Rational Risk Management Ends The Fear Factor


  • Amrit has an article up now in which he christens the term “Fear Factor Marketing”, which I find hilarious. From the article:

    The screaming headlines have been running for years. Whether they’re in press releases about cybercrime exceeding international drug profits or the billions of dollars lost to breach disclosures or videos highlighting the meltdown of power generators due to a myriad of vulnerabilities, the anti-malware industry has long relied on fear to move their products…

    Amrit then goes on to say that he believes that the commodification of security features will eventually force less reliance on FUD and lead to a stronger bargaining position for those on the buying side of the table. While I agree (as long as the rate of innovation in the market remains at the current rate or lower), I do think there is a more effective method you can use to remove emotion from the buying equation right now.

    WHY YOU BUY THINGS

    It’s common knowledge that people buy on emotion, and then provide rational justification to reinforce the emotion. It’s like this in real life for me. I don’t need to upgrade my computers to Leopard or Vista, I’ve been doing my work on Tiger and XP just fine for years now. Why upgrade? Because ITS FREAKIN COOL, that’s why. I’m upgrading because of emotion. Increased security or productivity features or whatever are simply justifications.

    The InfoSec market has been playing the same game now for ages. Things are sold to you because you fear getting hacked, or being “non-compliant” with some paternalistic prescription of your purchasing budget, or because that’s what Smart Pundit(tm) says you should be buying.

    ON AVOIDING EMOTION AS A BUYING MOTIVATOR

    The key to avoiding buying based on emotion is to have a rational approach. Easy enough. But unfortunately we’ve been trained to buy on emotion, and then justify that emotional response with some perverted logic. So the next time someone is trying to scare you into opening your checkbook - how do we know if we’re just perverting our logic to buy based on feeling or not?

    The answer is to use a system or framework that can be objectively applied, regardless of vendor or technology, to examine the worth of a product or technology to you. Not the obligatory case study, not to protect from some projected aggregate industry loss amount or increase in attack numbers, but worth to your specific environment.

    USING A RATIONAL APPROACH

    You’ll remember that the only metrics we’re concerned with are:

    • Reducing Risk
    • Reducing Loss
    • Creating Operational Efficiencies

    Now for purchasing a product, Loss Reduction metrics don’t apply (Loss Reduction is proven when we compare loss in a current incident with a past similar incident). But the other two are key.

    OPERATIONAL EFFICIENCIES

    Will the product allow you to do the same or more with less over its lifetime? Then buy it!

    RISK REDUCTION

    Ah, the key. You’ll recall that one of the arguments made against “risk management” is that you’re going to just buy the same things as everybody else anyway. Ummm, if you have an unlimited budget and are lazy, maybe.

    If you choose not to just buy what’s thrust upon you from the floor of the last trade show, you can match the probable reduction in loss event frequency and impact with your risk tolerance, and see if the new technology really makes sense given resource limitations (budget/human costs of aquisition). You can even go so far as to compare how a product will reduce the frequency or impact of probable loss versus the status quo, similar products, or alternative mitigation strategies - you’ll have a rational reasons to buy or not buy.

    Remember, it’s not just about risk reduction qualities - it’s about how much risk reduction compared to expenditure, and your tolerance for risk.

    Risk can be used to determine if an investment is worth the time and effort in reducing risk.

    THE NAMES HAVE BEEN CHANGED TO PROTECT THE INNOCENT

    A few years back when DLP products were first on the market, they were very, very expensive. And all the rich kids with Fortune 500 budgets were buying them. Part of this success was because the vendors had a really great sales trick. They would put the box down in your environment for a month, and then at the end of the trial period, they would show you all the confidential information that was going across your wire. Common Practices and paranoia dictate that this information going out unencrypted was a bad, bad thing. In fact, legend has it that a particular vendor had never lost a sale using this trick.

    Until they got to a risk aware (FAIR using) organization. You see, this organization had plenty of visibility and compliance concerns. And like all the other companies before them, they had lots of Data Leakage to see at the end of their 30 day trial. But something didn’t add up right:

    If they were leaking all this data, where were all the incidents?

