Automation · Customer service and scheduling

Answer prices and availability on WhatsApp from a Notion database

An AI agent answers prices, stock and terms on WhatsApp by reading your Notion database, and logs whatever it could not answer.

Show me this with my own operation See every automation

How do I make my WhatsApp answer prices with the list my team keeps in Notion?

A Met connected to Notion finds the database where your team keeps products or services, reads its columns and queries the row that matches the customer's question: price, availability, terms or delivery time. It answers with that value and, if the row has an attached catalog or sheet, it shares the download link. When something is not in the database it does not make it up: it logs the question as a new row in a pending questions database and leaves a comment for your team. If you change the price in Notion, the next answer already uses the new price.

What starts it

When the customer writes

It starts when a customer writes on WhatsApp to ask how much something costs, whether it is available or which terms apply. 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

Support Met with a price list in Notion

Sales team

This is not a named, seeded Met. It is the Met that already answers the business's WhatsApp, with Notion connected. Its work ends when the customer gets the price or the term the database states, or when the question is logged for someone on the team to answer.

Run by the customer support Met

Channels: WhatsApp

What you need connected

Notion Automate and collaborate

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
Find the price database

The Met searches Notion for the database your team named in the skill, for example "Price list", and keeps its ID. Search returns up to 50 results, so the database should have a name that is not repeated in the workspace.
- notion_search Notion
2
Read how the database is built

Before querying, the Met opens the database to learn its columns: product name, price, availability, terms. That tells it which column to filter on and which one to read, instead of assuming a field your team never created.
- notion_fetch Notion
3
Query the row the customer asks about

The Met queries the database filtering by the product or service the person named and reads that row's price, availability and terms. Each query returns up to 100 rows; if the customer asks about a whole category, the Met summarizes instead of pasting the table.
- notion_query_data_sources Notion
4
Share the attached sheet or catalog

If the row has an attached PDF or image, the Met asks Notion for the signed download link and shares it in the conversation. That link expires within minutes, so the Met requests it on the spot and never reuses an old one.
- notion_get_file_download_urls Notion
5
Log what the database does not answer

If the product is missing or the term does not appear, the Met does not improvise a price. It creates a row in the pending questions database with the question, the customer's phone number and the date, and tells the person someone on the team will reply.
- notion_create_pages Notion
6
Alert the team with a comment

The Met leaves a comment on the new row so the person in charge sees it in Notion. Once the team adds the product or the term to the database, the Met answers it next time without escalating.
- notion_create_comment Notion

Before you start

What you need ready

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

A price database in Notion

Your team keeps a database with one row per product or service and clear price, availability and terms columns. Whatever is not in that database, the Met does not know.
Notion connected with the writes you need

In Met, open Connections → Notion → Connect and authorize with an account that can see the price database. To log pending questions, the admin leaves page creation and comments switched on.
A WhatsApp line with the Met assigned

The business connects its WhatsApp Business API line and assigns the Met that answers customers. The answer happens inside that conversation; Notion remains the source of data.

The limits

What this automation does NOT solve

Better said here than discovered halfway through the rollout.

It does not charge or hold stock: it only answers what the database says. To sell, the Met needs another payments or store integration.
It does not fix the price database. If the row has an old price, the Met answers with that old price; keeping the database up to date is your team's job.
It does not ask for confirmation in the chat before creating the pending question row: if the action is on, it creates it. The admin decides which writes stay on.
It does not see databases the connected account cannot access. If the price list lives on someone else's private page, the Met does not find it.
It does not answer from a database of thousands of rows without a filter: each query brings up to 100 rows, so the question has to lead to a specific product or category.

FAQ

Questions about this automation

The next answer already uses the new price, because the Met queries the database on every question. There is no copy of the list stored in Meteor that goes stale.

Yes, if the database has a column for it. The Met filters by the columns that exist; if the difference is not written in the database, it cannot apply it.

It should not: this procedure leads it to log the question as pending and tell the customer someone on the team will reply. That is why the database should cover what people ask most.

Not to search, open, query a database and create rows. Unlimited SQL queries and some advanced searches need Business or Enterprise with Notion AI, and this procedure does not need them.

Keep going