zoho-billing-ai

Show Annual Recurring Revenue with QoQ and YoY trend and a growth trajectory. Use when the user asks about ARR, annual recurring revenue, annual run rate, or year-over-year revenue growth. Do NOT use for MRR, NRR, or monthly breakdowns.

Identify overdue invoices most likely heading toward write-off, while a save is still possible. Use when the user asks about bad debt, write-off risk, invoices unlikely to be recovered, or wants to see which AR is at risk of becoming bad debt. Do NOT use for general overdue lists or collections prioritisation.

List subscriptions whose payment card expires in the next 30 or 60 days, ranked by MRR at risk. Use when the user asks about expiring cards, cards about to expire, or wants to prevent involuntary churn from payment method expiry. Do NOT use for current payment failures or dunning analysis.

Show the full product catalog: all products, their plans with pricing and billing frequency, addons, and one-time items. Use when the user asks what plans are available, wants to see the product catalog, pricing, or available addons and items. Do NOT use for subscription status, customer-specific data, or coupon lookup.

Decompose churn into voluntary (customers who chose to leave) vs involuntary (failed payments) and identify which plans and products are driving losses. Use when the user asks why customers are churning, churn composition, voluntary vs involuntary split, or which products have the highest cancellation rate. Do NOT use for MRR, NRR, or retention rate questions.

Daily AR worklist: top 20 overdue invoices ranked by recovery probability (outstanding amount × payment-behaviour score). Use when the user asks who to chase for payments today, wants a collections prioritisation, or needs a daily AR action list. Do NOT use for aggregate overdue totals or bad-debt analysis.

List all subscriptions currently in PAST_DUE or UNPAID status, ranked by MRR at risk. Use when the user wants to see which subscriptions are in trouble, past-due subs, or subscriptions with failed renewals. Do NOT use for invoice-level overdue (use /overdue-invoices) or dunning analysis.

Identify customers underpriced for their actual usage — hitting addon limits, sustained high utilisation, or buying repeated one-time charges — and propose plan upgrades with supporting evidence. Use when the user asks about upsell opportunities, upgrade candidates, or revenue expansion from existing customers. Do NOT use for new customer acquisition or cross-sell of unrelated products.

List all overdue and unpaid invoices ranked by amount and days overdue. Use when the user wants to see what's unpaid, which invoices are overdue, total AR exposure, or a quick unpaid invoice list. Do NOT use for collections prioritisation (use /collections-today) or customer-specific invoice lookup.

Full invoice history for a customer: charges, payments applied, credit notes, and any gaps. Use when a customer disputes a charge, claims to have paid, asks about a specific invoice, or when the user needs to investigate a billing discrepancy. Do NOT use for aggregate overdue lists or collections worklists.

Break down MRR movement into new, expansion, reactivation, contraction, and churn components for any period. Use when the user asks about MRR growth, what drove MRR change, new vs expansion MRR, MRR waterfall, or why monthly recurring revenue went up or down. Do NOT use for ARR, NRR, total revenue, or churn rate questions.

Compute the MRR Quick Ratio (new + expansion + reactivation) / (contraction + churn) and classify business growth quality as efficient, leaky, or accelerating. Use when the user asks about growth quality, quick ratio, MRR efficiency, or whether growth is healthy. Do NOT use for raw MRR amounts or ARR questions.

Compute Net Revenue Retention and classify it against SaaS industry benchmarks. Use when the user asks about NRR, net revenue retention, whether existing revenue is growing or shrinking, or how well the business retains and expands customer revenue. Do NOT use for MRR growth waterfall, churn rate, or new customer acquisition questions.

Diagnose why a payment failed for a customer or subscription: look up the gateway error code, translate it to plain English, and recommend the fix. Use when the user asks why a payment failed, what a dunning error means, or how to resolve a failed charge. Do NOT use for churn analysis or dunning reports.

Rank all open quotes by staleness and expected close value, then recommend the next move per quote. Use when the user asks about open quotes, stalled deals, which quotes to follow up, or the sales pipeline status. Do NOT use for accepted/closed quotes or invoice creation.

Weekly refund anomaly digest: refund volume by product and country, spike detection, repeat-refund customers, and root-cause hypotheses. Use when the user asks about refunds, refund spikes, which products have high refund rates, or refund anomalies. Do NOT use for credit notes, payment failures, or churn analysis.

Upcoming renewal worklist: every subscription renewing in the next 14 days with a recommended motion per renewal (hold price, increase, pitch annual, or save-call). Use when the user asks about upcoming renewals, what's renewing soon, or renewal action planning. Do NOT use for past renewals or renewal failure recovery.

ASC 606 / IFRS 15 health check: deferred vs recognized revenue, waterfall variance, and anomaly flags for the controller or finance team. Use when the user asks about revenue recognition, deferred revenue, rev-rec health, or month-end close compliance. Do NOT use for MRR, ARR, or cash collection questions.

Weekly or monthly KPI digest covering MRR, ARR, active subscriptions, activations, churn rate, and NRR with anomaly callouts. Use when the user wants a top-level business health summary, weekly metrics, monthly report, or KPI dashboard. Do NOT use for detailed breakdowns of any single metric.

Full lifecycle view of one subscription: plan, billing history, renewal status, payment track record, and current health. Use when the user asks about a specific subscription, wants to know renewal date, billing amount, or full sub history. Do NOT use for a customer overview (use /customer-360) or subscription lists.

Rank all active trials by conversion likelihood and recommend the conversion nudge per trial. Use when the user asks how trials are converting, wants to prioritise trial follow-up, or needs to know which trials are most likely to convert. Do NOT use for expired trials or historical conversion rates (use /subscription-kpis).

