WooCommerce can push order data outwards on its own, and Zapier or n8n are just the translator between it and the rest of your tools. Zapier is a paid cloud service where you click together ready-made connections with no code. n8n is the same kind of tool, except you can host it on your own server and pay for compute rather than per operation. Both read the same source: a new order, a status change or a new customer in the store, and pass it on to a CRM, an email platform, a warehouse system or a spreadsheet.
What you can actually automate
Automation in WooCommerce always looks the same: a trigger (something happened in the store) plus an action (something should happen in another tool). The most common pairs:
| Trigger in the store | Action outside | What for |
|---|---|---|
| New order | Contact and deal in the CRM | Sales sees the customer without retyping anything |
| Order paid | Invoice issued in the accounting system | No more manual data entry |
| Status changed to “shipped” | SMS or message to the customer | Fewer “where is my parcel” questions |
| New customer | Added to a mailing list with a tag | Segmentation from day one |
| Low stock | Slack notification or email to the supplier | You do not run out of goods quietly |
The rule I stick to on every rollout: automate what you do by hand and at least a few times a week. Something done once a quarter will never repay the time spent building it.
Zapier or n8n - how they differ
They do the same thing in a different model: Zapier is more convenient to start with, n8n is cheaper at high volume and more flexible once the logic gets twisty.
| Zapier | n8n | |
|---|---|---|
| Where it runs | Vendor cloud only | Vendor cloud or your own server |
| Billing | Per operation | Per operation or just the server (self-hosted) |
| Entry barrier | Lowest, everything is clicked | Higher, looks like a flow chart |
| Branches and loops | Limited, get pricier with the plan | Natural, no surcharge |
| Custom code inside | Possible, but modest | A full code node |
| Customer data | Passes through the vendor’s servers | Can stay inside your infrastructure |
| Risk | The bill grows with turnover | Somebody has to maintain the server |
The pricing of both changes several times a year and both have free tiers, so check the current brackets at the source. Rule of thumb: a dozen-odd orders a day and two or three simple automations means Zapier. Hundreds of orders, or one event firing five things, means self-hosted n8n. Selling on marketplaces too? Compare it with multichannel platforms - I covered those in the piece on the Baselinker API.
How WooCommerce pushes data out in the first place
WooCommerce has two built-in mechanisms and needs no plugin for either.
Webhooks are messages the store sends the moment something happens. You set them up in the dashboard (WooCommerce → Settings → Advanced → Webhooks), pick a topic such as “order created”, and paste in the address you got from Zapier or n8n. From then on the store reports in on every order, immediately, without anything polling it in circles.
The REST API works the other way round: an external tool asks the store for data. You generate keys in WooCommerce → Settings → Advanced → REST API, choosing permissions (read, write or both). You use this when you want to write something into the store - change a stock level, add a note to an order, update a price.
A key with write permission has access to the whole store, so treat it like a bank password and grant read-only if the integration changes nothing. Remember too that webhooks fire from the store’s server - when the store is overloaded, some of those messages will vanish. In that case: performance first, automation second.
Connecting a CRM: HubSpot, Zoho, Klaviyo
Connecting WooCommerce to a CRM means moving the customer and their orders into the tool your sales team works in. The details depend on why you have a CRM at all.
- HubSpot - the ready-made connection handles contacts, companies and deals. It works well when an order should kick off a sales process: a call, a proposal, an upsell.
- Zoho CRM and Zoho Inventory - for companies that want CRM and stock in one ecosystem. Contacts and orders come across smoothly; the trouble starts with product variants and units of measure, which get mapped by hand.
- Klaviyo - not a CRM but e-commerce email marketing. What matters is passing purchase events, not just addresses: what they bought, for how much, and when they last did.
The same hygiene rule applies to all of them: one leading field. If a customer address can be changed both in the store and in the CRM, sooner or later it will diverge on an invoice. Decide which tool is the source of truth for each field, and switch off writes from the other side. I went into this at more length in the piece on connecting a CRM to an ERP and a store.
