Shopify automation
Open your app bill and count. Eleven subscriptions is normal; four of them overlap, two broke at the last theme update, and the one thing your shop specifically needs is still a spreadsheet somebody maintains by hand. Some of that is the honest price of Shopify automation and some of it is not, and telling the two apart is worth doing before the next renewal.
Where to draw the line
Being honest about what not to build is most of the value here, because building something the platform gives you free is a bad trade.
- Use Flow for simple triggers. Tag a customer, hold a risky order, send an internal alert. If a built-in rule does it, use the built-in rule and spend the money elsewhere.
- Buy the app when it is a solved commodity. Reviews, basic email, shipping labels. These are competitive markets with good products, and a build will lose to them.
- Build where your business is specific. Pricing rules nobody else has, stock across three sales channels and a warehouse, a bundling logic that is your own idea, a supplier feed in a format only your supplier uses. This is where the apps stop and the spreadsheets start.
- Build where the apps have to talk to each other. The failure is rarely inside one app. It is between them, at three in the morning, with nobody watching.
What is usually worth building
- One reliable stock position across Shopify, any marketplaces and the warehouse, with the rule about what happens when they disagree written down and enforced.
- Product data assembled once and pushed everywhere in the right shape, instead of maintained separately per channel.
- Order exceptions caught early: address that will fail, item that cannot ship where it is going, payment held for review, a supplier who has not confirmed.
- Answering customers from real order data — where it is, what was in it, what the policy says — with anything contentious handed to a person.
Hard limits, written into the build
- Change prices on its own beyond rules you wrote and can read back. Automated pricing that surprises the owner is how a bank holiday becomes expensive.
- Cancel or refund without a limit. Under a threshold with clear reasons, fine. Above it, a person.
- Email your whole customer list. Anything going to more than a handful of people passes a human first.
- Silently overwrite stock. When two sources disagree it must say so, not pick one.
Nothing here is worth building if
One sales channel, a few hundred orders a month, apps that work: there is nothing on this page worth paying for, and you should spend the money on traffic instead. Nobody watching whether the automations ran: fix that before adding more of them, because the most expensive failure in a Shopify shop is the silent one. And if you are about to replatform, wait — everything built now gets rebuilt anyway.
Built already, on other shops
- Recommendation shelves — cross-sell that maintains itself: 968 products tagged by their real attributes and the shelves rebuilt from those tags, instead of a list somebody has to keep current.
- New-arrivals radar — the scheduled half done plainly: a category scanned on a schedule and the movers ranked, with no judgement pretended. Knowing which half of a shop's automation only needs a schedule is what keeps the bill down.
Questions from shop owners
Is Shopify automation not just Flow plus apps?
For a lot of shops, yes, and I will say so. Flow handles simple triggers and the app market handles commodities well. What neither handles is logic specific to your business, and the joins between apps where the failures actually happen.
How do I know whether to build or buy?
If a well-reviewed app does it and the app market is competitive, buy. If you are maintaining a spreadsheet to patch around the app, or paying for four apps to achieve one outcome, that is the signal to build.
What breaks most often?
The joins. One app updates, another stops receiving what it expected, and nothing complains. Any build worth paying for includes something whose only job is to notice that silence.
Can it manage stock across Shopify and marketplaces?
Yes, and that is usually the highest-value build for a multi-channel shop. The hard part is not the syncing, it is deciding what happens when two systems disagree — and that rule has to be yours, written down, before anything is built.
Will custom automation survive a theme change?
It should, because it should not live in the theme. Anything built into a theme is a hostage to the next redesign, which is one reason so many app-based setups break at the worst moment.
The spreadsheet is the specification
Whatever you still maintain by hand is the exact description of what your apps do not do, which makes it the most useful thing you can bring to a conversation. Send it with your app list and their monthly cost. Quite often the answer that comes back is that you need fewer apps rather than more code, and you will get that answer in the free hour rather than in a quote.
Book that hour →