Shopify Flow is the automation tool many store owners overlook because it sits quietly in the admin and does not advertise itself. Yet for a large share of routine work, such as tagging, alerting and holding orders, it is the cheapest and most direct option, because it lives inside the platform and needs no separate login. The difficulty is knowing what to build with it. A blank canvas with a trigger box is not a prompt.
This guide answers that with twelve concrete Shopify Flow examples, each explained in terms of its trigger, its conditions and its actions, and with a note on what to watch for. It also explains how Flow is structured, how to test a workflow safely, the traps that catch beginners, and the signs that a job has outgrown Flow. Features and plan availability change over time, so confirm the details in the Shopify Help Center before you build anything that touches orders or stock.
How Shopify Flow works in one minute
Every workflow in Flow has the same three parts. A trigger is the event that starts it, such as an order being created, a customer being created or an inventory quantity changing. Conditions are optional checks that decide whether to continue, such as order total above a value or shipping country not in a list. Actions are what happens next, such as adding a tag, sending an email to your team, creating a note or changing a product status. You can chain conditions and use branches, so one trigger can lead to different outcomes.
| Part | What it is | Example |
|---|---|---|
| Trigger | The event that starts the workflow | Order created |
| Condition | A test that must be true to continue | Order total is greater than 250 |
| Action | What the workflow does | Add the tag high-value and notify the team |
| Data | Fields from the event you can use | Customer email, line items, shipping country, risk level |
Flow also connects to some apps, which can add their own triggers and actions, and it can send an HTTP request to an outside address, which is how many stores bridge to other tools. Which of these are available to you depends on your plan and on the app, so read the current documentation for your setup.
Twelve Shopify Flow examples
1. Tag high-value orders
Trigger: order created. Condition: order total above your threshold. Action: add the tag high-value and notify the team. This is a classic first workflow, because the tag then drives other rules, such as priority packing, a handwritten note or a manual check for fraud. Choose the threshold from your own data, for example the top five percent of orders.
2. Tag and hold risky orders
Trigger: order risk analyzed. Condition: risk level is high. Action: add the tag review-risk and hold fulfilment until a person looks. Flow can read Shopify’s own risk analysis, which makes this a low-effort protection. Make sure a person actually reviews the held orders, and set a time limit, otherwise good customers wait forever. Our guide to chargeback reason codes explains why fraud holds matter.
3. Low stock alert
Trigger: inventory quantity changed. Condition: quantity is at or below your reorder point. Action: send a message to the team channel or an email, naming the product and the quantity. To avoid a flood of alerts, add a condition that checks whether the previous quantity was above the threshold, so that you are told once when it crosses, not every time it changes.
4. Hide a product when it sells out
Trigger: inventory quantity changed. Condition: quantity is zero and the product is not marked backorder. Action: set the product status to draft, or add a tag that your theme uses to hide it. The reverse workflow can restore it when stock returns. Be careful with search rankings: temporarily hiding a product can lose its page position, so many stores prefer a sold-out label to a hidden product.
5. Tag customers by lifetime spend
Trigger: order paid or customer updated. Condition: total spent above a threshold. Action: add the tag vip. Segments built from tags can feed your email platform, support priority and special offers. A more careful method of grouping customers combines recency, frequency and spend.
6. Flag first-time buyers
Trigger: order created. Condition: customer order count equals one. Action: add the tag first-order, and use it to trigger a welcome note in the packing slip or a different email sequence. First-time buyers are the most uncertain and the most valuable to win over, so a small touch is well spent.
7. Hold preorders separately
Trigger: order created. Condition: any line item has the tag preorder. Action: tag the order preorder-hold and add a note with the expected ship date. This prevents preorders being packed with in-stock items by mistake and gives support a single place to look.
8. International order check
Trigger: order created. Condition: shipping country is outside your usual zones. Action: add the tag international and, if you need it, notify the person who prepares customs details. Rules like this make exceptions visible instead of discovered at the packing bench.
