14
Mar
15

The 2015 Verizon PCI Report

A lot has been written about this year’s Verizon PCI Compliance Report particularly about how 80% of organizations cannot maintain their compliance. And at the very end of the report are a number of issues raised by Verizon regarding why maintaining compliance is so difficult for most organizations. It is those issues that I would like to discuss.

Scale and Complexity of Requirements

“I just don’t understand why this ERP upgrade is going to take 18 months to complete. Can’t we just put the DVD in the drive and upgrade it like Microsoft Office?” – Anonymous Executive to IT Management

The same could be said about any security framework. If organizations are struggling with PCI compliance, imagine how they are struggling with HIPAA, FISMA or ISO 27K compliance. Compliance with any of the security frameworks is not easy.

I disagree with Verizon’s claim that it is related to the fact that most organizations do not know the PCI DSS. After six years and three versions, I rarely run into an organization today that does not have a basic, overall understanding of the PCI DSS. These organizations may have some interesting ideas on what sections and requirements of the DSS mean, but they have definitely studied it and read about it. Therefore the idea that organizations are ignorant on the subject is far from the truth in my experience.

In my opinion, where the problem lies is that most organizations have not truly managed their technology environments thanks to interference with mergers and acquisitions, partially implemented applications, bring your own device (BYOD), the Cloud and the plethora of other disruptions that complicate organizations. Today, IT is a very important part of any organization, but it is not managed like it was in the “good old days”. There are too many stakeholders and the consumerization of technology has not helped the situation by making everyone an IT “expert”.

Most organization’s IT operations these days are a hodge-podge of technologies, applications and networks. I would equate it to the technological equivalent of a house’s attic and garage combined. We all know we should clean and straighten them out, but that project always sits on the back burner as there are other, more important or fun things to do.

As a result, for most organizations, there is just no easy way to simplify, segregate and isolate cardholder data (CHD) and comply with the PCI DSS without making the environment even more complex. Starting over is not an option for a lot of organizations.

That said I have encountered a few very brave organizations that have done just that, started over. Management at these organizations came to the realization that fixing the problem was too complex and expensive and that starting over was the cheaper, safer and easier way to go.

Uncertainty about Scope and Impact

“I don’t know much about PCI, but I do know my scope.” – Anonymous Manager to QSA

When application developers cannot explain how their applications work on a technical level. When anyone in any department can be in the IT business. When security personnel are order takers for firewall configuration changes reviewed and approved by management that have no clue as to the implications of those changes. When network people are providing a communications utility for communications traffic but have no idea how that traffic traverses the network.

Is it any wonder we have no idea how to scope a PCI assessment?

But there are larger problems as to why scoping is difficult. The root cause of why scoping is such an issue is that everyone’s risk tolerance is different. I drive race cars at very obscene speeds on race tracks (mostly) that I am sure a lot of people would view as insane. However, I think that people that skydive and do rock climbing are the insane ones. All of this points to everyone’s acceptance and avoidance of risk based on their own views.

There is a sidebar in the Verizon report calling the PCI SSC to provide guidance about scoping. Good luck with that. The Council had a scoping SIG a number of years ago that imploded due to the aforementioned issues with everyone’s risk tolerance. The result was a small band of people from the SIG that published the PCI Open Scoping Toolkit. The PCI Open Scoping Toolkit is not perfect, but it provides a framework to have an intelligent discussion about how to go about scoping and determine what is in-scope and why.

The key to solving the scoping issue resides with the organization, not their QSA, acquiring bank or any other external entity. Organizations need to use the PCI Open Scoping Toolkit to come up with their scoping framework and definitions. Once that has been agreed, then an organization needs to map out their applications and networks to determine their true scope. This is where tools from vendors such as Tufin, FireMon, SolarWinds and the like can provide assistance by documenting the network and then simulating data flows over the network.

With that approach, it is incumbent on QSAs and other auditors to accept these definitions for their assessment unless there is some significant or gross error in the organizations definitions. This will address the complaint that organizations have with QSAs. How often have we heard something such as, “The last QSA told us this was compliant.” If we all play by the same risk definitions the client has provided, then statements like that should go away.

