Skip to content
Civware

Use cases

Use cases,team by team

Common first projects in each team: what the AI does, the systems it works in, and how it's measured. Each one starts the same way, with one job and one measure.

Support

Support

Replies drafted from your help center

Reads each new ticket and the customer's recent orders, drafts a reply from your help center articles with the source linked, and leaves it for an agent to edit and send.

Works in
Help desks such as Zendesk, Freshdesk or Intercom
Measured by
Time to first reply, and how many drafts go out with few or no edits
AI Assistants
Support

Tickets tagged and routed on arrival

Reads the subject, message and attachments, sets the category, product and priority, and assigns the right queue. Tickets it isn't sure about go to a person, with its best guess shown.

Works in
Help desks such as Zendesk, Freshdesk or Jira Service Management
Measured by
Tickets reassigned after routing, and time until the right team picks them up
AI Agents
Support

Customer questions answered on your site

Answers common questions in your website chat using only your published help content, links the page it used, and hands over to a person with the full conversation when it can't help.

Works in
Your website chat, connected to your help center
Measured by
Questions resolved without a ticket, and answers rated correct in weekly samples
AI Assistants

Operations

Operations

Incoming requests sorted and routed to the right team

Reads requests arriving by email and web form, picks the request type, urgency and owner, fills in the fields and creates the task. Anything unclear goes to a person to sort.

Works in
Shared inboxes and work tools such as Jira, ServiceNow or Asana
Measured by
Requests sent to the wrong team, and time until the right owner picks them up
AI Agents
Operations

Orders checked against stock and delivery dates

Checks each new order against stock levels, supplier lead times and the promised delivery date, and flags conflicts before they reach the warehouse. A planner decides what changes.

Works in
Order and inventory systems such as NetSuite, SAP Business One or Shopify
Measured by
Exceptions found after picking starts, and orders that miss their promised date
AI Agents
Operations

Procedures answered from your manuals

Answers how-to questions from your operating procedures and equipment manuals, shows the page and section it used, and says so when the manuals don't cover it.

Works in
Slack or Microsoft Teams, reading documents in SharePoint, Google Drive or Confluence
Measured by
Questions passed to senior staff, and answers checked as correct in weekly samples
AI Assistants

Finance

Finance

Invoices read and matched to purchase orders

Reads PDF and scanned invoices from your accounts payable inbox, pulls the supplier, line items, totals and tax, and matches each one to its purchase order. Mismatches go to a person with the difference highlighted.

Works in
Accounting systems such as NetSuite, Xero or QuickBooks
Measured by
Invoices keyed in by hand, and mismatches caught before payment
Document Processing
Finance

Expense claims checked against policy

Reads each receipt, checks the amount, date, category and limits against your expense policy, and returns claims that break a rule with the reason. A manager still approves every claim.

Works in
Expense tools such as Expensify, Ramp or SAP Concur
Measured by
Claims returned more than once, and policy breaks found after payment
Document Processing
Finance

Month-end reconciliations prepared for review

Matches bank and card transactions to ledger entries, groups what doesn't match, and drafts an explanation for each difference. Your accountant reviews and posts every adjustment.

Works in
Your ledger, bank feeds and reconciliation spreadsheets
Measured by
Hours spent reconciling at month end, and differences left unexplained
AI Agents

Sales

Sales

Account briefs drafted from your CRM before each call

The day before a meeting, gathers the account's open deals, recent emails, support tickets and notes into a one-page brief. Every fact links to the record it came from.

Works in
CRMs such as Salesforce or HubSpot, plus email and calendar
Measured by
Prep time per meeting, and facts in briefs that turn out wrong
AI Agents
Sales

First drafts of proposals from approved content

Drafts proposals and security questionnaire answers using only your approved product descriptions, pricing rules and past answers, and marks anything it couldn't source for the rep to fill in.

Works in
Your content library in Google Drive or SharePoint
Measured by
Time to first draft, and wording changed in legal or pricing review
AI Assistants
Sales

CRM notes written up after each call

Turns the call recording or the rep's rough notes into a summary, next steps and field updates, then waits for the rep to confirm before anything is saved.

Works in
CRMs such as Salesforce or HubSpot, and your call recording tool
Measured by
Calls logged within a day, and edits reps make before saving
AI Agents

HR

HR

Policy questions answered from your own handbook

Answers staff questions about leave, benefits and expenses using only your handbook and policies, quotes the section it used, and says it doesn't know when the documents are silent.

Works in
Slack or Microsoft Teams, reading your handbook in SharePoint or Google Drive
Measured by
Repeat questions reaching HR, and answers checked as correct in weekly samples
AI Assistants
HR

Onboarding checklists tracked and chased

Tracks each new hire's steps from signed offer to first week: contract, laptop, accounts, training. It reminds whoever owns the next step and tells HR what is running late.

Works in
HR systems such as BambooHR, Rippling or Workday, plus email
Measured by
Steps missed or late by the first day, and reminders HR sends by hand
AI Agents
HR

Job descriptions drafted from your templates

Drafts job descriptions from your role templates, salary bands and the hiring manager's notes, following your inclusive language guidelines. HR edits and approves before anything is posted.

Works in
Your templates in Google Docs or Word, and your applicant tracking system
Measured by
Time from request to approved draft, and rounds of edits before posting
AI Assistants

IT

IT

Access requests checked and prepared for approval

Checks each access request against the person's role and your access policy, prepares the change in your identity system, and waits. Nothing changes until IT approves.

Works in
Identity systems such as Okta or Microsoft Entra ID, with requests from your service desk
Measured by
Time from request to access, and changes made without approval (the target is none)
AI Agents
IT

Internal IT questions answered from your knowledge base

Answers questions such as VPN setup, new device setup and printer problems from your IT knowledge base, links the article, and opens a ticket with the details filled in when it can't solve the problem.

Works in
Slack or Microsoft Teams, connected to your service desk
Measured by
Repeat tickets on covered topics, and answers checked as correct in weekly samples
AI Assistants
IT

Incident summaries drafted from logs and tickets

After an incident, gathers the alerts, ticket updates, chat messages and recent changes from that window and drafts a timeline and summary. The engineer on call checks it and adds the cause.

Works in
Monitoring and paging tools such as Datadog or PagerDuty, plus your tickets and chat
Measured by
Time to a finished write-up, and corrections made in review
AI Agents

Tell us one job you'd like AI to do

We'll tell you honestly whether AI is the right fit, and what a sensible first step looks like.

One job is enough to start.

We use your details to reply to you, as described in our privacy policy.