Founder and CEO
The whole company, in one question.
ARR, net revenue retention, burn and pipeline against plan, in the same definitions the board sees. Ask the follow-up yourself.
ARR, net revenue retention, churn, pipeline and CAC, defined once with your finance team and answered from billing, the CRM, the support desk and your warehouse together. Everyone asks in the AI they already use, within their own access. Then the board pack and the Monday report send themselves, and the follow-up gets done in the CRM.
ARR in the board deck, ARR in billing, ARR in the CRM. Three numbers, and a meeting every month spent reconciling them.
“Why did NRR dip?” goes to the data team’s queue, behind the last ten tickets. The dashboard shows that it dipped, not why.
Customer success wants the accounts at risk this quarter. By the time the list is pulled, two of them have churned.
ARR means what finance says it means, whoever is asking. Each team asks about what it runs, in its own words, and gets the number the board would see.
Founder and CEO
ARR, net revenue retention, burn and pipeline against plan, in the same definitions the board sees. Ask the follow-up yourself.
Finance
From billing, every time, with the same rules for upgrades, downgrades and credits. The board pack built from those numbers, on schedule.
Sales and RevOps
Win rates, what closed this month, which deals went quiet, from the CRM. Follow-ups logged where the rep already works.
Customer success
Usage down, tickets up, renewal inside 90 days. A list every Monday, with the check-in emails drafted.
Marketing and growth
Attribution and CAC from your own definitions. New leads researched and scored overnight, written into the CRM before the first call.
Product
Feature usage by plan and by account, from your warehouse, next to the revenue it carries.
One question can read usage from the warehouse, ARR from billing and owners from the CRM together. And the answer isn’t the end: the task gets logged, the email gets drafted, the list lands in Slack, with the same permissions and every step on record.
Which accounts over $50K ARR have usage down 30% this quarter and renew in the next 90 days?
Six accounts, $410K ARR between them. Four opened a support ticket in the last month. Owners from HubSpot: Maria has four, Dev has two.
warehouse · usagebilling · ARR
HubSpot · owners
Log a renewal-risk task on each one and draft the check-in emails.
Done. Six tasks in HubSpot, each assigned to the owner. Six drafts are waiting in your mailbox.
HubSpot · writeemail · draft
Asked by Ana · Customer success access · on record · Example numbers
What ARR, churn and NRR mean, including upgrades, downgrades, credits and multi-year deals, is written down with the people who own the numbers. AI drafts the definitions from your data; your team approves them.
Every question is checked against the approved definitions and tables before it runs. The AI doesn’t guess what a column means or which invoices count.
The questions you approve are asked again every week and scored against billing and the warehouse. If a system or a model change breaks a number, the Accuracy report flags it before the board pack goes out.
Accuracy report
Week 45
23 of 24 approved answers match
Net revenue retention for October changed after last week’s billing update. Flagged for review before the board pack goes out.
Checks numbers, names and lists for closed periods, like last month or last quarter. Example data.
Each team gets its own workspace and its own access, down to the table. Nobody sees payroll by asking nicely.
The AI reads your CRM and billing without being able to change them, until you allow writes, tool by tool. Every change is on record.
Works in Claude, ChatGPT or whatever your team uses next. Switch, and your connections, definitions and access rules stay put.
What changes
Keep your dashboards, your CRM and your billing. Add the one number, and the follow-up, on top.
Start with the number your board asks about every month and the report someone rebuilds by hand. Then open it up team by team, on the same definitions, and let it reach the systems each team works in.
No. If your data already sits in one, we start there. If it is scattered across billing, the CRM and the app database, we bring it together first, in your accounts, and define the numbers on top of it.
Keep it. A dashboard answers the first question, and the next ten go to the data team’s queue. FrontierOps answers the follow-up, on the same definitions, in minutes, and runs the report and the alert that come after.
Good. Your definitions come straight from the project, from dbt Core, with one dbt parse step in CI. There is a page for that. For dbt teams →
Only where you say so. It reads a system unless you give it write access, tool by tool, and every change is on record.
Finance signs off the definition once, and every answer uses it. Then every week the approved numbers are asked again and scored against billing and the warehouse, so a change in a system shows up before it reaches the board.
Whichever your people use: Claude, ChatGPT or whatever comes next. If you switch, your connections, definitions and access rules stay in FrontierOps and nothing breaks.
Then we prove it on your own data in a week.