
Before you start: Record how many change requests you raised last quarter. Count requests for briefs and guidelines in chat for two weeks.
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 costs of roughly $50 to $300 a month for AI model usage, the vector database and hosting are charged to your own accounts, rising with the size of the indexed client library and how often the team asks. Project tool, Google Workspace and CRM plans are unchanged.
AI usage during build and testing is billed the same way.
Ask about running costsCustom requirements
A fourth source, more than 2,000 documents, or more than three permission groups can be added through a note at checkout. You get a written Change Quote, and that work starts once you accept it. Larger or bespoke projects go through the Custom AI request form.
Request a custom versionKeep it running
Monthly maintenance for this edition covers monitoring, fixes, model and platform updates, a monthly health report and monthly hours for changes, such as indexing a new client's onboarding pack. It starts after the 30 days of support.
$620/mo (5% of the build price a month, includes 6 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 project tool's AI add-on bills per seat on top of your plan | A custom build shop quotes after workshops | Ours is one fixed price, shown before you buy |
| Scope questions | Searches tasks, not the statement of work | Depends on what was scoped | Ours cites the clause from the client's statement of work |
| Approvals | Sign-off sits in comments you scroll back through | Depends on the build agreed | Ours returns the cited record of what was approved |
| Delivery | Switched on fast, filled slowly | Usually several weeks | Ours has a fixed delivery time shown before you buy, with faster options |
Questions about this product
Can the Agencies hub tell an account manager whether a request is in scope?
It returns what your documents say: the relevant clause from the client's statement of work or retainer, with the citation, so the account manager can see the wording and decide. The judgment stays with your team. It does not raise change orders; for detecting scope creep and drafting change orders, see AI Scope Management.
How is the Agencies edition different from AI Agency Knowledge Base?
AI Agency Knowledge Base makes brand guidelines, campaign learnings and creative templates searchable. This edition is built around operational questions on client onboarding, reporting, approvals and scope tracking, drawing on your project tool, Google Workspace, CRM or ad and analytics platforms. Pick this one when coordination questions eat billable hours.
Can clients log in and ask the Agencies hub themselves?
No. It is an internal assistant for up to 50 users on one interface, such as Slack, Microsoft Teams or a web app. The intent library is tuned to how clients phrase onboarding and reporting questions, so your account team can paste a client's question in and answer it quickly with a citation.
Can the hub use data from our ad and analytics platforms?
Yes, if one of those platforms is among your 3 content sources and its data can be exported or reached by API. Indexed reports can then answer questions such as which metrics a client's reporting covers. The hub answers questions only; for building white-label performance reports, see AI Agency Reporting.
Can freelancers be kept away from retainer and pricing documents?
Yes. Up to 3 permission groups are set at intake, for example leadership, account management and delivery including freelancers. Retrieval enforces them, so someone outside the commercial group never receives a passage from a retainer or pricing document, whatever they ask. You approve the groups before go-live.
How soon can our account teams use the Agencies hub?
The hub goes live on a fixed delivery schedule from a verified intake, and the delivery panel shows that date plus express and rush options before you order. Intake items are due within 72 hours of ordering, and each day of delay moves delivery by 3 working days, so ready access to your project tool keeps the date.
What does an agency pay each month to run the hub?
Roughly $50 to $300 a month in your own accounts for AI model usage, the vector database and hosting, depending on how many documents are indexed and how often the team asks. The build itself is paid once, and monthly maintenance is optional.
Spend billable hours on clients, not on finding answers.
Order AI Company Knowledge Hub for Agencies and account and delivery teams get cited answers on onboarding, reporting, approvals and scope on a fixed delivery date.
Works well with
Agencies
AI Agency Reporting
Builds white-label campaign performance reports for agency clients; the agency version of AI Client Reporting.