n8n Shopify Automation: 10 Workflows Worth Building
Ten workflows that connect Shopify, email, ads and AI in one place, the math on the hours they give back, and the setup trap most n8n tutorials still skip.
n8n Shopify automation means using n8n to connect your store to email, ads, spreadsheets and AI inside one workflow. Use Shopify Flow when a job stays inside Shopify. Use n8n when it crosses tools. The ten workflows below are the ones I'd build, because each replaces a boring job you repeat every week. On a hypothetical store, together they give back about 12 hours a week after upkeep.
Now the hot take. Most lists of n8n ecommerce ideas start with abandoned cart emails and order confirmations.
Don't build those. Klaviyo already runs abandoned cart flows better than your homemade version ever will, and Shopify already sends the confirmation. Rebuilding them gets you a more fragile copy of something you already own.
I'm not guessing about the tool, either. We run n8n in our own business. The lead forms on our site post into an n8n webhook, and our product importer hands its batches to n8n too. So this list comes from someone who's paying the upkeep bill.
What can n8n do that Shopify Flow can't?
Less than the vendors selling n8n hosting will tell you.
Several of those guides say Flow can't call outside tools. Wrong. Flow has a Send HTTP request action on Grow, Advanced and Plus. It waits up to 30 seconds for a response and can retry for up to 24 hours.
So the honest line is about weight. Flow is great at "this happened in my store, so do that." It gets clumsy when you want to pull from four tools, have an AI model read the pile, wait for your approval, then write the answer to a fifth.
That's the gap n8n fills.
| Shopify Flow | n8n | Zapier | |
|---|---|---|---|
| Starting cost | Free app on Basic, Grow, Advanced and Plus | Free Community Edition if you self-host, or cloud from €20 a month on annual billing | Free for 100 tasks, Pro from $19.99 a month on annual billing for 750 |
| How usage counts | Limits vary by plan | One execution per full run, however many steps | Successful steps use tasks, triggers don't |
| AI with human approval | Not covered on the Flow page I checked | Approval on agent tools through Slack, Gmail and 7 other channels | AI steps use tasks based on model tier |
| Who fixes it when it breaks | Shopify, mostly | You | Zapier for the platform, you for the logic |
| My pick for | Jobs that stay inside your store | Jobs that cross 3 or more tools | Small jobs with low volume |
Flow is the light switch. n8n is the wiring. Don't rewire the house to turn on a lamp.
If you want the wider map of where this fits, start with the AI commerce guide. And before you build anything, check whether Shopify Sidekick can already handle the admin job for free.
Which 10 Shopify automation workflows are worth building?
Three of these use AI. That's on purpose. The n8n AI workflows worth your time are the ones where a model reads messy text. Everything else is plain plumbing, and plumbing doesn't need a model.
1. The morning numbers report
A schedule trigger pulls yesterday's orders with the Shopify node, adds them up and posts revenue, order count and average order value to Slack. Start here. It's the easiest build on this list, and you'll learn the tool on something that can't hurt a customer.
2. The big order alert
Fire on every new order. If the total clears a line you pick, say $400, ping Slack with the customer's name and what they bought. A handwritten note on a $400 order costs you 2 minutes.
3. The low stock warning
Shopify offers an inventory_levels/update webhook, but I'd run this one hourly instead. You'll spend fewer executions, and one Slack message beats 40.
4. Review triage with AI
When your review app sends a new review, a model tags the topic and the objection. One-star and two-star reviews go to support. Every objection lands in a sheet, and that sheet is worth more than the alert, because it's the exact wording your product page should be answering.
5. Support drafts with AI
The Gmail trigger catches a new message. An AI agent looks up the order and writes the reply using Gmail's Create Draft operation. A human presses send. Any refund tool sits behind n8n's approval step, so the agent can ask but can't spend.
6. The Klaviyo product event
This is the one Flow users ask for. When someone buys your hero product, send a custom event through Klaviyo's Create Event endpoint using the HTTP Request node. Those events can trigger flows by default. Now your Klaviyo flows can follow up based on what they bought, with logic you control. Pair it with a Bought A, never B segment and the follow-up only reaches people who still need the next product.
7. Ad spend against revenue
Meta exposes spend by ad account, and n8n has a Facebook Graph API node to call it. Pull yesterday's spend, pull yesterday's orders and append one row to Google Sheets. After 30 days you've got your return on ad spend in a sheet you own, instead of the number Meta grades itself with.
8. The catalog check with AI
Once a week, pull every product and flag the gaps: no image alt text, an empty description, a missing barcode. Then have a model draft the fixes into a sheet for you to approve. Clean data pays twice if you're feeding Google Shopping. Never let it publish on its own.
9. The refund reason digest
Every Monday, pull last week's refunds through the HTTP Request node (the Shopify node only covers orders and products) and group the notes by product. One sheet. One glance. You'll spot the bad batch before it costs you a month of returns.
10. The product leaderboard
This is my favorite, and it's the one nobody puts on their list. Pull sessions per product page with the Google Analytics node, pull orders per product from Shopify, and rank every product by revenue per visitor. Pair it with profit per product and you'll know exactly which page deserves your next hour.
How much time do these n8n workflows give back?
