About Exness

The units every step is measured in

• A step measured in minutes is handled by software alone. It behaves the same at two in the morning as at midday, and watching it changes nothing.
• A step measured in hours has a person in it. People work in shifts, so the same step started at different hours produces very different elapsed times.
• A step measured in working days has left the account altogether and joined something that runs on its own calendar, with weekends and public holidays simply not counted.
• One step has no unit at all: the time spent deciding. Nothing on the map starts until it ends, and nothing on the map can shorten it.
• The unit a step runs in is stable for years; the number inside it moves constantly. Reading the unit first is what keeps a normal wait from looking like a fault.
• A step that has overrun its usual unit is the only reliable signal that something needs attention rather than patience.

Which clock each step runs on

Minutes, hours, working days: three different kinds of waiting

A step measured in minutes is a step handled by software alone. Nothing about it improves by being watched, and nothing about it depends on the hour of day. When such a step runs long it is almost never busy — it has stopped, and the correct response is to start it once more rather than to wait longer.

A step measured in hours has a person inside it. People work in shifts and in order of arrival, so the same step begun at different hours produces elapsed times that look nothing alike. That spread is not a quality problem; it is what a queue looks like from the inside, and it says nothing about the individual case.

A step measured in working days has left the account and joined something with its own calendar. It runs on business days, ignores weekends and holidays, and reports to nobody on this map. Its one useful property is predictability: begun at the same hour, it lands at roughly the same hour later on.

The waits a broker cannot shorten

Three of the waits on this map belong to the trader. The first is the quality of whatever gets uploaded: a file that is legible on the first attempt is read once, and one that is not is read twice with a round trip in between. The second is the machine — an install and a first connection are bounded by disk, network and whatever else the computer is doing. The third is the decision itself, which has no upper limit and no external cause.

A fourth sits with neither side. An intermediary in the middle of a transfer keeps its own cut-off, its own working calendar and its own queue. It is the single largest source of variance in the whole sequence, and it is worth learning once rather than re-discovering each time.

Sorting waits by owner changes what a delay means. A wait on your own side is an instruction to act. A wait on the broker's side is an instruction to let a normal duration finish. A wait on an intermediary is an instruction to look at a calendar rather than at an account.

Why the unit matters more than the number

The number inside a unit moves constantly. Queues lengthen and shorten, a check that usually takes an afternoon takes a morning, a batch runs early. None of that is worth tracking. The unit, by contrast, barely moves: a step that runs in minutes has run in minutes for years, and a step that runs on a working day has never once run in minutes.

That stability is what makes the unit usable as an alarm. A minutes-step that has taken an hour has failed. An hours-step that has taken an hour has not started yet. The same elapsed time reads as an emergency in one row of the map and as nothing at all in another.

It also settles when to act. Most of the effort spent chasing a step is spent inside the range where the step was always going to finish on its own. The effort that pays is spent on the small number of steps that have already left their range.

Where each clock is actually shown

None of these units are published as promises, and this page does not invent figures for them. What a step is currently doing, and what it is waiting on, is shown where the step lives: the account area for records and checks, the platform for anything to do with an order, and the intermediary's own channel for a transfer in transit.

Orders of magnitude are the honest level of detail. A step described here as taking minutes may take a fraction of one; a step described as taking a working day may finish the same afternoon. The map exists to tell a normal wait from a stalled one, not to predict a finishing time.

Reading a wait in six moves

1. Name the step before naming the delay. Creating a record, having something read, moving money and getting an answer are four different clocks.
2. Find the unit that step normally runs in — minutes, hours or working days — and compare the elapsed time with that rather than with your patience.
3. Decide whose side the wait is on. Yours means act, the broker's means let it run, an intermediary's means check a calendar.
4. Wait out the unit before doing anything. Inside the unit there is nothing to diagnose.
5. If a minutes-step has run long, start it once more. Repetition is the right response to software and the wrong response to a queue.
6. Write down when each step began. Without a start mark none of the above can be applied later.

Durations here are orders of magnitude for reading a wait, not commitments. The current state of any step is shown where that step lives.

What actually changes the length of each wait

Read the table by column: the third column is the only one anybody controls, and the fourth is where most of the effort usually goes.

Related Exness pages