The failure people expect from a copier is a wrong trade. The failure that actually costs them is no trade at all, for a fortnight, without a single alert. A terminal that quietly logged out places no orders and reports no errors, and from the trader's chair a dead connection and a quiet week look identical. So the question monitoring has to answer is not whether the last order succeeded. It is whether the thing is still alive.
The three states worth checking weekly
Connection first. Each connected trading account has its own MT4 or MT5 terminal running on our side, logged in with the credentials you supplied, and that login has a real state: connected, refused, or dropped because something changed at the broker. If it shows failed, the signals that arrived during that window are recorded as not placed rather than passed over in silence. That record is the entire reason the state is exposed instead of hidden behind a green tick.
Second, the Telegram side. The copier reads channels through your own Telegram account, so anything that ends that session ends the reading. Leaving a channel counts. Being removed from a paid group when a subscription lapses counts, and it is the version people miss, because the trades stop for a billing reason that has nothing to do with software.
Third, the account itself. The terminal reports the balance, the currency and the open positions back from the platform rather than assuming them from what was sent. If the balance shown does not match what your broker's own portal says, stop and find out why before the next signal arrives.
Refusals are the most useful record you have
Fills tell you what happened. Refusals tell you what your configuration is doing to you, and each one carries its reason. The four you will see most:
| Recorded reason | What it means | What to change |
|---|---|---|
| Spread above your limit | The market was wider than the guard when the order was due | Nothing, usually. Around a scheduled release this is the guard earning its keep |
| Open position cap reached | Several channels fired near each other and the account hit your ceiling | Check whether the cap is too tight or the channels are too correlated |
| Symbol not on the account | Your broker does not list that instrument under any name the matcher found | Add a rename rule, or accept the account cannot trade it |
| Insufficient free margin | The size the rule produced does not fit the account | Sizing, or the exposure already open |
Read them monthly, in a batch. One refusal is noise. Fourteen refusals all citing the spread limit on the same instrument means your guard is set against an instrument that is simply wider on your account than you assumed, and you have been excluded from a whole class of setups rather than protected from it. That is a configuration finding you would never reach by looking at your fills.
Separating the four ways a trade can be missing
When you notice a signal in the channel that never reached your account, there are four candidates and only one of them is execution. The post did not parse cleanly, so nothing was produced and it was recorded and left alone. The symbol did not exist. A gate refused it. Or the broker rejected the order and said why. These are logged separately on purpose, because they lead to four completely different fixes and people otherwise merge them all into a general feeling that the copier is unreliable. The full path from post to filled order is set out in inside a copier, and it is worth reading once so you know which segment you are looking at.
A refusal is not a fault. A copier that never refuses anything has no guards switched on, which is a worse state than one that declines a trade every few weeks with a reason attached.
What to compare at the end of the month
Trade count before anything else. Count what the channel published and count what your account took. If the two numbers agree, any disappointment is the channel's, and no amount of copier tuning will fix it. If they disagree, the gap has reasons attached and you can go read them.
Then percentages, never money and never lot sizes. Two accounts running identical instructions at different balances show wildly different figures in currency and nearly identical figures in percentage terms, which is the whole argument in scaling lots between accounts. Any gap in the percentages is the thing to investigate.
Then entries against exits. If your fills sit consistently worse than the levels the provider published but your exits match, you have a timing problem at entry, which is usually the provider posting after price has already moved rather than anything mechanical. The segments that actually contribute delay are broken down in what copier latency actually looks like.
Keep a record the software does not own
The copier records what it did. It does not record what you thought at the time, and after eight weeks you will not remember why you doubled the risk on one channel and demoted another. A short note per configuration change, with the date, converts a vague sense that things got worse into a comparison you can actually run. The habit is the same one described in keeping a trading journal, applied to settings rather than trades.
One rule makes those notes usable: change one thing at a time. A month in which you adjusted sizing, target handling and the trailing stop together tells you nothing about which change mattered, and a month does not contain enough trades to separate three variables anyway.
The first fortnight is different
Early on, watch the individual trades rather than the aggregate. Did the first copied order arrive on the instrument you expected, at a size you recognise, with the stop the channel posted? If the size looks wrong, it is almost always the sizing rule rather than a bug, and the choices are laid out in risk settings for copied trades. If the instrument looks wrong, check the symbol list your broker publishes on that specific account, because raw and standard accounts at the same broker carry different suffixes.
After that, drop to a weekly connection check and a monthly review. Monitoring a copier is not a screen you sit in front of. It is four questions asked on a schedule, and the most important of them is whether silence means nothing happened or nothing was working. Current product detail is at copysignals.io.
Leveraged trading carries a high risk of loss. Monitoring tells you what the software did. It makes no signal a good one, and SINGUARD takes no view on any channel a customer follows.
"The dangerous state is not a bad trade, it is a week of no trades that you read as a quiet market. Absence has to be something you can detect, or you find out at the end of the month."
— Roman Onta, Executive Director, SINGUARD
Key Takeaways
- A logged out terminal and a silent channel look identical from the outside, so connection state is the first thing to check, not the last.
- Every refused trade is recorded with its reason, and reading a month of refusals says more about your configuration than reading your fills.
- A missing trade is one of four separate things: an unclean parse, a symbol that does not exist, a gate refusal, or a broker rejection.
- Compare percentage return and percentage drawdown across accounts, never lot sizes or currency amounts, and change one setting at a time.
Frequently Asked Questions
How do I know the copier is still connected?
Each connected account shows the state of its own terminal, and the Telegram reading side shows whether your session is live. Check both weekly, because a dropped session produces no errors and no orders, which is indistinguishable from a quiet market.
Why does the log show trades that were never placed?
Those are refusals, and each carries the reason: a spread wider than your guard, the open position cap, an instrument missing from your broker's list, or insufficient free margin. They are recorded so a gap in your statement has an explanation.
What should I review at the end of a month?
Count the channel's published signals against your account's trades, read the refusals and their reasons, then compare percentage return and drawdown rather than money. Investigate any gap in the percentages before changing settings.
About the Author
Roman Onta is an Executive Director at SINGUARD. He builds the Prop Firm CRM, the Broker CRM, Scalegram and CopySignals side by side with his brother Alex Onta, and he helped on the design of eTrader, the division Alex built and leads. His ground is worldwide payment processing, AML compliance and the corporate structures brokers are built on, work the two of them carry together, shaped by executive roles in the UAE and international corporates. He lives and works in Dubai for most of the year. Meet the executive duo leading Singuard's five divisions.