As IT professionals who have a special interest in monitoring (because why ELSE would you be here on THWACK, except for maybe the snuggies and socks), we understand why monitoring - and more importantly, automation in response to monitoring events - is intrinsically useful and valuable to an organization. We understand that automation creates efficiencies, saves effort, creates consistency, and most of all, saves money in all it's forms - labor, downtime, lost opportunity, and actual cost.
To us, the benefits of automation are intuitively obvious.
And that's a problem. When I was in university and a professor used the phrase "it's intuitively obvious" that meant one thing: run like hell out of the class because:
A) the professor didn't understand it themselves, and
B) it was going to be on the exam
The idea that things that we see as intuitive are the same things we find either difficult or unnecessary to explain was driven home to me the other day when I solicited examples of the measurable business value of monitoring automation.
"I’m putting together some material that digs into the benefit of automated responses in monitoring and alerting. I’m looking for quotable anecdotes of environments where this has been done. What’s most important are the numbers, including the reduction in tickets, the improvement in response, etc.
A certain company “enjoyed” an average of 327 disk full alerts each and every month, which required roughly 20 minutes of staff time to investigate. After implementing a simple vbscript to clear the TEMP directory before opening a ticket, the number of alerts dropped to 57 per month. Now each of those tickets required one hour to resolve, but that’s because they were REAL events that required action, versus the spurious (and largely ignored) quantity before automation was introduced.)
If you have any stories like this, I’d love to hear it. Feel free to pass this along to other colleagues as you see fit."
I received back some incredible examples of automation - actual scripts and recipes, but that's not what I wanted. I wanted anecdotes about the results. So I tried again.
What I am looking for is more like examples that speak to how monitoring can justify its own existence to the bean counters who don’t care if technology actually works as long as it’s cheap.
I got back more examples of monitoring automation. Some even more jaw-droppingly awesome than before. But it's still not what I wanted. I started cornering people one-on-one to see if maybe email was the wrong medium.
What I discovered upon intense interroga... discussion was that IT pros are very familiar with how automation is accomplished. We remember those details, down to individual lines of code in some cases. But pressed for information that describes how that automation affected the business (saved money, reduced actionable tickets, averted specific downtimes which would have cost X dollars per minute), all I got was, "Well, I know it helped. The users said it was a huge benefit to them. But I don't have those numbers."
That was the answer in almost every case.
I find this interesting for a couple of reasons. First, because you'd think we would not only be curious, we would positively bask in the glory that our automation was saving the company money. We are, after all, IT professionals. The job practically comes with a cape and tights.
Second, we're usually the first ones to shout "Prove it!" when someone makes a claim about a particular fact, event, effect, or even opinion. Did someone say Star Trek IV was the highest grossing movie of the franchise? Show me the BoxOfficeMojo.com numbers for that!* At a dinner party and someone says sharks are a major cause of death in the summer? You're right there to list out 25 things that are actually more likely to kill you than sharks.**
But despite our fascination with facts and figures when it comes to ephemera and trivia, we seem to have a blind spot with business. Which is a shame.
It's a shame because being able to prove the impact that monitoring and alerting has on the business is the best way to get more of it. More staff, more servers, more software, more time, and most importantly, more buy-in.
Imagine providing your CEO with data on how one little alert saved $250 each and every time it triggered, and then opening up the ticket logs to show that the alert triggered 327 times last month. That's a savings of $81,750 in one month alone!!
Put those kind of numbers against a handful of automated responses, and you could feel like Scrooge McDuck diving into his pool of money every time you opened the ticket system!
So prove me wrong. In the comments below, give me some examples of the VALUE and business impact that monitoring has had.
More than just giving me grist for the mill (which, I'll be honest, I'm totally going to use in an upcoming eBook, totally giving credit where credit is due, and THWACK points!) what we'll all gain is insight into the formulae that works for you. Hopefully we can adapt it to our environment.
* In actuality StarTrek IV ranked 4th, unless you adjust for inflation in which case it was 3rd. First place is held in both categories by the original motion picture. The more you know.
**Sharks kill about 5 people annually***. Falling out of bed kills 450. Heck, bee stings claim 53 lives each year. So go ahead and dive in. The water's fine and Bruce the shark probably will leave you alone.
*** Unless you are watching "Sharknado". Then the number is closer to 16 people.