Where no-code stops and what it costs
Zapier and n8n handle the “one event, one action” scenario brilliantly. Trouble starts with:
- High volumes - several hundred operations a day on per-execution billing turns automation into a second hosting bill.
- Twisty logic - “if the customer is wholesale and ordered above a threshold, then a different warehouse, price and document template”. You can assemble it out of blocks, but you end up with a tangle nobody understands a year later.
- Two-way sync - stock levels flowing both ways is the classic trap for an infinite update loop, described in the piece on inventory sync.
- Documents and formal requirements - invoices, credit notes and KSeF, Poland’s mandatory e-invoicing system, which has covered the largest companies since February 2026 and everyone else since April 2026. Here what counts is correctness and the ability to resend, not how pleasant the clicking is.
When you hit two items on that list, an integration written directly is cheaper: the store talks to the CRM or ERP with no middleman. The code sits in a store plugin, it has logs and retries, and you do not pay per event.
| Option | Recurring cost | When it makes sense |
|---|---|---|
| Zapier, entry plan | Subscription growing with the operation count | 1-3 simple automations, low volume |
| n8n on your own server | Server plus maintenance time | Medium and high volume, twisty logic |
| Custom-built integration | Maintenance only, zero per-event fees | Formal requirements, two-way sync, high volume |
Prices change often enough that I do not quote them from memory - work them out from the current price lists and multiply by a realistic, not optimistic, order count. The arithmetic from the piece on the real cost of plugins helps here: single automations are cheap, added up they can surprise you.
On my side, projects are billed fixed-bid, usually 5,000-10,000 PLN net, and day-to-day maintenance runs under technical support (Basic from 1,000, Pro from 2,000, Premium from 3,500 PLN net per month).
Rollout step by step, and the traps
- List processes, not tools - the actions clicked by hand plus how many times a week they repeat.
- Pick one process to start with, the most frequent and the simplest, for example a new customer landing in your mailing list.
- Take a backup and test on a staging copy of the store, not on the live one.
- Set up the webhook or API keys with the minimum permissions needed.
- Test on a test order - did the data arrive, and did it land in the right fields.
- Add further automations one at a time, giving each one a week of observation.
Three things break these rollouts most often. Duplicates: a status change can fire the same scenario several times, so every event needs a marker saying it has been processed. Ordering: payment information sometimes reaches the CRM before the order itself, and the integration has to know how to wait. Silent failures: an alert for no traffic matters as much as an alert for errors, because a connection can die on Thursday and nobody notices until Monday. And if the store does not deliver email at all, no automation will fix that - it is a separate problem.
If you want this sorted without experiments
I start with a free diagnosis: I look at the store and at the list of things done by hand, then say plainly what can be wired up with off-the-shelf tools in a day and what needs an integration written. Often half the list is Zapier and a few clicks, and the real cost sits in one process nobody counted.
For bigger work you get a written quote before the start: what gets built, in what order, what happens when it fails and what it costs. I write the code under optimization, and if the integration then simply has to keep working with someone on call, ongoing support takes over.
Frequently asked questions
Do I need a plugin to connect WooCommerce to Zapier? You do not have to install anything. WooCommerce has built-in webhooks and a REST API, and that is enough for Zapier to receive orders and write data back. Dedicated plugins can be handier for rarer events, but they are not required.
Zapier or n8n for a store? With a few simple automations and low volume, Zapier is faster to get running. With a high order count, branched logic, or when customer data should not leave your infrastructure, self-hosted n8n comes out cheaper.
Will automation slow my store down? Done properly it should not, because a webhook sends a short message and does not wait for a reply. Problems come from integrations that poll the store in circles or run heavy operations on every customer click - those get moved into background jobs.
When is writing a custom integration better than using Zapier? When the monthly operations bill matches the cost of maintaining code, when the logic has many conditions, when the sync has to work both ways, or when documents and formal requirements are involved. Compare the annual cost of the off-the-shelf option against the one-off cost of building.
Sitting on this yourself and unsure where to start? I speed up and repair existing sites and stores - Core Web Vitals, outages, the features that are missing. Tell me what is happening and you will get a concrete quote with a date.