Once an organization truly understands and has defined its scope, it can then understand the impact of existing operations and any changes.

The Compliance Cycle

This is what the Council is attempting to address with business as usual (BAU). The idea is that with security practices and monitoring embedded within an organization’s operations, security issues can be quickly identified and addressed before they become serious.

However, for this to work, organizations need to have their scope known as well has how their IT environment actually works. Without that knowledge, embedding the PCI DSS into the organization is a futile exercise.

Lack of Resources

Every organization is running “lean and mean” these days. Cost control is king. As a result, resources are stretched, sometimes to the point that any additional activities just cannot be accommodated without hiring someone. And hiring is not allowed. So implementing BAU is not going to go well if it goes at all.

On the information security front, finding qualified people is nearly impossible, even for consultancies. Organizations are finding that most information security professionals are heading to consultancies because the pay is better. Since security is hard on both the mind and the body, most people want to be reimbursed as much as possible for their efforts. As a result, most organizations cannot pay for in-house security resources. And then, even if they do ante up, typically the person that takes the position either gets bored once they fix everything, or gets frustrated when the organization refused to make required changes to ensure or enhance security.

Enter the managed security services provider or MSSP. The concept is that the MSSP provides the security talent at a more reasonable price yet organizations get the quality personnel needed to enhance and stabilize their security.

Where this goes wrong is that the MSSP and the customer are not on the same page as to each other’s responsibilities. This is from a mixture of sales people over promising as well as prospective customers hearing what they want to hear. Never mind that it is all documented in a contract.

To address this situation, the PCI SSC has come up with a new requirement, 12.8.5, which states:

“Verify the entity maintains information about which PCI DSS requirements are managed by each service provider, and which are managed by the entity.”

Under the v3 Attestation Of Compliance (AOC) form, this will not be as big a problem for an organization to maintain. However, if an organization has a lot of service providers and/or the service providers have v2 AOCs; this could be a very daunting task.

Lack of Insight in Existing Business Processes

“I’ve only been in this position for [2, 3 or 4] months. So I’m not fully up to speed on everything we do.” – Anonymous Manager to QSA

“I’d give you an organization chart, but it would be out of date by the time I printed it.” – Anonymous Human Resources Manager to QSA

In today’s fast changing business world, people get shuffled out of departments and divisions faster than people can manage the changes. As a result, finding anyone with any sort of insight into an organization’s business processes can be extremely difficult, if not impossible.

Then we go back to my earlier comment about lack of IT management. With the advent of the Cloud, some business divisions and departments have totally sidestepped the formal IT organization and set up their own operations in the Cloud. Did they know what they were doing? No! But that was beside the point, they at least now have IT solutions, never mind if they are secure or implemented properly. The only way to find these rogue operations is to quiz everyone in the organization about how they operate and what they use to operate.

Even then, I have run into situations where a new payment channel pops out of the woodwork at the last moment. Next year’s assessment issue or we will not get the one we are currently doing out the door.

Misplaced Confidence in Existing Information Security Maturity

A lot of organizations that have been doing IT for years and years get caught in this trap. Just because you have been doing IT for an eternity does not mean that you have been doing it right for the same amount of time or that you are doing it correctly now.

In a lot of IT organizations it is an unfortunate fact of life that areas such as special projects, business continuity planning or information security were used as those “safe” places to put the former IT Vice President or Manager out to pasture so they could retire. It did not matter if the individual could handle the job; it was a place to park someone and provide a gentle way out of the organization.

A rare few individuals made the transition and actually took up the challenge of mastering their new responsibilities. However, the vast majority just checked out, collected their pay check and then retired. This left the organization with a very immature security operation compared to the rest of IT’s operations. Add into the mix the changing landscape of IT with business divisions and departments doing their own thing unbeknownst to anyone and you can see how the maturity of information security could be easily misunderstood.

Then along comes the QSA to do the PCI gap analysis and it all comes to a head as the organization comes to the rude awakening that all is not as good as they thought and that significant gaps exist. To add insult to injury, the organization finds that fixing the gaps is going to take a lot longer than the 90 days they had set aside for that activity so that they could get their Report On Compliance (ROC) done in the same year.

