Get tasks doneBusiness size: Small team

A Philippines automation freelancer says his n8n and Claude flows answer about 70% of a Malaysian brand's support questions and moved 5,437 reviews for $1

Aaron Cabantugan · Philippines · n8n, Claude, Apify, Shopify

What the agent did

For a Malaysian direct-to-consumer brand he replaced a chat tool with five self-hosted n8n flows across WhatsApp, Messenger, Instagram and Facebook. Rules route each question and Claude handles the unclear ones. An Apify and n8n pipeline moved customer reviews into the Shopify store.

Result, as claimed

He says about 70% of the brand's support questions are now handled without a person, five flows are live across the four channels, and 5,437 reviews moved to Shopify for $1 in API costs against $1,500 or more for a migration service.

This is the result as claimed in the public source linked below. It has not been checked and is not a typical outcome.

Proof

Read the public source

Try it yourself

Sort a month of customer questions into ones a bot can answer from your own facts and ones a person must take.

Difficulty: Intermediate

Prompt

Here are [number] recent customer messages from my shop, with personal details removed: [paste]. Here are my real shop facts, such as shipping times, prices, returns and product details: [paste]. Group the messages by intent. For each group, say whether it can be answered only from my facts, and draft one reply that uses only those facts. Put anything about refunds, complaints, health or legal topics, or anything my facts do not cover, in a hand to a person list. If a fact is missing, write I don't know and say what I should add. Return a draft only. Never send, spend, change settings or write to a live system.

Prompt: starter by Solopreneur Code

Tools

  • n8n Used to run the five support flows and the review migration pipeline.
  • Claude Used for the questions the routing rules could not sort.
  • Apify Used with n8n to move the reviews into Shopify.

Steps

  1. Export one month of customer messages and remove names and numbers.
  2. Write a short page of shop facts the bot is allowed to use.
  3. Run the prompt and check that every draft reply uses only those facts.
  4. Start with one channel and the two or three most common question types.
  5. Read the handed over messages each week and add missing facts.

Watch out: A bot that guesses can promise refunds or delivery dates you cannot keep, so keep a clear I don't know rule and a fast way to reach a person.

Starter version by Solopreneur Code

Sign up for the newsletter

Get the next story from Solopreneur Code in your inbox.

Form not loading? Subscribe on Substack

Back to all stories

More like this

Browse all stories

Once a week it reads his spam and archived email, groups it by sender and looks for events and articles that fit his goals for the year. For each useful sender it gives up to 5 action points and suggests pulling it out of spam or unsubscribing. It now checks his calendar and working hours before it shortlists events. Claude also names new files and chats to one convention and drafts unusual workshop formats for him to pick from.

Result, as claimed

The first run went through 718 emails and it now handles about 30 a week. He says AI gives him back 3 to 4 hours a week, and one webinar it flagged put him on a path to new work.

Get tasks doneBusiness size: SoloIncludes prompt

A marketing educator has Claude route her inbox into CRM deals, tasks and draft replies, and says agents save her 3 to 4 hours a week

Cristina Peñaloza Morales · Country unknown · Claude, Gmail, HubSpot, Todoist, Notion, Riverside, Gamma

Reads new email and routes each one. Training enquiries become a HubSpot contact and deal with a drafted reply. Community messages get a draft reply. Possible collaborations become a Todoist task. Invoices and admin get a label. Also runs a scheduled Monday AI news digest.

Result, as claimed

She says agents save her between 3 and 4 hours a week.

Get tasks doneBusiness size: Small teamIncludes prompt

A marketplace CEO says Claude cut his company's monthly close from 2 weeks to 2 days and builds a 70-slide board deck with one click

Brendan Candon @brendancandon · Country unknown · Claude, Claude artifacts, Claude Tag (Slack), Jira, Amplitude, Metabase, SigNoz, GitHub, Iterable, Google Drive, Excel, PowerPoint

Claude, connected to the company's systems, pulls data for board meeting prep, the monthly close and weekly leadership dashboards and presents it. Custom apps built with Claude replaced Monday, ClickUp and Airtable.

Result, as claimed

For the last two quarters Claude has built a 70-slide board deck with one click, and the monthly close went from 2 weeks to 2 days.