Segregation of duties: no one person from start to finish
Segregation of duties splits a transaction so that committing fraud or hiding an error requires collusion. Most breaches of it are accidental, and invisible until tested.
IT general controls decide whether you can rely on anything a system tells you. Weak ITGC does not fail on its own — it invalidates the controls above it.
IT general controls are the controls over the environment in which financial systems run: who can get into them, how changes reach them, how they are operated, and how the data behind them is protected. They are called general because they are not specific to any one transaction — they underpin every automated control and every report the system produces.
That is also why they matter more than their unglamorous name suggests. If anyone can change the code, or grant themselves rights, or run an unlogged data fix, then no automated control in that system can be relied upon and no report from it is evidence of anything. Weak ITGC does not produce one finding; it invalidates a whole layer of controls above it.
Access is where most findings land, and within access, privileged accounts are where the serious ones land. A developer with standing production access defeats the change control domain entirely, no matter how well the change process is documented.
The single most common cause of an ITGC failure is not a missing control but an incomplete population. If you cannot demonstrate that the user list or the change list is complete, samples drawn from it prove nothing — and the tester is required to treat the control as untested rather than effective.
When the accounting system is somebody else's cloud service, the change and operations domains largely move to the provider, and their assurance report becomes part of your evidence. What does not move is access: you still decide who has rights in that system, you still onboard and offboard, and you still have to show that this was reviewed.
Read the provider’s report properly rather than filing it. It will list complementary user entity controls — the things the provider assumes you are doing. Those are your controls, and testers check them.
Because access is the domain that stays with you, the roles have to be explicit rather than conventional. Ettex Teams keeps permissions per workspace and per role, so producing the user list and evidencing what each role can do is a report rather than an investigation — and the segregation of duties analysis runs against the same list. Change and operations for the platform itself remain the provider’s side of the line, which is exactly what the assurance report is for.
Application controls operate inside a process — a three-way match, an approval limit, a validation rule. ITGC govern the environment those controls run in. Application controls can only be relied on where the relevant ITGC are effective.
No. They appear in financial audits generally, in SOC 2 and ISO 27001 work, and in most customer security questionnaires. SOX is simply where they are tested most formally.
Those that produce or hold data relevant to the financial statements, including reporting tools and material spreadsheets fed by them. Scope is a documented judgement, not a list of everything IT runs.
Typically logging every action performed with the privileged account and an independent review of that log. It is weaker than removing the access and testers treat it accordingly.
Segregation of duties splits a transaction so that committing fraud or hiding an error requires collusion. Most breaches of it are accidental, and invisible until tested.
A remote joiner misses everything that is normally absorbed rather than taught — who to ask, how things really work, whether their pace is right. All of it has to become explicit or the first month is guesswork.
Most hybrid policies say three days in the office and stop. The questions that cause the arguments — which days, who decides, what happens to meetings — are the ones left out.