List all active trials sorted by expiry date, showing days remaining and plan. Use when the user asks which trials are expiring, wants to see active trials, or needs to know which trial customers to engage before they expire. Do NOT use for trial conversion rates or historical trial analysis.

Run the full analyst pipeline end-to-end — sales performance, collection performance, and subscription growth — then synthesise all three into a single business-health report with an executive narrative and a rich visual dashboard. Use for any request to analyse "overall business health", "monthly business review", "how is the business doing", "full analysis", or any request that spans more than one of the three pillars. Propose-only; writes a Markdown report and renders a live dashboard; never calls mutating endpoints.

Analyse collection performance by comparing two equal-length date windows — this month day 1 to today, vs the same day range in the prior month (e.g., May 1–19 vs April 1–19). Pulls Zoho Billing payments received, refund details, payment failures, and renewal failures. Surfaces collection efficiency, refund pressure, failure rates, and MoM trends. Use when the user asks about cash collected, payment trends, refund rates, dunning health, or collections vs. invoiced comparison.

Analyse sales performance by comparing two equal-length date windows — this month day 1 to today, vs the same day range in the prior month (e.g., May 1–19 vs April 1–19). Pulls Zoho Billing invoice details, credit notes, sales by item, and sales by plan. Surfaces revenue trends, top-performing items/plans, credit-note pressure, and MoM delta. Use when the user asks about sales performance, revenue breakdown, invoice trends, best-selling plans or items, or credit note analysis.

Analyse subscription growth by comparing two equal-length date windows — this month day 1 to today, vs the same day range in the prior month (e.g., May 1–19 vs April 1–19). Pulls Zoho Billing active subscriptions, MRR, activations, and product/country breakdowns. Surfaces MRR trajectory, churn rate, net subscriber growth, and geographic/product mix. Use when the user asks about subscriber growth, MRR trends, churn, activations by product or market, geographic expansion, or any subscription-level performance question.

Takes a scored dunning cohort (from dunning-cohort-builder) and applies error-code playbooks to produce three outputs: (1) a prioritised action list per subscription, (2) an error-code segment summary table, and (3) playbook reference cards grouped by error code. Propose-only — all recommendations require human approval. Use after dunning-cohort-builder has run, or when the user asks for a recommended action plan for at-risk subscriptions.

Fetches every subscription currently in dunning from Zoho Billing, enriches each record with a composite priority score (weighted by amount at risk, LTV, lifetime duration, and retry count), then groups the cohort by gateway error code and error message. Output feeds directly into dunning-action-planner. Use when the user asks which subscriptions are in dunning, wants to see the current at-risk pipeline, or needs a scored and grouped cohort for downstream action planning.

Top-level entry point for the dunning agent. Runs the full dunning analysis pipeline end-to-end: fetches and scores the dunning cohort (dunning-cohort-builder), applies error-code playbooks to produce prioritised actions (dunning-action-planner), then synthesises everything into a single Markdown report with an executive summary, error- code segment table, prioritised action list, and playbook reference. Propose-only — never calls mutating endpoints. Use for any request that asks to "analyse dunning", "show dunning report", "what subscriptions are at risk", "prioritise dunning actions", or "dunning pipeline overview".

Flags subscriptions with payment cards expiring within 30 days to enable proactive payment method updates. Propose-only output for human review.

Identifies subscriptions currently in dunning status or at risk of entering dunning soon, segments by risk level for targeted recovery. Output feeds into decision matrix for recovery strategy recommendations.

Analyzes payment failures from a date range, categorizes by error type and gateway, identifies systemic issues. Propose-only analysis for human review.

Recommends targeted recovery actions for at-risk subscriptions based on failure reason and risk level. Uses decision matrix configuration to tune strategies. Produces propose-only Markdown report for human review.

Analyzes all open quotes, ranks by expected close value, identifies stalled/dead deals using staleness scoring. Recommends next moves (follow-up, discount, convert, mark lost) for human review and execution.

Build the recovery cohort by fetching the last 7 days of lost opportunities and abandoned carts from Zoho Billing. Normalizes both sources into a unified JSON array sorted by lost_value DESC — lost opportunities first (higher intent), then abandoned carts. Use this skill whenever you need to pull fresh data before a recovery run, or any time someone asks "what leads or carts did we lose this week?", "show me what we need to recover", or "build the recovery list".

Run the full recovery pipeline end-to-end — fetch last 7 days of lost opportunities and abandoned carts, rank by lost value, apply the decision matrix, and write a Markdown proposal report. Use for any recovery run request. Propose-only; no automated emails, no API mutations. ALWAYS use this skill when the user says things like "run recovery", "what leads did we lose", "recover failed signups", "abandoned cart recovery", "lost opportunities report", "who didn't sign up this week", "recovery run", "show me what we missed", or asks to analyse or act on revenue lost from incomplete signups or cart abandonment. Trigger even for vague phrasing like "let's do the recovery review" or "pull up recovery".

Apply the recovery decision matrix to a normalized recovery cohort and produce per-record action recommendations (human_outreach, email_with_discount, email_reminder, email_sequence, accept). Run after the cohort is built. Pure rule evaluation — deterministic, zero LLM tokens, re-runnable whenever the YAML changes.

Classify free-text cancel reasons into canonical buckets (price, feature_gap, temporary, quality, business_end, duplicate_account, unknown) using the YAML taxonomy. Run after cohort is built, before the strategy engine.

Build the retention cohort — at-risk (non-renewing) or recently churned subscriptions normalized with cancel reason, LTV, LTD, plan, and MRR. Use for any request to pull cohort data before a retention run.

Apply the retention decision matrix to a classified cohort and produce per-subscription recommendations (discount, pause, downgrade, human_triage, standard_offer, accept). Run after cancel reasons are classified.