Shopify Automation Without Apps: Tested on Five Stores
Shopify automation without apps, tested on five stores — what Flow, the Admin API, and webhooks replace for free, and where I write code instead.
Key points — AI summary
- An app-by-app audit across five stores found many paid apps had become redundant with free native Shopify features — the test was whether anything actually breaks if you delete the app today
- Shopify Flow (free on every plan) handles order tagging, risk flagging, customer spend tiers, and low-stock alerts in production, but each Flow lives inside one store's admin — five stores means five copies that can drift
- Flow's "Send Admin API request" action reaches most of the GraphQL Admin API for free, including bulk tag updates and metafield writes with no dedicated Flow action yet — but test every mutation in GraphiQL first, since an overly broad mutation once mistakenly tagged a pile of unrelated orders
- Two apps survived the cutting exercise — a timed abandoned-checkout email sequence and a review app — because native Shopify had no equivalent good enough to replace them
- Native automation hits a hard wall at cross-store workflows and events Shopify can't see, like a parcel stuck at a carrier — that gap is what a consolidated multi-store dashboard and a custom tracking-lifecycle bot stack exist to cover
Summarized from this article by our writing pipeline; reviewed by the author.
On this page
In April 2026 I opened the Apps page on each of the five Shopify stores we operate and asked the same question of every line item: what actually breaks if I delete this today? The audit took an afternoon per store. For an uncomfortable number of apps, the honest answer was "nothing" — the job they were installed for had quietly become something Shopify does natively, for free.
I should declare the bias up front. Every installed app is a recurring charge, a new set of data permissions, and one more dashboard somebody has to remember to check. On one store you can shrug that off. On five, all three multiply, and the shrug gets expensive. So this post is my working map of Shopify automation without apps: the native rules that genuinely run our stores, the one free power feature most merchants never touch, the two apps I tried to delete and couldn't, and the structural line where I stop deleting and write code instead.
If you take one action from this article, make it the audit itself. The rest is what I found on mine.
The rules that run five stores without a paid app
Shopify Flow is free on every plan from Basic up, and it's the workhorse. Every workflow is a trigger (an order was placed), optional conditions (total over $500), and actions (tag it, hold fulfillment, notify someone). Here's what actually lives in Flow on our stores — running in production, not theoretically possible:
- Order tagging. International orders, PO-box addresses, and orders above a value threshold each get a tag our team routes by. This alone replaced a paid tagging app on two of our stores in the April cleanup.
- Risk flagging. High value plus a billing/shipping mismatch gets tagged and held for a human. Flow never cancels an order on its own here — that's a choice, not a gap.
- Customer spend tiers. Repeat buyers past a spend threshold get tagged, so email segments build themselves instead of someone maintaining a spreadsheet.
- Low-stock pings. When a variant drops below its threshold, we hear about it before the storefront sells inventory we don't have.
Since this spring, Sidekick drafts most new rules for us from a plain-language sentence, which quietly changed how many small annoyances we bother to automate. I've written the full Flow and AI audit up separately, including the caveat that matters most in this article: a Flow lives inside one store's admin. Five stores means five copies of every rule, and copies drift — I found the proof in our own workflows this March.
Before you install anything, also check what the admin already ships outside Flow: automatic discounts (where discount and shipping logic belongs now that Shopify Scripts stopped executing in mid-2026), native abandoned-checkout emails, staff notifications, and scheduled reports all exist now without a single install. Flow itself ships 150+ pre-built templates, per Shopify's docs — I treat them as a catalog of what's possible, not things to switch on unmodified.
The free power feature almost nobody uses
Flow has an action called Send Admin API request that runs most mutations in the GraphQL Admin API directly — including operations that have no dedicated Flow action yet. GraphQL has been Shopify's primary API surface since the REST Admin API went legacy in October 2024 (Shopify's timeline, not my estimate), so this is the modern path, and it means Flow's ceiling is far higher than the visual builder suggests: bulk tag updates, metafield writes, fulfillment adjustments, all without an app.
One habit from a mistake I made in May: test every mutation in the GraphiQL explorer before it touches production. I once scoped a tag mutation wider than I intended and tagged a pile of orders that had no business being tagged. Recoverable — tags are cheap to remove — but only because the mistake happened to be a cheap one.
Webhooks, and the plan-gating nobody mentions
On Grow, Advanced, and Plus plans, the Send HTTP Request action turns Flow into a bridge to anything with an endpoint: Slack, Discord, Google Sheets, your own services. This is how our Flow rules reach the channels our team actually reads instead of dying in admin notifications. On Basic that action is gated, and the usual workaround is Flow Webhooks — which is, yes, technically an app. It's also free, and my quarrel was never with the word "app"; it's with recurring charges and permission sprawl. For the details on triggers, conditions, and limits, Shopify's Flow documentation is genuinely good.
The two apps I tried to delete and couldn't
The audit wasn't a clean sweep, and pretending otherwise would make this article useless.
The abandoned-checkout app survived. Shopify's native recovery email is a single message; our app runs a timed sequence, and the sequence recovers enough carts that deleting it felt like burning money to make a point. Native gets you one shot — if a sequence is paying for itself on your store, keep it. The review app survived too, because there's still no native equivalent I'd put in front of customers.
That left me with the rule I now apply quarterly: an app survives only if it does something native can't, well enough that I'd notice its absence within a week. Everything that failed that test in April got deleted, and between April and June I haven't reinstalled one of them.
Where native automation ends for me
Two boundaries, and both are structural rather than fixable with cleverer Flows.
First, anything cross-store. Flow has no concept of "across stores" — a morning briefing or a consolidated revenue view spanning five admins simply isn't a workflow it can express, no matter how many copies you maintain.
Second, events Shopify can't see. A parcel sitting silent at a carrier for six days is not a Shopify event, so no trigger will ever fire for it. Our stuck-shipment alerting runs on a tracking lifecycle in our own bot stack, which over time grew into a full multi-store agent — a different article's worth of story.
That second layer is what we ended up building. A consolidated dashboard for operating many Shopify stores — orders, revenue, shipping in one view — with automatic order sync to Google Sheets, bulk shipment tracking with stuck-shipment alerts, and consolidated financial reporting across the fleet.
But I'll end on the honest version: if you run one store, you don't need any of this. Turn on Flow, run the app audit, keep your money. It's when every new rule has to be copied four more times that a cross-store layer stops being a luxury.