Route — Demo account as a line test
The Exness demo account measures the connection, not the profit
A practice account is created in the Personal Area after registration on the official exness.com, needs no deposit and trades virtual funds at real market prices. The money in it is imaginary; the traffic it moves is not. It streams the same quotes, runs the same terminal build and recovers from a dropped link exactly the way a funded account does — which makes it the only place where a thin, unstable or metered connection can be put under load with nothing at stake.
What is identical, and what is not
The money is virtual; the line is not
Nothing about a practice account is a reduced version of the real thing on the network side. That is the whole basis of the exercise below, so it is worth being specific about it.
The same quote stream
A demo receives the same market prices a real account receives, from the same servers, at the same rate. Watching a chart on a demo asks the connection for the same work it would ask for on a funded account.
The same software
MT4, MT5, the Web Terminal and the Exness Trade app are not different builds for practice accounts. The client that will carry real orders is the client being tested.
The same recovery behaviour
Reconnection after a break, the refilling of missing chart history, the state the terminal comes back in — all of it happens on a demo exactly as it will happen later, because none of it depends on what kind of funds are in the account.
One difference, and it is the useful one
A failure costs nothing. An order that never arrives, a session that dies halfway, a reconnect that takes far too long — on a practice account these are readings, not losses.
Deliberate failure
The experiment nobody runs on a funded account
The single most informative thing a poor line can be asked is what happens when it dies in the middle of an instruction. On real money that question is answered by accident, at the worst possible time. On a demo it is answered on purpose, in four steps, as often as required.
-
Fill in an order and stop short of confirming
Instrument chosen, volume set, the ticket ready to go. Nothing has left the device yet, which is the state the test starts from.
-
Cut the signal, then send
Switch the device to flight mode or walk out of coverage on purpose, and press the button. The terminal now has an instruction it cannot deliver, and how it says so is the first reading of the test.
-
Bring the connection back and time it
Count the seconds from restoring the signal to the moment the terminal is genuinely usable again — connected, quotes moving, the position list trustworthy. That number is worth more than any advertised speed.
-
Read what actually survived
Check the position list and the journal against what was intended. An instruction that never reached the server was never an order; one the server had already accepted stayed with the server through the outage. Knowing which of the two this connection produces is the point of the whole exercise.
Repeat it a few times rather than once, because a bad line is not consistently bad — it is occasionally bad, and one attempt cannot tell the difference.
What to write down
Four numbers worth taking from a practice run
| Reading | Where it comes from | What it decides later |
|---|---|---|
| Data per hour | The device's own data counter, reset before the session and read at the end of it | Whether a terminal can be left running all session or should be opened only to act. |
| Seconds to reconnect | Timed by hand from signal restored to terminal usable | How long a gap has to be tolerated before a position is left unattended by force. |
| Fate of an instruction sent into a gap | The position list and the journal after the outage | Whether an unanswered button press can be safely pressed again, or has to be checked first. |
| Whether this network reaches the server at all | The terminal's connection status on the network in question | Some networks are slow and some simply do not complete the connection. This is the reading that separates them. |
Swipe the table sideways →
Take the four readings once per route rather than once in total — a browser tab, an installed desktop terminal and a mobile app do not spend a connection the same way, and the routes themselves are laid out on the downloads page and on the platform board.
What counts as an outcome
Reading the result of the exercise
The output is the timings and the counter
A completed practice session leaves two useful artefacts: a set of timings and a figure on the data counter. Those carry over to the funded route unchanged, because they describe the network and the software rather than the account.
They are also the only part of the exercise that is repeatable. Run it again next month on the same route and the numbers can be compared honestly.
The virtual balance is not a result
- It is not a track record. A demo proves that the platform and the connection behave as expected; it does not promise the same outcome on real money.
- It excludes the two things that matter most later. Live trading adds slippage — a fill at a different price than the one quoted — and the weight of a real balance behind every decision.
- It rewards the wrong behaviour during a line test. Deliberately breaking a connection mid-order is bad for a balance and good for the readings, and the readings are what the session was for.
Trading is risky and may not be suitable for everyone.
Timing the test
When to run it, and for how long
At the worst hour, not the quietest one
A network is at its kindest when nobody else is using it. Testing late at night produces flattering numbers that describe conditions the trading session will never take place in. Pick the hour the connection is normally busiest.
One whole session, not a few minutes
Data consumption is only meaningful over a stretch, and the interesting failures — a hand-off between networks, a drop and recovery — are the ones that arrive when they choose to, not when the test is watching.
On the connection actually being paid for
Testing on somebody else's fast link measures somebody else's link. The reading is only worth writing down when it comes from the network and the plan that will carry the funded account.
All of this happens before a deposit exists, which is the practical argument for doing it at all: the route can be rejected on evidence while rejecting it is still free. Standard and Standard Cent accounts have no minimum deposit when the funded step does arrive, and the five account types are compared on the account types page.
Quick answers
Questions about bandwidth, reconnects and quotas
Does a practice account use less data than a funded one?
No. The quote stream, the chart history and the reconnect traffic are identical — the account type changes what the numbers on screen mean, not how many bytes are needed to show them. That equality is exactly what makes the measurement transferable.
Does an open terminal keep spending the connection when nothing is being traded?
Yes. A terminal that is signed in and showing a chart is receiving prices continuously, and that is what the data counter picks up. The practice session is where the size of that background cost gets established, so the decision to stay connected or to open the terminal only when acting can be made on a number.
If the connection dies while an order is being sent, is the order placed?
It depends on how far the instruction travelled, and that is the reading the deliberate test produces. An instruction that never reached the server does not exist; one already accepted by the server stays with it and appears in the position list once the terminal reconnects. Checking the list after a break, before pressing anything again, is the habit the demo is there to build.
Can the same test compare two routes?
Yes, and that is the strongest use of it. Repeat the identical sequence on an installed terminal, in the Web Terminal and in the Exness Trade app on the same connection, and the four readings become directly comparable. A browser tab is not automatically the lighter option, which is a conclusion worth reaching from measurements rather than assumptions.
Do the prices on a demo differ from real ones on a poor connection?
The prices are the same market prices a real account receives. What a weak line changes is how promptly they arrive on screen, and that lag is part of what the exercise is measuring.
Do the readings stay valid once real money is involved?
The network readings do — they describe the connection and the software. What they cannot describe is execution on a funded account, where slippage and the pressure of a real balance both appear. The timings tell what the line can support; they say nothing about what should be traded on it.
Test the route before it carries anything
Registration on the official Exness website takes an email and a short questionnaire, and a practice account can be created straight afterwards — no deposit, virtual funds, real market prices, and a connection that can be pushed until it breaks without anything being at risk.
Open Exness AccountOne click on the button is itself the lightest request on this page — it forwards to exness.com over this site's partner route, and the account is created on the broker's side. What that first page asks for is described in the registration guide.