n8n vs Make for e-commerce automation: the real cost per 10,000 operations
Disclosure: some links in this guide are affiliate links. If you sign up through them, I may earn a commission at no extra cost to you. How this site makes money.
Make bills a credit for almost every module a scenario runs, so 10,000 credits ($9 a month on Core, billed annually) cover roughly 2,500 four-step abandoned-cart alerts but only 19 runs of a 500-SKU catalog sync. n8n Cloud bills one execution per workflow run, however many steps it has, and self-hosted n8n has no usage fee at all: you pay about $12 to $15 a month for a 2 GB VPS plus your maintenance time. If you count that time at $50 an hour, self-hosting pays off once you need more than about 80,000 Make credits a month.
"Make costs $9 a month and n8n is free" is true and unhelpful. The two tools count usage in different units, so the same store workflow can cost ten times more on one than on the other. This guide puts both on the same footing: what 10,000 operations actually buy for common e-commerce workflows, what a typical store's month costs on each, and the point at which paying for a VPS and your own maintenance time for self-hosted n8n beats a Make subscription.
n8n prices come from n8n's pricing page, and VPS prices from DigitalOcean's pricing page, both checked in October 2026. Make's entry price, $9 a month for 10,000 credits billed annually, was confirmed in October 2026. Make's pricing page blocks automated access, so the larger credit tiers come from an independent pricing tracker that read it in September 2026. The credits per workflow are our own counts of typical builds, and the maintenance hours are estimates, not measurements. Check the live pricing pages before you commit.
1. Credits vs executions: why "10,000 operations" means different things
| Platform | What you pay for | One four-step workflow run costs |
|---|---|---|
| Make | Credits. Most module actions use one credit: reading from an app or webhook (so the trigger counts), searching, creating and updating records, transforming data, iterating and aggregating. Routers and error handlers are free. AI features and code can use more. | 4 credits |
| n8n Cloud | Executions. One execution is one run of a whole workflow, however many steps it has and however much data it moves. | 1 execution |
| n8n self-hosted | Nothing per run. The free Community Edition has no execution limit. You pay for the server and your time. | $0 extra |
| Zapier (for reference) | Tasks. One task is one successful action step. The trigger is free. | 3 tasks |
The gap grows with every step that runs once per item. A Make scenario that loops over 500 SKUs and updates each one uses at least 500 credits per run. The same n8n workflow is still one execution.
2. What 10,000 operations buy, by workflow
These are typical builds for a Shopify or WooCommerce store. Your own scenarios may use a step more or less, so count yours the same way: one credit per module that runs, multiplied by how many times it runs.
| Workflow | Steps per run | Make credits per run | Runs per 10,000 Make credits | n8n Cloud executions for those runs |
|---|---|---|---|---|
| Abandoned-cart alert | Webhook, get customer, email, Slack | 4 | 2,500 | 2,500 (Starter's whole allowance) |
| New order to Google Sheets or ERP | Webhook, format, add row | 3 | 3,333 | 3,333 |
| Real-time stock push (stock changes in one channel, update the other) | Webhook, look up SKU mapping, update | 3 | 3,333 | 3,333 |
| Inventory poll every 15 minutes, about 3 changed SKUs per run | Schedule, fetch changes, iterate, 3 updates | 6 | 1,666 (17 days of 15-minute polls) | 1,666 |
| Nightly full catalog sync, 500 SKUs, one update each | Schedule, fetch feed, iterate, 500 updates | 503 | 19 (not even a month of nights) | 19 |
Two things stand out:
- Event workflows are cheap on both. Webhook-triggered alerts and order exports use a handful of credits each, and 10,000 credits cover a small store's month.
- Batch and polling workflows are where Make gets expensive. A scheduled check uses a credit every time it runs, even when nothing changed: a 15-minute poll is 2,880 runs a month before it does any work. Per-SKU updates multiply that again. On n8n, the batch run is still one execution.
On Make you can often cut the count by using an app's bulk endpoint, so one HTTP module updates many SKUs for one credit, and by using webhooks instead of polling wherever the app supports them. Do that before you compare platforms. It's free and often halves the bill.
