Revenue Intelligence

Data Dictionary

Every term on the site, what it means, and where it is.

Active 5 terms

Active company

A company where at least one person did real work in the last 7 days.

Where: ANALYTICS.WEEKLY_ACTIVE — IS_ACTIVE per account per fixed week (DATA-57, landed 2026-09-14) · on accounts, dailies, daily use, funnel
Active team (weekly active team)

A company where at least two people did real work in the last 7 days.

Where: ANALYTICS.WEEKLY_ACTIVE — IS_ACTIVE_TEAM per account per fixed week (persons_active at least 2; DATA-10, landed 2026-09-16) · on accounts, dailies
Daily use (week)

An onboarded account with at least one hands-on or with-Lana action on every weekday of a fixed Monday-Friday week.

Where: ANALYTICS.WEEKLY_ACTIVE.IS_EVERY_WEEKDAY; rendered on daily_use.html · on accounts, daily use
Daily habit (four weeks)

Daily use in each of the last four complete non-holiday weeks.

Where: Computed over ANALYTICS.WEEKLY_ACTIVE by load/export_daily_use.py; rendered on daily_use.html
Human-active account

An account with at least one event from a human actor in the last 30 days.

Where: ANALYTICS.HUMAN_ACTIVITY_EVENT · on product adoption

Paying 4 terms

Paying customer (product)

A company with an active paid platform subscription.

Where: proposed — ANALYTICS.DEF_PAYING_PRODUCT · on funnel
Customer (canonical)

A company being billed something in any revenue engine — an active Sage contract net of discounts, or a paid platform subscription. A fully discounted contract is not billing evidence.

Where: ANALYTICS.ACCOUNT_360 · on carr report, scorecard
Purchase date (product)

The day a company first moved onto a paid product tier (Core, Premium or Max), read from the nightly plan-tier snapshot.

Where: ANALYTICS.PRODUCT_PURCHASE (one row per company ever paid; PURCHASE_DATED_BY says how). Transitions in ANALYTICS.PRODUCT_TIER_TRANSITIONS. · on carr report
Non-billing account

A company with a live contract that is being billed nothing, because its discounts cover the full amount. A comp or a save, not a departure.

Where: ANALYTICS.CARR_MONTHLY.to_non_billing_net · on carr report

Churn 3 terms

Product churn

A paying product customer at the end of week N who is on Free or canceled at the end of week N+1. A tier downgrade is contraction, reported separately.

Where: ANALYTICS.PRODUCT_CHURN_WEEKLY · on carr report
Realized churn (gross revenue churn)

Recurring revenue that stopped this month — an account had recurring revenue last month and zero this month. Recognized when the contract ends, not when notice is given.

Where: ANALYTICS.CARR_MONTHLY.churn_net · on carr report, scorecard
Committed churn (forward)

Signed non-renewals not yet effective — revenue contractually ending in a future month.

Where: not measurable — see note · on carr report

Base 2 terms

Person

One human, not one login. Employee rows sharing a lower-cased email are one person across organizations; within one organization, the employee id.

Where: ANALYTICS.PERSON_KEY (employee_id to person_id) · on accounts, dailies
Company (the counted unit)

A canonical account that has completed onboarding. A company that has not finished the onboarding flow has not entered the application and is not counted as a company.

Where: ANALYTICS.COMPANY_BASE · on dailies, daily use, funnel

Usage 2 terms

Raw usage dollars

What the work cost before the plan-tier multiplier: the ledger's cost_attribution.rawCredits times 0.10 dollars. Multiplier-neutral, so it is the figure for every usage or behaviour chart.

Where: ANALYTICS.CREDIT_USAGE_DAILY.USAGE_DOLLARS_RAW; rendered only through report_style.usage_money(v, 'raw')
Visible usage dollars

What the customer was charged: the ledger deduction, already after the plan-tier multiplier, times 0.10 dollars. Only where the question is what customers were charged.

Where: ANALYTICS.CREDIT_USAGE_DAILY.USAGE_DOLLARS_VISIBLE; rendered only through report_style.usage_money(v, 'visible')
Confidential — Internal

Generated 2026-09-26 07:21 UTC from reports/definitions.yml. Definitions are authored in that file and nowhere else; this page is rendered from it, and status is derived from the review dates rather than stored, so it cannot drift from the truth. Changing a definition bumps its version, reopens a review window, and registers a restatement so published trend lines are corrected rather than silently re-based.