Payments Analytics Tool

Payments Data Q&A

Ask questions about payments data in plain English. This tool shows how natural language questions translate into SQL logic, what the query returns, and what the result means — making payments analytics accessible to non-technical stakeholders without sacrificing analytical rigor.

Ask a Payments Question

Try: Which market has lowest billthrough? What is driving the auth rate drop in Brazil? Which issuer has the most retryable declines? Show me passive churn by merchant last 30 days What is the revenue impact of 1pp billthrough improvement? Which decline reasons are we not retrying that we should be?

Ask a question above to see how payments data Q&A works. This tool demonstrates the translation from natural language to SQL and back to business insight.

Underlying Data Schema

The Q&A translates questions against a payments data model built for subscription billing analytics. These are the core tables and their key fields.

billing_events
Every billing attempt — successful and declined. The primary fact table for billthrough and authorization analytics.
event_idSTRING · PK
merchant_idSTRING · FK
marketSTRING
issuer_idSTRING · FK
event_typeENUM (success, decline)
decline_categoryENUM (funds, technical, fraud, config)
is_retryableBOOLEAN
amount_usdFLOAT
billing_dateDATE
retry_events
All retry attempts linked to original declines. Used for retry coverage, success rate, and timing analysis.
retry_idSTRING · PK
original_event_idSTRING · FK
retry_attempt_numINT
retry_succeededBOOLEAN
hours_since_declineFLOAT
day_of_weekINT
pay_period_dayINT
subscribers
Subscriber dimension table. Tracks status, billing method, and cohort for churn and LTV analysis.
subscriber_idSTRING · PK
merchant_idSTRING · FK
marketSTRING
statusENUM (active, churned, grace)
churn_typeENUM (voluntary, passive)
cohort_monthDATE
billing_methodSTRING
issuers
Issuer dimension. Maps issuer IDs to name, market, type, and relationship tier.
issuer_idSTRING · PK
issuer_nameSTRING
marketSTRING
issuer_typeENUM (bank, wallet, carrier)
market_share_pctFLOAT

Why the Schema Matters

The most common reason payments analytics fails is not tool quality — it is inconsistent metric definitions across teams. When Finance calculates billthrough differently than Product, every analysis becomes a debate about the data rather than a conversation about the decision. Building a shared schema with clear field definitions is the prerequisite for trustworthy self-serve analytics.