# Walk into every customer call knowing what you promised. Customer call brief is an AI agent for customer meeting preparation. Name an account and it reads your customer, usage and ticket data, the last call's notes and your calendar, then sends you a short Slack brief: plan and renewal, open tickets, the usage trend, every promise and whether it happened, and two or three things to raise. ## Why do customer calls go wrong? Because the customer remembers what you said last time, and you don't. The account details sit in billing, the urgent ticket in the help desk, the usage drop in analytics and the promise in a call note from August. Without them all in front of you, a routine check-in turns into an apology. Without it: - Five tabs open a minute before the call. - A promise from the last call you forgot you made. - The customer telling you about their urgent ticket. - A usage drop you hear about at renewal. With Customer call brief: - One brief per account, in a few short lines. - Every promise marked delivered or outstanding. - Open tickets and the usage trend up front. - Two or three things worth raising. ## What goes into a customer meeting brief? Four things you'd otherwise look up in four places, then two or three things to raise. ### Where the account stands Plan, seats, monthly revenue and renewal date, with the days left and the CSM who owns it. For example: Plan, MRR, Renewal. ### What's broken for them Every open ticket with its severity, subject and age, so you know what they'll raise. For example: Severity, Status, Age. ### How much they use it Active seats from the first to the latest of the last eight weeks, against seats bought, as one line. For example: Active seats, Change. ### What you promised Requests and promises from the last call's notes, each marked delivered, outstanding or not evidenced. For example: Requests, Promises, Status. ## How does AI customer meeting preparation work? Every run follows the same five steps for one account, then sends one brief. Only the last step writes anything. 1. **Finds the account** (Tables query, reads): Plan, seats, MRR, renewal date and the CSM who owns it. 2. **Checks tickets and usage** (Tables query, reads): Open tickets, and active seats over the last eight weeks. 3. **Reads the call notes** (Document search, reads): What they asked for, what was promised, and any sign it was delivered. 4. **Looks for the next meeting** (Google Calendar, reads): The next event naming the account, from now on, or none found. 5. **Sends the brief** (Slack, posts): Six short sections, ending with two or three things to raise. When it runs: - @mention: @customer-call-brief I have a call with Customer B on Tuesday - Schedule: Every Monday at 8:00, for the account named in the sentence - Webhook: Run when your scheduling tool books a customer call ## From one sentence to a call brief. Describe the job, let Arclyx draft the agent, test it once on a real account, create it, then read the brief that lands in your Slack messages. Every promise says whether it was kept. Example output in your Slack DMs, "Customer call brief · Customer B" (placeholder companies): - Customer B · Account: Team plan, 25 seats, $1,250 MRR. Renews 5 December, in 42 days. Owned by Marcus. The renewal is close enough to come up on the call. Source: example-a.com/customers - Customer B · Open tickets: Two open: an urgent integration failure, 16 days old, and a question about paying for 25 seats. The urgent one blocks the reason they bought. Know its fix status first. Source: example-a.com/support-tickets - Customer B · Usage: 22 to 9 active seats of 25 bought over eight weeks (−59%). Expect a right-sizing question at renewal. Source: example-a.com/product-usage - Customer B · Last call: 86 days ago. Promised: a training session for their front desk in August. Not scheduled, per the notes. Outstanding and overdue. Two of their requests show no sign of delivery either. Source: example-a.com/call-notes - You · To raise: Confirm the integration fix. Ask what changed in adoption. Book the overdue training. Each one drawn from a ticket, the usage or the notes. Source: example-a.com/call-notes Next meeting: none found on your calendar · 2 flags ## One brief, before every call. The same brief answers a different question for each person who talks to the customer. - CSMs: Every check-in prepared, promises included. - Account managers: The renewal and the usage trend before the price talk. - Support leads: Which open tickets the customer will bring up. - Founders: A key customer's state before an exec call. ## How do you prepare for a customer meeting? Check four things before every call: where the account stands, what is broken for them, how their usage is trending and what you promised last time. Then pick two or three things to raise. An AI agent can gather all four from your own data in under a minute. In Arclyx that's five steps, starting with one sentence. 1. **Describe the job**: One sentence: what the brief covers, and where it goes. 2. **Upload your data**: Customers, weekly usage and tickets as CSV tables, and your call notes as a document. 