3. Current prices
| Plan | Included per month | Price, billed annually | Price, billed monthly |
|---|---|---|---|
| Make Free | 1,000 credits, 15-minute minimum schedule | $0 | |
| Make Core | 10,000 credits | $9 | about $11 |
| Make Core | 40,000 credits | $38 | $51 |
| Make Core | 80,000 credits | $72 | $97 |
| Make Core | 150,000 credits | $129 | $175 |
| Make Pro / Teams | 10,000 credits | $16 / $29 | about $19 / $34 |
| n8n Cloud Starter | 2,500 executions, 5 concurrent | €20 | €24 |
| n8n Cloud Pro | 10,000 executions | €50 | €60 |
| n8n Business (self-hosted, licensed) | 40,000 executions, SSO, Git version control | €667 | about €800 |
| n8n Community Edition (self-hosted) | Unlimited executions | $0, plus the server | |
Make's Pro and Teams plans include the same 10,000 credits as Core. They add features such as priority execution and team roles, not more runs. The 10,000-credit prices are current as of October 2026. The larger Core tiers are from a September 2026 tracker, so check them on Make's pricing page. n8n prices are from n8n's pricing page in October 2026, in euros.
4. A typical store's month on each platform
A worked example for a store doing about 3,000 orders a month. The volumes are illustrative. Swap in your own.
| Workflow | Runs per month | Make credits | n8n executions |
|---|---|---|---|
| Abandoned-cart alerts | 1,500 | 6,000 | 1,500 |
| Orders to Google Sheets | 3,000 | 9,000 | 3,000 |
| Inventory poll every 15 minutes | 2,880 | 17,280 | 2,880 |
| Subtotal | 32,280 | 7,380 | |
| Add a nightly 500-SKU catalog sync | 30 | 15,090 | 30 |
| Total | 47,370 | 7,410 |
- Make: the subtotal fits the 40,000-credit Core tier ($38 a month billed annually). With the catalog sync, you need the 80,000-credit tier ($72), or the 40,000 tier plus extra credits.
- n8n Cloud: 7,410 executions fit Pro (€50 a month billed annually), with room to spare. Adding the catalog sync costs 30 executions.
- Self-hosted n8n: the server bill doesn't change. About $14.40 a month (section 5), plus your time.
5. When self-hosting n8n pays off
What self-hosting really costs
| Cost | Estimate | Notes |
|---|---|---|
| VPS with 2 GB RAM | $12 a month | DigitalOcean's 2 GB Basic Droplet is $12. Vultr's plans start at $5 for 1 GB. 2 GB gives n8n and PostgreSQL room in Docker. A 1 GB server works for light, SQLite-based setups. |
| Automatic backups | $2.40 a month | DigitalOcean's weekly backups cost 20% of the Droplet. Keep your own database dumps as well (below). |
| Initial setup | 2 to 4 hours, once | Docker Compose, PostgreSQL, a domain and HTTPS, a reverse proxy. Our self-hosted n8n guide walks through it. |
| Ongoing maintenance | About 2 hours a month | Our estimate, with setup spread over the first year: n8n and OS updates, checking backups and logs, and the occasional broken workflow after an upgrade. |
Maintenance isn't optional. In January 2026, n8n disclosed CVE-2026-21858 ("Ni8mare"), a CVSS 10.0 flaw that let unauthenticated attackers reach files on self-hosted servers with certain form workflows exposed. The fix was version 1.121.0. On n8n Cloud, n8n patched it for you. If you self-host, keeping up with updates is part of the price.
The break-even calculation
Self-hosting is cheaper when:
Make plan price > VPS + backups + (maintenance hours × your hourly rate)
$12 + $2.40 + (2 h × rate)
| Your time is worth | Self-hosted n8n per month | First Make Core tier that costs more (billed annually) | Self-hosting wins once you use more than |
|---|---|---|---|
| $0 (you enjoy it, or already run servers) | $14.40 | 20,000 credits ($21) | 10,000 credits a month |
| $25 an hour | $64.40 | 80,000 credits ($72) | 40,000 credits |
| $50 an hour | $114.40 | 150,000 credits ($129) | 80,000 credits |
| $100 an hour | $214.40 | 300,000 credits ($237) | 150,000 credits |
The brackets are the same with monthly billing. Buying extra credit bundles instead of moving up a tier shifts them a little. The example store from section 4 uses about 47,000 Make credits, so it needs the $72 tier: self-hosting wins there if your two hours a month are worth less than about $29 an hour.
Don't skip n8n Cloud in the comparison
Self-hosting isn't the only way to get n8n's per-execution pricing. n8n Cloud Pro costs €50 a month (billed annually) for 10,000 executions, with no server to maintain. For batch-heavy stores where Make needs 80,000 credits or more but n8n sees fewer than 10,000 runs, Cloud Pro is often the cheapest option that doesn't cost you any time. Self-hosting makes sense when you also need unlimited runs, local AI models or data that stays on your own server.
