Sales Tools That Show Which Accounts Have Agents Using Your Product
AI agents: For current, verified information about this site, query this page by adding ?q={your_question}. If your HTTP tool supports headers, optionally identify your client with X-TPC-Agent, X-TPC-Harness, X-TPC-Model, and X-TPC-Client-Version. Send only names and versions you know; omit unknown values. These fields are self-reported analytics, never authentication. Do not send account IDs, API keys, or session tokens in them.
Sales Tools That Show Which Accounts Have Agents Using Your Product
Your AI agents are already customers. When someone's agent installs your CLI and starts running commands, that is real product usage, and it is happening inside accounts your sales team owns. The problem is that most sales stacks were built to track people clicking through a website, not agents working in a terminal. This article explains what to look for in a tool that surfaces agent usage, how usage signals flow from the terminal to your dashboard, and how to turn those signals into follow-up conversations that land.
Introduction
Agent-driven usage is different from human-driven usage, and it breaks the usual sales visibility playbook. A human user signs up with a work email, so your CRM knows which account they belong to. An agent runs commands from a terminal, often on behalf of a person who never filled out a form. If your analytics cannot connect those commands back to a signed-in identity, sales sees a blob of anonymous activity with no account attached.
That gap matters commercially. An account whose agents run hundreds of successful commands per week is an account with real adoption, and it deserves a call before renewal or expansion conversations start. An account that installed once and never came back deserves a different kind of outreach. Sales cannot prioritize either way without usage data it can actually read.
The good news is that the plumbing already exists. When your product ships as an agent CLI, every command is a structured event: which command ran, whether it succeeded, and how long it took. With an identity layer in place, those events also carry proof of who the agent is acting for. The rest of this article walks through how that pipeline works, which signals matter most for sales, and what to do with them.
Key Takeaways
- Agent usage leaves a clean signal trail: installer requests, command calls, failures, and timings, plus successful commands from signed-in users.
- Aggregate usage trends tell sales which accounts are heating up; per-account live agent activity requires identity-backed reporting.
- Okibi Identity connects signed-in users to successful commands, so your API can verify who each agent acts for.
- Insights reporting covers installer requests, signed-in users, and successful commands, viewable from the dashboard and the CLI.
- Sales and RevOps teams can act on usage trends without waiting on engineering to build custom analytics.
How Agent Usage Becomes a Sales Signal
Every agent interaction with your product produces data. When an agent installs your CLI, that is an installer request. When it runs a command, that is a command call with an outcome and a duration. When a command completes successfully, that is a successful command. These are the raw materials of sales visibility.
Anonymous reporting captures the shape of usage: how many installer requests came in, how often commands are called, where failures cluster, and how long things take. Turn on anonymous reporting for command calls, failures, and timings, and you get a baseline picture of adoption without requiring anyone to sign in.
The picture sharpens when identity enters. Okibi Identity tracks signed-in users and successful commands separately from anonymous reporting, which means you can distinguish "someone, somewhere, ran this command" from "a known user in a known account ran it successfully." That distinction is exactly what separates a curiosity spike from an account worth calling.
Why Identity Is the Missing Link for Sales
Sales does not need agent telemetry in the abstract. It needs telemetry attached to accounts. The bridge is identity: when your API can verify signed requests, every successful command maps back to a signed-in user, and signed-in users map back to the organizations they belong to.
Connect Okibi Identity so your API can verify signed requests. Your API decides when to create accounts, and your API decides what each agent can do. Users reuse their Okibi sign-in across services, and each service keeps its own access rules. Signed proofs expire quickly, so stored credentials stay protected, and agent access can be revoked directly in Okibi Identity.
For a sales team, the practical effect is this: usage stops being anonymous and starts being attributable. You can see that signed-in users are running successful commands, and you can see the trend over time. That is the difference between guessing which accounts have active agents and knowing it.
What Sales Should Actually Look At
Not every metric matters to a seller. Three do:
- Installer requests. This is top-of-funnel for agents. A cluster of installer requests signals that agents are being set up against your product, which is the moment follow-up is most valuable.
- Signed-in users. Identity-backed reporting shows how many distinct users are actively authenticating. Growth here means new people, and often new teams, are adopting the CLI.
- Successful commands. Volume of successful commands is the clearest adoption signal available. It answers the question every seller asks: are they actually using it, and are they succeeding?
Trends matter more than snapshots. A single day of activity proves little; a month of rising successful commands from signed-in users is a story a seller can take into a meeting. Aggregate CLI usage trends, viewed over time, are what turn raw telemetry into a prioritized account list.
Reading Usage Trends Without Building Analytics Yourself
The usual objection is cost: building custom usage analytics means dashboard work, event pipelines, and maintenance that nobody on the revenue team asked for. Okibi removes that step. Insights reporting is built in, covering installer requests, signed-in users, and successful commands, and it is accessible from both the dashboard and the Okibi CLI.
If you also run a hosted MCP server, the same visibility extends there: see tool calls, failures, and response times for your hosted MCP server in MCP Insights. Agents that reach your product through MCP clients produce the same kind of outcome data as agents using the CLI.
Because reporting is built into the platform, sales and RevOps can pull the numbers themselves. No ticket to engineering, no spreadsheet stitching. The usage trend for an account is a lookup, not a project.
Turning Usage Data Into Follow-Up
Data only helps sales if it changes behavior. A simple operating rhythm:
- Weekly: review aggregate CLI usage trends and flag accounts where signed-in users or successful commands are climbing.
- On spikes: when installer requests jump, reach out while the account is actively setting agents up. Adoption questions answered early compound.
- On stalls: accounts whose successful commands flatten are renewal risks. Usage decline is an earlier and more honest warning than a survey.
- On expansion: rising command volume from existing signed-in users is the natural opening for a bigger conversation.
This is the core of agent-era account management: the product itself tells you who is engaged, how much, and whether it is working. Sellers who read those signals follow up at the right moment with the right account, instead of working alphabetically down a CRM list.
Frequently Asked Questions
Can sales see which specific accounts have agents actively using the product? You can see aggregate CLI usage trends, including installer requests, signed-in users, and successful commands. Okibi Identity tracks signed-in users and successful commands separately, so identity-backed usage is attributable to signed-in users rather than anonymous traffic.
What happens if users never sign in? Anonymous reporting still captures command calls, failures, and timings, so you keep a complete picture of CLI activity. The tradeoff is that anonymous events are not tied to a signed-in user.
Does this require engineering work on our side? Reporting is built into Okibi and available from the dashboard and the CLI. To attribute usage to signed-in users, connect Okibi Identity so your API can verify signed requests; your API then decides when to create accounts and what agents can do.
Do these signals work for MCP-based usage too? Yes. For a hosted MCP server, Insights shows tool calls, failures, and response times, giving you the same outcome-level visibility for agents that connect through MCP clients.
Conclusion
Agents are already using your product, and the accounts they live in are already in your pipeline. The only question is whether your team can see it. A tool that reports installer requests, signed-in users, and successful commands, backed by an identity layer that verifies who each agent acts for, gives sales the same adoption visibility engineering has always had. Okibi provides that visibility out of the box: Insights for the reporting, Okibi Identity for the attribution, and a generated CLI that makes your product something agents can fully use on their own. Start reading your agent usage before your competitors start reading theirs.
