Store pulse watches your Shopify store and surfaces low-stock products and high-risk orders automatically, in the room where the team is already working, with the product or order attached.
Plenty of store problems are visible in the data long before a person spots them. Stock crossing a sensible threshold. An order flagged as risky sitting unactioned while fulfilment runs.
Relying on somebody noticing works until the day they are busy, which is the day it matters. Alerts convert noticing into responding, and the difference in reliability is large.
Store pulse works from the moment the app is installed, with nothing to configure before it is useful.
The test worth applying to any alert: what will the person receiving this do differently in the next five minutes? If the answer is a specific action, it earns the interruption. If the answer is "be aware", it is a report and belongs in a summary.
This is why alerts land in a room with the product or order attached rather than as an email saying something happened. Acting on it is one step, not three.
Most alert fatigue is an audience problem rather than a threshold problem. An alert reaching someone with no authority to act and no responsibility for the outcome is not information, it is noise, and it teaches that person to skim the channel that also carries the alerts they needed.
Worse, it removes a safety net rather than adding one: when several people receive something, each reasonably assumes one of the others has it.
Roles and room membership decide who sees which alerts, so stock alerts can reach whoever places orders and nobody else.
An alert reports a fact. A task assigns a response. Without the second step, a low-stock alert is something somebody saw and intended to deal with, which is not the same as work that will happen.
Alerts turn into tasks with an owner and a due time in a click, with the product already attached.
An alert that says "stock is low" has moved the problem rather than solved it, because the reader still has to work out which product, how low, and whether it matters.
A useful one carries the product itself, so the person reading it can see the variant, the current level and the recent rate without opening another screen. That is the difference between an alert that prompts an action and one that starts a task.
It also has to arrive while there is still time to act, which is a question about your replenishment time rather than about the alert. A threshold set at a round number will fire too late on anything with a long lead time, and the lead time people use is usually the supplier's quoted figure rather than the measured one including the days a delivery sits in receiving.
A high-risk label means several fraud indicators matched patterns associated with fraudulent orders. It is a prompt to look, not a verdict, and plenty of legitimate customers trip it: gift purchases to a different address, travel, a corporate VPN.
Hold fulfilment first. It is reversible, costs nothing, and it is the only genuinely time-sensitive step. Investigating while the order ships is how a recoverable situation becomes a chargeback.
Then read which indicators fired, rather than the overall label. Payment signals such as a failed address check carry far more weight than location signals, which produce most false positives.
Then pick one of three: fulfil, verify with a light-touch message, or cancel and refund promptly. Sitting in a fourth undecided state is the outcome that costs the most.
Alert fatigue is not usually caused by volume. It is caused by alerts reaching people who cannot act on them, which teaches those people to skim the channel that also carries the ones they needed.
Route by who can act. Stock alerts to whoever places orders, and generally nobody else. Wider distribution feels like a safety net and does the opposite: when several people receive something, each reasonably assumes one of the others has it.
The exception is a product in an active campaign. Then three people need to know quickly, because you are otherwise paying to advertise something you cannot sell: whoever reorders, whoever can pause the advertising, and whoever will answer the customers asking where it went.
An alert reports a fact. A task assigns a response. Without the second step, an alert is something somebody saw and intended to deal with, which is not the same as work that will happen.
Because the alert lands in a room rather than an inbox, turning it into an owned task with a due time is one action, and the product stays attached. That matters most on the days it is most likely to be skipped.
Most stores start with alerts by email because it is what the platform offers, and it works until volume rises. Then three things go wrong at once.
It arrives for one person. If that person is packing, driving or off, the alert has been delivered and not received, and there is no way to tell the difference from the outside.
It cannot be replied to usefully. A colleague's answer to an alert you got by email is not visible to anybody else, so the same question gets asked again next week.
It competes with everything else. An inbox contains suppliers, couriers, marketing and a stock alert, ranked by arrival time rather than urgency, and the alert loses.
A room fixes all three by default: the right group sees it, the response is attached to it, and the room contains only operations. Email still has a place as a backup for the person who genuinely needs to be interrupted when they are not looking at the app.
Alerts tell you what is happening now. The pattern across a fortnight tells you what to change, and nobody sees the pattern unless somebody deliberately looks.
Ten minutes a week is enough. Scan the alerts, and pick out anything that has fired more than twice. A product that repeatedly hits its threshold either has the wrong threshold or the wrong reorder quantity, and either is a five-minute fix that removes a recurring interruption permanently.
Do the same with high-risk orders. If the same indicator keeps firing on orders that turn out fine, that is worth knowing before you cancel a real customer over it.
Low-stock products and high-risk orders, surfaced in Store pulse. Both work from install with no configuration, and both arrive with the product or order attached.
Yes, through roles and room membership. Sending stock alerts only to whoever places orders makes action more likely, not less, because shared visibility spreads the assumption that somebody else is handling it.
Email notifications cover you when you are away from Shopify admin, and you control which events reach you.
It is a different job. Shopify tells you something happened; Store pulse puts it in front of the team with the record attached so it can be discussed and assigned without leaving the conversation.
Shopify's fraud analysis reviews each order and labels the risk. A high-risk label means several indicators matched patterns associated with fraud, which is a prompt to look rather than a verdict.
Free for teams up to five, and about two minutes to connect your store.