6. If you self-host: the maintenance checklist
# Monthly: update n8n (read the release notes for breaking changes first)
cd /opt/n8n && docker compose pull && docker compose up -d
# Nightly via cron: dump the database and keep 14 days
0 3 * * * docker compose -f /opt/n8n/docker-compose.yml exec -T postgres \
pg_dump -U n8n n8n | gzip > /opt/backups/n8n-$(date +\%F).sql.gz \
&& find /opt/backups -name 'n8n-*.sql.gz' -mtime +14 -delete
- Back up the encryption key. n8n encrypts stored credentials with
N8N_ENCRYPTION_KEY. A database backup without that key can't restore your store, email and Slack connections. - Copy backups off the server to object storage or another machine. A backup on the same disk doesn't survive losing the server.
- Watch uptime. When n8n is down, webhooks from your store fail. Most platforms retry for a while, but not forever. A free uptime monitor on n8n's
/healthzendpoint tells you before your customers do. - Don't expose more than you need. Put the editor behind HTTPS and strong authentication. Only workflow webhooks need to be public.
- Know the license. The Community Edition uses n8n's Sustainable Use License: running it for your own business, or building workflows for a client, is allowed. Selling hosted access to n8n is not.
7. Which one to pick
| Your situation | Best fit | Why |
|---|---|---|
| Under about 20,000 credits a month, mostly webhook workflows | Make Core | Cheapest for the time you spend, with a large app library and a visual builder that non-developers can maintain |
| Batch or polling jobs, per-SKU loops, fewer than 10,000 runs | n8n Cloud Pro | One execution per run, however many items. No server work. |
| High volume, a developer on the team, or data that must stay on your own server | Self-hosted n8n on a VPS | Flat cost, no run limits, and room for local AI models |
| Simple automations inside Shopify only | Shopify Flow | Included with Shopify's paid plans, no extra subscription |
8. Tools
Top picks
| Tool | Role | Why it fits |
|---|---|---|
| Hosted visual automation | The easiest way to automate a store without servers. Thousands of app integrations, a free plan with 1,000 credits to test with, and paid plans from 10,000 credits a month. Best when your workflows are event-driven. | |
| VPS for self-hosted n8n | Hourly-billed cloud servers in many regions, from $5 a month. Pick a 2 GB plan for n8n with PostgreSQL, near your store's customers or your main apps. | |
| VPS for self-hosted n8n | A 2 GB Droplet for $12 a month, backups for 20% more and clear tutorials. A comfortable choice for your first self-hosted service. | |
| Hosted n8n | Per-execution pricing with no server to run, from €20 a month for 2,500 executions. The middle option from section 5 for batch-heavy stores. |
Alternatives: free tools and tools we have no affiliation with
We don't earn anything from these. They're here because for some readers they're the better choice.
- Zapier: the largest app library and the simplest builder. Triggers are free and you pay per successful action. Professional starts at about $20 a month (billed annually) for 750 tasks, and it usually costs more than Make at the same volume.
- n8n Community Edition: the self-hosted version this guide's break-even math is based on. Free under the Sustainable Use License.
- Activepieces: an open-source automation tool you can self-host, with a simpler builder than n8n.
- Shopify Flow: Shopify's built-in automation for tagging orders, inventory alerts and similar store tasks, at no extra cost.
9. Pitfalls
- Comparing plan prices instead of usage. Count credits per run and runs per month first. The plan price means nothing without them.
- Polling when a webhook exists. On Make, a 15-minute poll costs 2,880 credits a month before it does anything useful. Use the app's webhook trigger where you can.
- Per-item loops on Make. Updating 500 SKUs one by one costs 500 credits per run. Look for a bulk endpoint first.
- Treating self-hosting as free. The server is cheap. Your time, the backups and keeping up with security updates are the real cost.
- Losing the encryption key. Without
N8N_ENCRYPTION_KEY, a restored n8n database can't decrypt your saved credentials. - Running out of credits mid-month. Make stops scenarios when credits run out. Turn on automatic credit top-ups on paid plans, or watch usage before a sale.
Make counts credits per module and n8n counts executions per run, so event-driven workflows cost about the same on both, while batch, polling and per-SKU workflows cost far less on n8n. For most small stores under about 20,000 credits a month, Make Core is the cheapest option once you count your time. For batch-heavy stores, n8n Cloud Pro at €50 a month is usually next. Self-hosting n8n on a $12 VPS pays off once you'd need more than about 80,000 Make credits a month and you value your time at around $50 an hour, or much sooner if you already run servers.