Skip to content

Automation · Collections and receivables

Answer on WhatsApp how much a client owes according to FreshBooks

An AI agent answers on WhatsApp how much a client owes by reading their invoices, payments and credits in FreshBooks, without opening the books.

How do I answer on WhatsApp how much a client owes me without opening FreshBooks every time?

A client writes on WhatsApp to find out what they have pending. The Met picks the right business in FreshBooks, finds the client, lists their invoices and opens the ones still open, checks the recorded payments and the credits in their favor, and tells them which invoices are pending and for how much, with what FreshBooks says. If the client insists they already paid and the payment is not there, the Met does not mark it: it tells them the team will check. It is all reading; the FreshBooks connector writes nothing.

What starts it

When the customer writes

It starts when a client writes on WhatsApp to ask how much they owe, which invoice is pending or whether a payment arrived. There is no trigger to configure: the Met assigned to the line runs the procedure inside that conversation.

Nothing to set up. The Met assigned to that channel picks up the conversation and runs the procedure inside it.

Who runs it

Billing Met that answers balances with FreshBooks

Finance and operations team

It has no seeded name: it is a billing Met with FreshBooks connected. Its work ends when the client knows which invoices are pending, or when the difference they claim is in the hands of a person on the team.

Run by the quotes and receivables Met

Channels: WhatsApp

What you need connected

6 steps · 6 tools

The procedure, step by step

Each step shows the tool that runs. They belong to the real connector: if one of them stopped existing, this page would not build.

  1. 1

    Pick the right business

    The Met lists the businesses the FreshBooks account can access and keeps the one you named in the skill. If the account holds several businesses, querying the wrong one would give a balance that looks normal and belongs to another company.

  2. 2

    Find the client

    The Met looks for the person or their company among the FreshBooks clients. Before giving a balance, it compares what the person says with the client's details, so it never tells someone what another client owes.

  3. 3

    List their invoices

    The Met lists that client's invoices and separates the ones still pending from the ones already settled. If the client asks about a single one, it keeps that one.

  4. 4

    Open the pending invoices

    The Met opens each pending invoice to get its detail: what was billed, for how much and when. That lets it answer the client with concrete data instead of an unexplained total.

  5. 5

    Check the recorded payments

    The Met lists the payments recorded for that client. If the person says they already paid and the payment shows up, it confirms it; if it does not, it does not assume it and says the team will check.

  6. 6

    Take the credits in their favor into account

    The Met lists the credits the client has in their favor and factors them into the answer. Then it tells them which invoices are pending and for how much, with what FreshBooks says.

Before you start

What you need ready

The Met solves none of these for you. If one is missing, the automation stops there.

The limits

What this automation does NOT solve

Better said here than discovered halfway through the rollout.

FAQ

Questions about this automation

The Met checks the recorded payments. If the payment shows up, it confirms it; if not, it says the team will check. It does not mark it as paid, because the connector does not write to FreshBooks.

It should not: this procedure leads it to compare what the person says with the client's details before giving a balance. The skill should state which details it has to confirm.

It works to answer when the client asks. To reach out on its own it needs a scheduled task and, outside the 24 hours window, a template approved by Meta.

Keep going

Other automations built on the same tools

Want this automation running on your own data?

We will show it to you with your accounts connected, not with a canned demo.