9. Notify support about a refund
Trigger: refund created. Condition: refund amount above a set value. Action: notify support with the order number and the reason, so that a person can follow up. This turns a silent money event into a visible one and catches mistakes early.
10. Archive completed orders
Trigger: order fulfilled. Condition: order is paid and delivered or a set number of days have passed. Action: archive the order. A tidy admin makes it easier to see what still needs attention. Check that your accounting or reporting does not depend on the archived status.
11. Tag orders by sales channel
Trigger: order created. Condition: source name matches a channel, such as an online store, a social channel or a point of sale. Action: add a tag with the channel name. Reporting by channel becomes a simple filter instead of a spreadsheet exercise.
12. Send an HTTP request to another system
Trigger: order created. Condition: order contains a product from supplier A. Action: send an HTTP request with the order details to a supplier or a workflow tool. This is how a simple Flow reaches outside Shopify. It is powerful, and it is also where problems begin, because a failed request must be noticed and retried. We return to this below.
Build a workflow safely: a short method
- Write the rule in plain words first. When X happens and Y is true, do Z. If you cannot say it in one sentence, split it.
- Start with a harmless action. Add a tag or send a notification before you hold, cancel or change anything.
- Test with a real order. Place a low-value test order that meets your conditions and check that the run history shows what you expected.
- Check the negative case. Place an order that should not trigger it and confirm that nothing happens.
- Add a guard against repeats. Include a condition that checks whether the tag or note already exists.
- Name and describe it. A workflow called Untitled 4 is a future problem. Write what it does and who owns it.
- Review the run history weekly for a month. Look for errors and surprises before you trust it with larger changes.
Common traps
- Triggers can fire more than once. An order can be updated several times, and a workflow that adds a note on each update will add several notes. Guard against it with a condition.
- Two workflows that fight. One adds a tag and another removes it. Keep a list of every tag and which workflow owns it.
- Silent failures. A workflow that errors will not tell you unless you look. Check the run history and, for important ones, build in a notification when an action fails.
- Overlap with an app. If an app already tags orders, a Flow that does the same may conflict. Know who owns each job.
- Rate and plan limits. Plans and connected apps can limit what Flow can do. Confirm limits in the documentation instead of discovering them in production.
- Changing the shop without a test. Editing a live workflow is editing live behaviour. Duplicate it, test the copy and then replace the original.
When Flow stops being enough
Flow is excellent when the job lives inside Shopify. It becomes awkward when the job needs several outside systems to agree, when a person must approve a step, when you need retries and a queue to survive an outside outage, or when you need a clear log that a non-technical colleague can read. Those are the situations where owners start bolting together HTTP requests and workarounds, and where a separate engine or a properly built system pays for itself.
| Sign Flow is no longer enough | What helps instead |
|---|---|
| The workflow calls an outside system and you cannot tell when it fails | A workflow engine or a build with alerts that name the step and record |
| Several systems must agree on one fact, such as stock | A designated source of truth with a sync, not several competing writers |
| A person must approve before something happens | An approval step with a queue and a record of who decided |
| You need to re-run a failed step safely | Idempotent actions with keys, so a retry does not duplicate |
| Volume spikes hit limits during a sale | Queues and batching that respect each system’s rate limits |
For stock across channels, see our inventory sync service. For routing orders to the right supplier and recording what was sent, see order routing to suppliers. For the wider question of which tool suits which job, start with how to automate a Shopify store.
A starter set to build this week
If you want a short list rather than twelve, these four give the most value for the least risk, and each is easy to reverse.
| Workflow | Why start here | Reversible? |
|---|---|---|
| Tag high-value orders | Makes valuable orders visible and drives other rules | Yes, remove the tag |
| Low stock alert | Prevents stockouts without changing anything | Yes, turn off the workflow |
| Tag first-time buyers | Gives you a segment for welcome and support | Yes |
| Notify on large refunds | Makes money events visible | Yes |
After a month, review which tags are actually used. A tag nobody uses is clutter. A tag used by three other rules is the foundation of your automation, and it deserves a written definition so that nobody changes its meaning by accident.
