FINANCE

Agentic Dunning Recovery Orchestrator

On each failed Stripe payment, an AI agent reasons over the customer's history and warehouse signals to choose a personalized recovery plan, drafts tailored outreach.

CategoryFinance
Enginepaperclip
Difficultyadvanced
Triggerwebhook
Steps5
Setup~25 min

How it runs

The automated pipeline, trigger to output.

  • TriggerStripe invoice.payment_failed webhookStripeStripe
  • ActionPull customer history and signals from BigQueryGoogle BigQueryBigQuery
  • LogicAgent reasons over context to choose recovery plan
  • ActionReschedule retry in Stripe and send tailored emailOutlook
  • OutputEscalate low-confidence or high-value cases to SlackSlack

What it does

Replaces fixed dunning rules with an agent that decides per customer. For each failed payment it weighs the decline reason, account value, prior recovery behavior, and engagement signals from the warehouse, then composes a recovery plan: how aggressively to retry, what to say, and whether a human should step in.

When to use it

Use it when rigid cadences leave money on the table and your accounts are heterogeneous enough that one-size outreach underperforms. Best for teams comfortable letting an agent draft and execute outreach with guardrails.

How it works

  1. 1A Stripe `invoice.payment_failed` webhook starts the run.
  2. 2The agent pulls customer history and engagement signals from BigQuery to assess recovery likelihood.
  3. 3It reasons over the context to pick a retry timing and drafts a personalized email matching the situation.
  4. 4It executes the retry reschedule in Stripe and sends the drafted message via Outlook.
  5. 5Low-confidence or high-value cases are escalated to a human in Slack with the agent's reasoning attached.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect StripeCustomers, subscriptions, payments.
  2. 2
    Connect BigQueryDatasets, queries, schemas.
  3. 3
    Connect OutlookMail, calendar, contacts.
  4. 4
    Connect SlackChannels, DMs, threads, mentions.
  5. 5
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  6. 6
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  7. 7
    Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.

Run this workflow in your colony.

14-day trial. No DevOps. No Sales call. Provisioned in under a minute.