Having good security will create opportunities for an attacker to leave evidence behind.
If businesses are not looking for that evidence regularly within their networks, there is a risk that the intruder will go undetected, according to Trend Micro strategic markets VP, Blake Sutherland.
Learn how smart CIOs are protecting customers from security breaches
To back up his point, Sutherland points to a Verizon study that found it may take hours or days to breach a network, but weeks, months and years to determine someone whether someone is already in and "exfiltrating" data.
"In the past, if my security was better than my neighbours', then my neighbour is more likely to get attacked," he said.
"But if someone truly wanted something from me, if they have the patience and resources, they will find a way in."
It has reached the point where if a business has something of value, cyber criminals "can and will go in."
The question then is whether the controls that provide visibility quickly and determine whoever is attacking the network are in place.
"In a lot of cases we have talked about putting on suits of armour but not having a visor," Sutherland said.
He added that approach up to now has been about trying to stop the attack but not investing in figuring out who is doing it.
"A lot of these new technologies are trying to give you far more information but maybe in the past it was about keeping the attacker away," Sutherland said.
"Now we can find out more about attackers themselves."
Not that advanced
Email based attacks are getting more sophisticated, and Sutherland has seen a lot of developments from the social engineering perspective in particular.
"There is far more information in the social network now about people, allowing cyber criminals to come up with detailed emails," he said.
Of note, Sutherland said 90 per cent of targeted attacks start with a spearphishing attempt.
"It is far easier to get the information to craft that attack," he said.
Information gathering, which the security used to call recon, is step one, and Sutherland said this followed by looking for the first point of entry, usually an indirect one.
As an example, Sutherland said that a cyber criminal might go after a translator as means to access a law firm.
"The attack is designed to be social, sophisticated and stealthy," he said.
Cyber criminals are persistent these days, which has given birth to the term advanced persistant threats (APT).
However, Sutherland is wary of the term APT, particulary when it comes to the "advanced" aspect.
"They are persistent threats, but they are only as advanced as they needs to be," he said.
So if somebody knows a target is running Windows 2000 at home, they do not have to be that advanced to break into the system.
"Not every targeted attack is using five zero day flaws to break in," Sutherland said.
Because the attacks are being customised in this way, Sutherland said Trend Micro talks now about "custom defence."
"For anything that is targeted directly at you, you have the have the tools to respond to this sort of custom attack with custom defence," he said.
Patrick Budmar covers consumer and enterprise technology breaking news for IDG Communications. Follow Patrick on Twitter at @patrick_budmar.