Verticals · SaaS
Your whole company runs on a handful of metrics — ARR, MRR, churn, net revenue retention, active users, CAC and LTV — and finance, product, and GTM each define them a little differently. "What counts as churn?" and "what is an active user?" aren't schema questions; they're arguments your teams have never fully settled. A text-to-SQL tool doesn't settle them — it writes fresh SQL and quietly picks one interpretation each run, so the board deck, the product dashboard, and the investor update disagree and no one can say which is correct.
The alternative today
Every option in front of a SaaS data team either re-guesses the metric each run or hard-codes a different version of it per team — and none of them make ARR mean one thing across finance, product, and GTM.
Text-to-SQL & LLM copilots
The tool writes SQL from the schema and infers what "churn" or "active user" means on each run. Ask the same question next week and it can pick a different definition — so the number moves for reasons that have nothing to do with the business.
A dashboard per team
Finance, product, and GTM each build their own ARR and retention logic into their own dashboards. Each is internally consistent and mutually contradictory — and the gap only surfaces when someone puts two of them on the same slide.
The cost of waiting
The week before the board meeting becomes a fire drill to agree on one NRR number, not to discuss what it means. When investors ask how a metric is defined, the honest answer is "it depends who you ask" — and that costs credibility.
The outcome
Instead of regenerating SQL and re-guessing your terms, Spotonix — the AI analyst that interprets before it queries — surfaces the ambiguity in a metric, lets your teams settle on one definition, and then composes every answer from that definition. Churn is defined once and referenced everywhere, so the board deck, the product dashboard, and the investor update are built from the same logic and actually tie out.
The who → Segments
Active users, paying accounts, and expansion cohorts are defined once as Segments in the Context Graph — referenced everywhere instead of re-derived a different way on each team's dashboard.
The how much → Calculations
ARR, gross and net churn, NRR, and CAC/LTV live as Calculations that carry the logic your finance, product, and GTM leads agreed — the same definition behind every answer that uses them.
The reusable intent → Analysis Patterns
A cohort retention analysis or an expansion-vs-contraction cut becomes an Analysis Pattern — the semantic glue binding Segments and Calculations into a question the business can re-run every board cycle and get the same shape.
Governed from intent through execution
How it works
A plain-English metric question stops being a guess. Intent Algebra resolves it into a visible plan built from your own definitions, asks when a term is contested, and only compiles when the plan closes.
Someone asks "what was net revenue retention for the enterprise cohort last quarter?" No SQL, no special syntax — the question a board deck actually needs answered.
Intent Algebra binds the terms to the Segments, Calculations, and Analysis Patterns your teams approved, then shows the plan — which definition of the cohort, which retention formula, which window. If "the enterprise cohort" is still contested, it asks rather than silently guessing, and you agree the definition once.
When the plan closes it compiles to a query, and the generated SQL is validated against the bindings you approved before it executes. The approved definition is persisted in the Context Graph and retrieved on the next question that mentions NRR — so the company stops re-teaching the same business language and the numbers reconcile across teams.
Representative questions
Contested definitions
"Finance, product, and GTM each report churn differently. Which definition do we actually use?"
Spotonix surfaces the competing definitions, shows the difference, and lets the business settle on one Calculation that then drives every answer that mentions churn.
Board reconciliation
"Why does ARR on the board deck not match the number in our product dashboard?"
Because they were built from two definitions. Once ARR is one Calculation referenced everywhere, both are built from the same logic — and the reconciliation fire drill goes away.
Self-service under governance
"Can GTM pull their own retention cuts without inventing a fourth definition of active user?"
They ask in plain English; Intent Algebra composes only from the Segments and Calculations your teams approved, so self-service can't quietly fork the metric.
Consistency
"If two people ask for NRR next quarter, will they get the same logic?"
Once a definition is approved, it's persisted and retrieved for the next NRR question, and the generated SQL is validated against it before running — so the interpretation is on the record, not re-guessed between runs or between people.
Proof & status
The mechanism behind every SaaS metric answer: how a question is resolved into a visible plan built from the definitions you approved, then compiled — and how the ambiguity gate asks instead of guessing when a metric is contested.
Ask a metric question and see the plan before any SQL runs — and watch the ambiguity gate surface a contested definition instead of guessing. That behavior is live today.
We're working with early design partners to validate interpret-before-execution on SaaS metrics. We're not publishing customer results or naming companies yet — if reconciling ARR, churn, and NRR across teams is the problem in front of you, talk to us about the design-partner program.