Here's the math on a hypothetical store doing 30 orders a day. Swap in your own counts, because these are my estimates and not a study.
| Workflow | Manual math (hypothetical) | Minutes a week |
|---|---|---|
| Morning numbers | 20 minutes x 7 days | 140 |
| Big order alert | 5 minutes x 7 days | 35 |
| Low stock | 15 minutes x 5 days | 75 |
| Review triage | 3 minutes x 40 reviews | 120 |
| Support drafts | 2 minutes saved x 60 tickets | 120 |
| Klaviyo event | 30 minutes x 2 list pulls | 60 |
| Ad spend sheet | 15 minutes x 7 days | 105 |
| Catalog check | 60 minutes x 1 | 60 |
| Refund digest | 45 minutes x 1 | 45 |
| Product leaderboard | 80 minutes x 1 | 80 |
| Total | 840 |
That's 14 hours. Take off 2 hours a week for fixing what breaks, and you've got 12.
Budget 2 hours a week for upkeep. The tutorials never do, and it's where most of these projects die.
What does n8n Shopify automation cost, and what breaks first?
Here's the math on executions. Each full run counts as one, however many steps it has. On that same hypothetical store, 2 order-triggered workflows use 1,800 a month (30 orders x 30 days x 2). The hourly stock check adds 720. Add the daily, weekly and message-triggered workflows and you land near 3,000.
That's over the 2,500 on n8n's €20 Starter plan and comfortably inside the 10,000 on the €50 Pro plan, both on annual billing. Or self-host the Community Edition for free, and accept that the server is now your problem.
Now the trap, and I barely see anyone covering it.
n8n's Shopify credential docs tell you to create a custom app in Shopify and copy its access token. But Shopify's own page now says custom apps are created in the Dev Dashboard, and only apps made before January 1, 2026 count as legacy apps you manage from the admin. If you're setting up today, you may use the client credentials grant, and those tokens expire after 24 hours. It only works when the app and store belong to the same Shopify organization.
So plan for it. Have the workflow request a fresh token before it calls Shopify, or it'll work on Tuesday and die on Wednesday.
Three more things bite people:
- Missing scopes. n8n's docs warn that a missing scope can throw a "Forbidden" error, and
read_ordersalso needsread_products. - Duplicate deliveries. Shopify sends an
X-Shopify-Webhook-Idheader so you can spot repeat deliveries. Use it, or you'll send two Klaviyo events for one order. - Silent failures. Build one error workflow that posts to Slack and attach it to all ten. Do it on day one.
Where do n8n AI workflows stop helping?
At the page. And this is my bias showing, so weigh it. I build product pages for a living.
But watch what happens when you put the two numbers side by side. Twelve hours a week at a hypothetical $40 an hour, over 4 weeks, is $1,920 of time back.
Now look at one page. On the Copper Bamboo Sheets rebuild, the conversion rate was 1.0% and the average order value was $125, so revenue per visitor was $1.25. On 10,000 visitors, that's $12,500. After the rebuild, the conversion rate was 4.3% and the average order value was $221, so revenue per visitor was $9.50. On the same 10,000 visitors, that's $95,000. Real client numbers, not typical results, and not a promise of what your store will do.
Time back is a hypothetical: 12 hours x 4 weeks x $40. Page lift is extra revenue on 10,000 visitors from one client product, not profit, and not a promise.
Those aren't the same kind of dollar. One's saved cost and the other's revenue. I'm still not going to pretend they're close.
You can automate a machine or you can automate a leak. n8n runs both at exactly the same speed.
That's why the leaderboard and the review sheet are my two favorites. They're the only workflows here that point you back at the page. The rest buy you the hours to go fix it.
And hand the rest of the repeat work to whatever's cheapest. This list of jobs for AI agents covers what you may not need n8n for at all.
What should you do next?
Build workflow 1 this week. Just the morning report: one schedule, one Shopify pull, one Slack message.
Then build number 10, and open the page at the bottom of that leaderboard before you touch anything else.
Book Your Profit Audit
n8n will give you your mornings back, but it can't make a weak page sell. Get your free profit audit and we'll find where your best product is leaking money and show you how to rebuild a high-converting product sales page in less than 15 minutes.
Or go here to check it out → revenueflows.ai
P.S. Automation makes a store cheaper to run. A better page makes it worth running. Do them in whichever order your numbers say.
Frequently asked questions
Is n8n good for Shopify automation?
Yes, when the job crosses several tools, like Shopify plus Klaviyo, Meta, Google Sheets and an AI model. For jobs that stay inside your store, start with Shopify Flow, because it's a free app and Shopify maintains it.
Does n8n have a Shopify integration?
Yes. n8n has a Shopify node for creating, updating, deleting and getting orders and products, plus a Shopify Trigger node. Anything else, like inventory or refunds, goes through the HTTP Request node and Shopify's API.
What's the difference between n8n and Shopify Flow?
Shopify Flow lives inside your store and is free on Basic, Grow, Advanced and Plus. n8n runs outside your store, combines many tools in one workflow and supports AI agents with human approval, but you maintain it yourself.
Is n8n free for Shopify stores?
The self-hosted Community Edition is free, though you pay for the server and handle the upkeep. n8n Cloud starts at 20 euros a month on annual billing for 2,500 workflow executions, and each full run counts as one execution.
Can n8n run AI workflows for a Shopify store?
Yes. n8n connects to model providers like OpenAI, Anthropic and Google, and you can require human approval before an AI agent uses a specific tool. Use that approval step on anything that touches a customer or money.