A worked scenario: one order, four workflows
It helps to follow a single order through a small set of workflows, because the value of Flow comes from how the pieces combine. Imagine a customer in Canada places an order for three items worth three hundred and ten dollars, one of which is a preorder. Here is what a well-organised set of workflows does without anyone touching it.
| Moment | Workflow that runs | Result |
|---|---|---|
| Order created | Tag high-value (over 250) | Tag high-value is added; the team channel is notified |
| Order created | International order check | Tag international is added; the customs note task is created |
| Order created | Hold preorders separately | Tag preorder-hold is added with the expected ship date in a note |
| Risk analyzed | Tag and hold risky orders | Risk is low, so nothing happens and the order flows on |
| Fulfilled and delivered | Archive completed orders | The order is archived after the delay you chose |
Nothing here required a person to read the order, yet the packing team sees the right tags, finance sees the high-value flag and support sees the preorder note. The same order without these rules would have been packed in the usual way, shipped incomplete or discovered as a problem after the customer wrote in. This is the kind of quiet gain that does not show up as a single big number but removes dozens of small mistakes.
Using Flow together with an email platform
Flow is not an email tool, but its tags are a convenient bridge to one. If your email platform reads Shopify customer and order tags, a Flow that adds vip or first-order creates a segment you did not have to build by hand. Be careful about the direction of the data: tags written by Flow flow outwards, and a segment in the email platform cannot change a tag back. Keep one system responsible for each fact, and write down which one.
This is also the moment to think about consent. A tag that means “bought in the last thirty days” is safe to use for service messages, but a marketing message needs clear permission from the customer. Do not let a convenient tag blur that line. Our compliance basics page links to the official guidance for the regions you sell to.
Permissions and who may edit workflows
A workflow can change products, hold orders and notify customers, so editing one is a sensitive act. Limit access to the few people who understand the rules, give them their own staff accounts instead of a shared login, and require two-step sign-in. If an agency or a freelancer builds workflows for you, ask them to document each one and to hand over the editing rights when the project ends, so that you are never locked out of your own store logic.
Peak weeks and Flow
During a large sale, order volume can rise many times over, and a workflow that works at forty orders a day may behave differently at four hundred. Before a peak, check three things. First, whether any action sends something outside Shopify, because outside systems have their own limits. Second, whether alerts will flood a channel to the point that nobody reads them. Third, whether any rule holds orders and could pile up faster than a person can review. A short rehearsal with a batch of test orders is far cheaper than discovering a bottleneck on the day.
Tags, naming and ownership
The long-term health of your Flow setup depends on discipline about names. Choose a pattern, such as a short prefix that says what a tag is for, and stick to it. Keep a one-page register that lists every workflow, what it triggers on, what it does, which tags it writes and who owns it. When something unexpected happens to an order, that page tells you in a minute which workflow to look at. Without it, you are searching a list of similar-looking names.
- Prefix tags by purpose, for example risk-, ship-, vip-, so that a glance tells you the kind of rule.
- Never reuse a tag for a second meaning. Make a new tag instead.
- Record the owner. The person who knows why a workflow exists is the person who can change it safely.
- Date your changes. A short note with the date of each edit explains behaviour that changed.
Measuring whether a workflow earns its keep
Every workflow should have a number that tells you whether it was worth building. For the high-value tag it might be the number of orders that received the extra care and the effect on repeat purchases. For the low-stock alert it is the number of stockout days before and after. For the refund notification it is the time it takes someone to respond. If you cannot name the number, the workflow may be tidying for its own sake. Review them quarterly and remove the ones that do nothing, because each workflow is a small piece of behaviour that someone must understand.
Sources and further reading
Figures and rules in this guide come from the primary pages below. Platform limits and prices change, so open the page before you build on a number.
- Shopify Help Center: Shopify Flow, triggers, conditions, actions and plan availability
- Shopify developer docs: webhooks, how Shopify delivers events and what can repeat
- Shopify Help Center, where each feature referenced here is documented