    You see, they understood that frequency was part of the equation. And they hadn’t removed any controls recently, so this leakage had to have been happening for years. So where was all the risk? The answer, as it turns out, is that the risk just wasn’t as great as you would think it would be, and certainly not enough to warrant the price tag the DLP vendor expected (and had received from dozens of similar organizations).

    So the vendor finally lost a sale.

    Now this was back when these things costs millions of dollars to large organizations. Since then, market pressures have dropped the price to something more reasonable, (1/10th of the cost) and the organization finally bought one (price, tolerance, and risk now make sense). But the point is that the CISO didn’t pull the emergency stop chord, expend massive amounts of budget and political resources, and buy herself a multi-milllion dollar DLP installation just because of FUD, or peer pressure.

    Fear didn’t work. Risk did.

    So take heart Amrit, the End of the Fear Factor is in sight, and it’s available now if you are willing to use it.

    Posted on

  • 14 comments

    1. Spock Nov 13

      I upgraded to Leopard specifically because of Time Machine. You irrational types make it hard for the rest of us.

    2. Jon Robinson Nov 13

      Brilliant. Well done. “How to Buy”

    3. Adam Nov 13

      “If they were leaking all this data, where were all the incidents?”

      The incidents are in the same place as the leaks: invisible before we got tools to look for them. Why would any ID theft victim call their hospital to report an id theft? (to grab a random incident from the front page of pogowasright.org)

      Adam

    4. Alex Nov 13

      Hi Adam,

      Are you suggesting a 1 to 1 correlation between the DLP data and frequency of incidents?

      If not, then the question to be answered remains: “To what extent does DLP in place allow you to prevent, detect, or respond”, and “Is that amount of additional control worth the investment”.

      This then becomes a risk question.

    5. Adam Nov 14

      I’m not suggesting correlation, I’m suggesting that the question, “if we’re leaking, where are the incidents?” is a poor question because incidents are missed, covered up, hidden, and otherwise unobservable.

      It’s a little like asking, “if this radioactivity stuff exists, where are the gamma rays” before the invention of x-ray film.

    6. Alex Nov 14

      I should write, then, that this IRM department was not naive enough to believe that they were without incident. The probability of current incident is (was) something to account for.

      However, when we use the word “probability” we stumble back into the “risk realm” and then must quickly address impact. For this organization, the impact - and past aggregate impact from incidents, were no where near the multi-million dollar price tag for the DLP solution they were pitched.

    7. Chris Hayes Nov 14

      There does not have to be an “incident” that results in a loss either. I think what compounds this concept is regulatory or industry compliance – especially those that rely on “self-reporting”. What is the appropriate cost to prove due diligence in absence of quantifiable loss events. Better yet, what is the cost of an organization being considered a responsible corporate citizen and trying to prevent (and manage) vs. reacting when there is a loss event because of ignoring risk.

    8. Adam Nov 15

      Color me a little skeptical (although obviously you have more details than I do).

      I’ve seen organizations with mature processes designed to cope with a lack of information. I’ve seen those organizations taken aback when they added new detection mechanisms.

      Overall, though, I agree. We need much better data about what goes wrong in order to make better decisions.

    9. Alex Nov 15

      I’d rather you be skeptical than swallow everything the blogosphere puts out.

    10. Sammy Nov 17

      Hi Alex,

      I’m glad I could help you set up your “my product is better” anecdote, even if you had to grossly misinterpret my words as arguing “against” risk management and fabricate the idea that I’m calling CISOs “stupid” in order to fit your needs. And this after you accused me of name calling in your previous post. Wow. Good times.

      –Sammy.

    11. Alex Nov 17

      Sammy,

      You’re completely right. I apologize for taking your quote out of context, and for my comments. I’ll edit those out.

    12. How to Get Six Pack Fast Apr 15

      The topic is quite trendy in the net right now. What do you pay the most attention to when choosing what to write ?

    1. Liquidmatrix Security Digest » Security Briefing: November 14th
    2. Not Bad For a Cubicle » Blog Archive » Too much of a good thing

    Leave a reply