There is a line in every trading product between "the system told me" and "the system did it". Where a platform draws that line tells you almost everything about how it will behave on your worst day.
PocketX draws it early and states it plainly: alerts notify you; they do not trade for you. Everything below is the detail of how that works in practice.
What an alert is
A PocketX alert is a condition you defined, evaluated against market data, that has become true. When it fires, you are notified — with the instrument, the current price, and what the alert was for.
There are two shapes it can take:
- Alert triggered. The condition is met. This is information. Nothing further happens automatically.
- Alert action ready. The condition is met and a corresponding action has been prepared for your review. The action still requires you to open it, review it, and confirm with your trading PIN.
The second is the one worth understanding properly, because it looks the most like automation and is the least like it.
An action ready is a prepared draft, not a pending order
When PocketX prepares an action from an alert, it has assembled the details — instrument, side, quantity, product type — into something you can review in one place. That is all it has done.
Nothing is submitted until you review it and enter your trading PIN. Not a partial fill, not a resting order, nothing. The action is a draft with an expiry on it.
That expiry matters. Action links are deliberately short-lived. A prepared action that sits around for hours is a prepared action based on a market that no longer exists, and confirming it would be acting on stale reasoning. If the link has expired, that is the system refusing to let you trade yesterday's idea at today's price.
Why strategies stop at alerts too
The same boundary applies to the strategy workspace. Arming a cash strategy evaluates it on closed bars and creates alerts only. It does not place orders.
This is a design decision rather than a missing feature. Live strategy order execution is explicitly not part of the current release. The reasoning is straightforward:
- A rule that can place orders needs to be right about far more than direction. It needs to be right about liquidity, about sizing, about what to do when a leg fails, and about what happens when data goes stale mid-session.
- The failure mode of a wrong alert is that you ignore it. The failure mode of a wrong automated order is a position you did not choose.
- A system that notifies keeps the human in the loop at the moment it matters most, which is the moment something unexpected is happening.
Alerts fire on closed bars
Cash strategy conditions are evaluated on closed bars, not on every tick. This has two consequences you should internalise.
It removes a class of false signals. A spike that pierces your level intrabar and reverses before the close never becomes an alert. The condition is judged on a completed observation.
It costs you immediacy. You hear about the signal at the close of the bar. On a daily timeframe, that is the end of the session. If your strategy depends on reacting inside the bar, this is not the tool for it — and you should be honest with yourself about whether such a strategy was ever realistic for a retail participant.
Alerts do not control state
A principle that runs through the whole platform: notification delivery never controls financial state.
If an alert fails to deliver, nothing about your positions, orders, or strategy has changed. The state in the app is authoritative. A missing notification is a missing notification — not a silent failure of the thing it was describing.
The inverse also holds. Receiving a notification does not mean anything has been executed. It means something was observed, or something is ready for you to review.
Set alerts you will actually respond to
The most common way to make alerts useless is to set too many. Twenty alerts firing across a session trains you to dismiss them, and the one that mattered gets dismissed with the rest.
A short discipline that works:
- One alert per idea, at the level where your view would actually change. Not at a round number, and not at a level you would ignore.
- Set the exit alert when you set the entry alert. If you cannot state what would make you leave, you do not have a position thesis yet — see how to set a stop-loss.
- Delete alerts for ideas you have abandoned. Stale alerts are the main source of alert fatigue.
- Decide the response before the alert fires. An alert should trigger a pre-decided action, not a fresh round of deliberation while the price moves.
That last point is the entire value of the mechanism. The alert exists so you can make the decision calmly in advance and merely execute it later, rather than making it under time pressure with money on the line.
Choose channels deliberately
Alerts can reach you on more than one channel. Two rules of thumb:
- Time-sensitive alerts belong on a channel you actually check within minutes. An alert about a level being hit is worthless in an inbox you read in the evening.
- Everything else belongs somewhere quieter. Routing non-urgent notifications to a loud channel is how you end up muting the loud channel.
Prepared actions in particular expire. Route those to whatever reaches you fastest, or you will find yourself opening an expired link.
The summary
- Alerts tell you a condition became true.
- Prepared actions are drafts that expire and require your PIN.
- Arming a strategy creates alerts, never orders.
- Delivery of a notification never changes financial state.
If that reads as a list of things the product will not do for you, that is the correct impression. The about page states the same principle in one line: research and alerts supply context, and the decision remains with the customer.
Investments in the securities market are subject to market risk. Read all related documents carefully before investing. Nothing in this article is a recommendation to buy or sell any security, and past performance is not indicative of future results.