The Verizon report is a great read and provides a lot of insights. Everyone should get a copy and read it, take it to heart and address your organization’s security shortcomings.

Advertisement

10 Responses to “The 2015 Verizon PCI Report”


  1. 1 Steve H
    March 16, 2015 at 2:23 PM

    Great article and I agree with pretty much everything. I’d just comment on the idea (I know it is what the standard says) that the merchant defines the scope. If only!! Most merchants dodge this by hiring the QSA to do a scoping exercise for them and then have great difficulty providing basic information like what servers and firewalls they have. As QSAs I think we need to push back harder on this.

    • March 16, 2015 at 3:25 PM

      To be fair though, most merchants and even a lot of service providers have no idea how to scope because it is deemed to be some sort of “Black Art”.

  2. March 16, 2015 at 11:51 AM

    Wow! Just wow! Nodding my head in agreement as I read this.

  3. March 15, 2015 at 12:21 AM

    Great post! Technology change is slow in large-scale, multi-national organisations. Therefore, there is a great difficulty in keeping up with PCI changes which happen quite frequently. Also, the budget is controlled by the business.

    Some of the wording of the PCI requirements can cause debate. E.g. What is the difference between truncation of PAN’s vs masking? Some would argue that there is a difference, others argue that there is none. I am of the view that they are different. What about you?

    • March 15, 2015 at 6:47 AM

      PCI and all security frameworks change as a result of changes in attacker tactics. Therefore you are implying that large organizations are unable to secure their environment because of an inability to change. That is a very sad state of affairs.

      Masking is when X’s, zeroes or asterisks for example are used to replace digits in the PAN to meet the first six, last four rule (i.e., a masked PAN is still 15 to 16 characters long). Truncation is when only the first six, last four digits (i.e., a truncated PAN is either four, six or ten characters in length) of the PAN are stored/displayed without any padding characters.

      • March 15, 2015 at 6:54 AM

        Do you know of many large organisations that do it well? What do they do differently do you think?

      • March 15, 2015 at 7:46 AM

        Yes, there are a few large organizations that do it well, but they are also very strict on the use of IT because they have to be for regulatory as well as legal reasons. It seems that unless organizations have a literal “gun” put to their head (i.e., legal and/or regulatory reasons), there are always excuses for why security is slipshod or non-existent. Even then, I know of organizations that just hope they do not get caught for their bad security practices.

      • 8 R-
        March 15, 2015 at 11:35 AM

        I have always considered the difference to be that truncation is a permanent removal of the middle digits, whether they are then replaced by other characters or not. Masking is a temporary change that takes place during the display of the PAN, but with the underlying full PAN still present. (A masked PAN can be unmasked, a truncated PAN cannot be un-truncated.)

      • March 16, 2015 at 5:06 AM

        When in doubt, go to the PCI DSS Glossary.

        According to the Glossary, Masking is defined as:

        “In the context of PCI DSS, it is a method of concealing a segment of data when displayed or printed. Masking is used when there is no business requirement to view the entire PAN. Masking relates to protection of PAN when displayed or printed. See Truncation for protection of PAN when stored in files, databases, etc.”

        Truncation is defined as:

        “Method of rendering the full PAN unreadable by permanently removing a segment of PAN data. Truncation relates to protection of PAN when stored in files, databases, etc. See Masking for protection of PAN when displayed on screens, paper receipts, etc.”

        Knowing how POS systems mask the PAN, there really is no underlying full PAN to go back to.

      • 10 R-
        March 16, 2015 at 9:19 AM

        The FAQ on the matter (FAQ 1146) is also worth referring to:

        “Note that even if a PAN is masked when displayed, the full PAN might still be electronically stored and would need to be protected in accordance with PCI DSS Requirement 3.4.”

        Notice that the glossary says “concealing” rather than “removing”.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


Welcome to the PCI Guru blog. The PCI Guru reserves the right to censor comments as they see fit. Sales people beware! This is not a place to push your goods and services.

March 2015
M T W T F S S
 1
2345678
9101112131415
16171819202122
23242526272829
3031  


%d bloggers like this: