For all the right reasons, your company has been thinking about deploying SIEM…to create an alert system when those with less than good intentions come knocking; to remediate potential network threats; to comply with federal, state or industry regulations; and identify the risks and vulnerabilities throughout the enterprise IT infrastructure and architecture. If you maintain even a modest (SMB -> Fortune 1000) organization that has any online identity, SIEM should be the cornerstone of your asset protection strategy.

First and foremost, SIEM (and to a certain extent log management) is about visibility. Who is doing what and when on your network. It is as much about understanding the holistic landscape of your infrastructure as it is protecting proprietary assets. Without it, it’s akin to coaching the Big Game without any idea who is the opponent; or for that matter if you even have a starting left guard.

But fun metaphors aside, SIEM is a critical enterprise tool. And just like any enterprise solution, it requires forethought, vigilance and most importantly, a good game plan. And when deployed properly it can change your IT department from infrastructure-based, to information-centric. And as such you get to make better decisions, faster.

And with every technology there are best practices and pitfalls. In past articles I have spoken at length regarding the advantages of deploying and managing SIEM in the cloud. Many of these surround the affordability, manageability, control and capability of the solution. For many, security from the cloud is still an emerging concept. But for those who’ve already made the leap, they are reaping the significant benefits. But I want to move beyond the arguments of “going cloud” when deciding on security solutions. Today I want to focus on what happens next. How do you start collecting that ROI once a cloud-based security-as-a-service has been chosen?

The reason most enterprise deployments fail (on premise or cloud) can be typically traced to two causes: (1.) Lack of buy-in from the executive level or employee resistance to change, but more often the culprit is (2.) lack of vision or process. Too many companies jump in and apply a solution because they heard it was important or were sold a Porsche when all they needed was a family SUV. Of course one of the benefits of cloud-based security is the ability to "buy" the SUV and instantly scale up to that Porsche, if and when, the business need requires it (without touching CapEx budgets!)! But with that here are 8 best practices you should implement when moving forward with your cloud-based security initiative:

Best Practice #1: Identify your goals and match your scope to them. There are five questions you need to ask before moving forward with any deployment. 1. WHY do you need SIEM (compliance? user and/or partner expansion? BYOD? Breach detection?) HOW will SIEM be deployed to properly address these issues (what processes, functionality and capabilities are needed; which needs to be outsourced/replaced/improved) WHAT needs to be collected, analyzed and reported? HOW BIG does the deployment need to scale to accurately and cost effectively meet your specific business need? And WHERE is the information situated that should/must be monitored?

Best practice #2: Incremental usage. The quickest route to success is taking baby steps. The idea is to prove the concept and then expand the scope. To some this might be to start with log management and add  SIEM once you understand the requirements, commitment and volume. Now because security-as-a-service is so flexible and can ramp up or down instantly, an easy entry point might be to start with only those elements that fulfill compliance. The project might be overwhelming, but if you take it in bite-sized phases, you will find the victories come easier and the ROI is justified. When dealing with a cloud security deployment, it is easy to turn on the fire hose when only a garden hose is needed. But the beauty of a cloud deployment is the ease and flexibility of scaling. Again, another example of incremental usage would be either to apply SIEM against specific use case scenarios or possibly just migrate a division or a department or a function (as opposed to the entire enterprise).

Best Practice #3: Determine what IS and ISN’T a threat to your network. Returning to the fire hose metaphor, when deploying a SIEM initiative, it is very easy to get lost in a sea of data. It can be like trying to drink from that proverbial fire hose. The trick is to recognize what constitutes a true risk and eliminate false positives. And this requires some internal analysis to create a series of rules that sift out the white noise and differentiate “normal” traffic from suspicious activity. For instance, if there is an attempted access to your partner portal from Russia—is that normal? Do you even have a partner in Minsk? But even a simple filter isn’t quite enough. Risk is three dimensional and it can hide in plain sight. That’s why you continue to filter based on time of day, IP address, server, attempts, network availability and a myriad of other forensic qualifiers before the alert is grave enough to require immediate attention.

Best practice #4: Map response plans. Now that an incident gets your attention, what do you do? Do you launch an account investigation, suspend the user, deactivate a password, apply a  denial-of-service against the IP or a number of remediations based on the severity, vulnerability and identity of the transgressor. This goes back to workflow and process. Who is going to what to whom and how? SIEM is a process-reliant technology. You simply can’t flip a switch and say you’ve put up a magic forcefield around your network. Your response plan is your blueprint to closing the vulnerability gaps and ensuring compliance.

Best practice #5 Correlate data from multiple sources. The practice of situational awareness is what adds the muscle into a SIEM initiative. Like #4, it isn’t enough to plug in a solution and press “go.” Situational awareness takes into account a multitude of different endpoints, servers, data streams, assets and inventories, events and flows, from across the enterprise and puts information into context.  Context is the most important portion of risk assessment.  For example, a shark is a threat. However if that shark is 10 miles away, it is not a direct or immediate threat. Doesn't mean you're not vulnerable if that shark gets hungry. Having an engine that not only creates accurate perspective, but analyzes, understands and acts upon behaviors is key. And to do that a centralized SIEM engine needs the data from more than just a single source or single server.

Best Practice #6: Requires Real time monitoring 7/24/365. For many companies this is a challenge, but hackers don’t sleep. And although a great deal of SIEM and Log Management is automated, it still requires the vigilance of 24 hour monitoring. Trees might be falling in the forest, but if there is no one to see them, breaches occur, networks are compromised. I’ve witnessed plenty of IT departments that don’t have the resources. Again, this is a considerable advantage that security-as-s-service provides and allows you to sleep just a little better at night. Knowing that this one crucial element of your security is professionally addressed without additional staff or budget makes the cloud that much more valuable.

Best Practice #7 Remain calm! One thing we’ve noticed is that soon after the deployment of a SIEM/Log Management it seems there are alerts and issues you never dreamed about. Things are bound to look worse before they get better and it can seem overwhelming; kind of opening a Pandora’s Box of malware and botnets.  For the most part it is because you now know what you didn’t know before.  In some respect it is like looking at your hotel room comforter under black light and a microscope. But once you realize what you’re looking at and that much or the remediation can be automated, soon, (with a bit of fine tuning and normalizing correlation feeds) you will be measure that the anomalous events lessen and the alert prioritizations allow you to make timely and intelligent decisions.

Best practice #8: Evolution. Security is a moving target. You need to revisit you processes and workflows every few months to make sure you are up to date with compliance requirements, new users/access points and expanded or redefined workflows. This is more than recognizing the latest virus threats. New users access your network with regularity. New layers of regulations are added. There are new applications requiring monitoring. All in all, by giving your cloud-based SIEM and log management solutions the new and necessary data, your enterprise will be more secure than it was yesterday.