3. **Connect Google Calendar**: The agent gets only tools that search and list events, so it can't create or change them. 4. **Watch a dry run**: Optional. Brief a real account and read it before anything is sent. 5. **Ask before a call**: @mention it with the account's name, or put it on a schedule. The sentence that built it (@mention, When you ask): > When I name a customer, brief me for the call from our customer, usage and ticket tables, the call notes and my calendar: plan, seats, MRR and renewal, open tickets, usage trend, what they asked for and what we promised, the next meeting, and two or three things to raise. ### Or start from one of these sentences - Renewal calls: "When I name a customer renewing this quarter, brief me with their seats, MRR, usage trend, open tickets and three things to settle before the renewal." - Shared with the team: "When I name a customer, post the call brief to #account-reviews so the CSM and the account manager both see it." - Escalations: "When I name a customer, list their open high and urgent tickets with ages, and what we told them about each in the notes." - A key account: "Every Monday at 8, brief me on Customer B before this week's check-in." ## AI meeting preparation vs your CRM record. Doing it by hand and reading the CRM record each give you part of the picture. Here is what each one gives you, and doesn't. | What you need | By hand | HubSpot record | Customer call brief | | --- | --- | --- | --- | | Account and renewal | Look it up in billing | Properties, if someone keeps them current | From your customer table | | Usage trend | Export from analytics | Only if usage is synced in | Eight weeks, first to latest, with the change | | Promises from the last call | Reread old notes | In logged notes and calls | Each one, delivered or outstanding | | Things to raise | If you plan them | No | Two or three, from the above | | What you get | Tabs and notes | An activity timeline | One short brief in Slack | Source: [HubSpot knowledge base: record timelines](https://knowledge.hubspot.com/records/filter-activities-on-a-record-timeline), checked 24 October 2026. ## Is it safe to give an AI agent your customer data? It reads your tables, notes and calendar, and writes only to you. Every run is recorded and has a spend ceiling by default. ### Every run is recorded - Every step, in order: Every run records the tools it called, with a short preview of each result. Only the person who started a run can open its content. - Every gap, flagged: Missing data, old notes and an empty calendar are listed in the brief, not papered over. - What each run cost: Tokens and cost are recorded per run and per agent, next to your plan's limits. - A spend ceiling, on by default: Each run has a dollar ceiling unless you remove it. A run stops before it goes over it, and tells you. ### It can - Read the tables and documents you can access in Arclyx - Search your Google Calendar - Message you on Slack, post to one channel you name, or email you ### It can't - Create, change or answer calendar events - Change your tables or notes - Contact the customer - Email anyone but you ## Customer meeting preparation, answered. ### What is customer meeting preparation? Getting ready for a call with an existing customer: where the account stands, what is broken for them, how much they use the product, and what you promised last time. Done well, the customer never has to remind you. The agent pulls all of it into one brief. ### How is it different from Meeting brief? Meeting brief researches people outside your company on the public web before external meetings. Customer call brief works from your own data on a customer you already have: their account, tickets, usage and the notes from your last call. ### Does it check whether promises were kept? Yes. For each request and promise in the last call's notes it looks for evidence in the tickets and notes, then marks it delivered, with the evidence, or outstanding. Without evidence it says 'not evidenced' rather than assuming. ### What data does it need? Three tables, uploaded to Arclyx as CSV: customers with plan, seats, MRR and renewal date; weekly active seats; and support tickets. Plus your call notes as a document, and Google Calendar for the next meeting. ### What if there is no meeting on my calendar? It says 'No meeting found on your calendar' in one line, and never lists unrelated events. It only searches from now on, and only for events that name the account. ### Does it change anything or contact the customer? No. It runs read-only queries, searches your notes and reads your calendar. It can't create or change an event, write to a table or reach the customer. Its one message goes to you. ### Can I brief several accounts at once? It briefs one account per run, so each brief stays short. Name several and it briefs the first and asks whether to carry on. ### Do I need to write code? No. Describe the job in one sentence. Arclyx drafts the agent, you upload your tables and notes and sign in to Google Calendar once, and you can watch a dry run before you create it. ## Agents that work well with it - [Renewal risk radar](https://arclyx.ai/agents/churn-prediction): Every renewal in the next 90 days, rated - [Success handoff](https://arclyx.ai/agents/customer-success-handoff): A handoff from every closed deal - [Meeting brief](https://arclyx.ai/agents/meeting-prep): A brief for every external meeting [AI for customer support and success](https://arclyx.ai/solutions/customer-support): AI customer service agents that work behind the scenes: the queue sorted, replies drafted for you, renewals at risk rated and every handoff written. ## Early access Build your customer call brief. One sentence in, a working agent out. Join the beta and we'll set you up. Join at https://arclyx.ai/#join or write to hello@arclyx.ai. Last updated 24 October 2026.