How I failed my company as a monitoring architect.

picard-facepalmI didn’t realize it until recently but now I know that I failed one of my previous companies when I was their monitoring architect. It’s not easy to admit past failures but it’s an important part of growing as a professional and as a person. Before I explain how I failed let me provide some short background.

Filling Gaps the Best Way Possible

As a monitoring architect it was my job to make sure that we had monitoring tools, training and processes in place to properly support the business. I was extremely passionate about my responsibilities and worked tirelessly to understand the gaps in our strategy and plug those gaps in the best way possible. And there’s the rub, my idea of the “best way possible” was flawed.

During my time as a monitoring architect I got to test out many different technologies. I put the tools through their paces in my lab. I put the companies through their paces by testing the sales and support organizations before making any purchases. I figured out how each new technology would fit into our existing processes and created new processes where they were needed. I documented everything thoroughly and ultimately ended up bringing in multiple new tools over the course of a few years.

Ignorance is Bliss, and Then It’s Not

The tools and processes were successful from the the perspective of helping to reduce customer and business impact. Unaware of my failure I was proud of what I had accomplished and I moved onto other things and new companies. Then I had a conversation this week that burst my bubble and woke me up like a slap to the face. With one sentence I realized my failure… “Our users only open the tool a few times a year when there are problems and they can’t remember how to use it!”. (Thankfully we weren’t discussing my current companies software)


Smack!!!! It hit me fast and without mercy. I realized that for years I had overlooked an absolutely essential aspect to successfully implementing technology. As a technologist I am well trained and very experienced in using “less than intuitive” products. But, I’m the exception to the rule in the real world. Most people don’t have the time nor the desire to become technology experts. They simply want and need a tool that solves their problems and is easy to use.

I had been so focused on features and processes and filling gaps that I overlooked that fact that power + ease of use = successful adoption. I had unwittingly robbed my company of the possibility of massive return on investment by selecting tools that were okay for me to use but very difficult for most of the rest of the company to use.

Immediately after this disturbing realization I recalled the timeframe shortly after I brought a new tool into my organization. I had set up a vendor led 1 week training session for the people on my team to learn our fancy new toy. The training session went well with only a few bumps in the road where the usage concepts were somewhat difficult. The problem was that within 2 weeks of the end of training most of the team had forgotten how to do most of what they were taught except for the simplest of tasks. That should have clued me in but I was blinded by pride and continued along blissfully ignorant until this week.

Never Repeat the Same Mistake

I hate losing. I hate getting something wrong. I hate the fact that it took me years to realize my failure but I’m grateful for having realized it at all. It’s a mistake I will never repeat and hopefully one that you will never have to make now that you have read this story. Please, when making decisions for a large group of people, really think about the people that will be impacted by your decision and make sure you consider the fact that everyone is different and what works for you wont necessarily work for them.

These days I’m happy to work for a company that is disrupting the enterprise technology market with powerful AND easy to use software. I didn’t truly realize how important it was until this week. If you’d like to check it out for yourself click here and start a free trial today.

Why I Joined The Leading APM Provider AppDynamics

A new year, a new iPhone and a new quarter. What else is new? How about a new company?

Last month I was fortunate enough to join a stellar marketing team at one of the fastest growing enterprise software startups in the bay area. The company you ask? AppDynamics, and did I mention we’re also the leading next generation Application Performance Management (APM) provider for modern architectures in distributed, cloud, virtualized and on-premise environments? We exceeded our targets for 2011 achieving an astonishing 400% growth in bookings. Not too shabby for being the new kid on the block in a competitive market already inundated with vendors. You have old school APM tools from megavendors like CA, HP and Compuware (was dynaTrace). Then you have the new school breed such as New Relic and AppDynamics. In fact, Gartner’s MQ lists over twenty vendors. So with such a crowded market why did I even consider such a move?

Well there’s a laundry list of reasons, but here are the top ones that come to mind.

1. Business Innovation. This is another kind of BI not just Business Intelligence. It’s really a breath of fresh air to be working with an organization that is not only obsessed with pumping out insanely great technology every few quarters or so, but also open to embracing innovative approaches to every discipline of the business including creative marketing and sales strategies. Often times enterprise software companies unabashedly attempt to cloak themselves in slideware selling a “vision” or an enterprise solution poles apart from reality. Unfortunately when it comes down to an actual evaluation, you end up having to attend a dozen meetings just to see an applicable demo, a one week to two month proof-of-concept followed by throwing millions of dollars at consulting and implementation services, which segues to my next point.

2. Ease-of-Use. This simple yet powerful concept has been repeatedly neglected or intentionally ignored by many enterprise software companies. Luckily, the Leaders of the New School such as Apple, Salesforce, Box, etc. (not Busta Rhymes group) have changed the way end users value an intuitive user interface and design. At AppDynamics, we’ve adopted a similar mindshare. “Easy” is the new world order in this industry because the managers, engineers and folks in IT operations are encountering enough complexity as it is with these modern architectures. I doubt the last thing that they want is another tool to further complicate their lives causing more frustration on the job. At the end of the day everyone is a consumer – the least common denominator – who wants to use software that helps us demystify our lives and makes us successful at our jobs (unless you’re a sadist).

Software that is easy to install, implement and use can have a tremendous impact on the bottom-line of a business. Suppose you end up rolling out a new system but end up having to spend a chunk of company change on implementation and training costs. What impact does that have on your productivity and ultimately your company’s bottom-line? Here’s an example from Avon’s Q3, 2011 earnings transcript,

“Despite extensive pre-implementation testing, we had greater than anticipated implementation challenges in the go-live. Significantly higher business complexity in this market contributed to a greater than expected level of disruption, as I said, when we went to the go-live environment.”

Many vendors make enterprise deployments akin to embarking on an IT version of manifest destiny. I’m sure you can think of a few applications in your own IT toolbox that fit the bill where at some point you ended up asking yourself, “Why can’t this be as easy as [fill in the blank with some consumer app]?”  Fig. 2. See empathetic frustrated user to your left.

That was compelling enough for me to join AppDynamics. We truly understand the business significance as to why software ought to be easy 360 degrees around especially in production. I’m not saying that the work designers and developers have to do to achieve this “Easy” goal is easy in itself. I have an unrequited love for the folks in engineering who possess the talent and perseverance in coding applications, but that doesn’t excuse a vendor from selling you a dream and then leaving you stranded to implement a nightmare all because there wasn’t enough emphasis on ease-of-use.

3. Application Performance. This one is near and dear to my heart and arguably the main reason for me to join AppDynamics. It takes me back to the challenging days and sleepless nights I endured while working on a massive global PDM implementation at LG Electronics jointly with Dassault Systemes. The year was 2008. Skynet hadn’t become self-aware yet. App Man was just A Man in the throes and woes of IT operations, and half way around the world over in Seoul, Korea I was managing juggling recurring performance issues on a weekly basis with our PMO having to answer to the beck and call of the LGE CIO. The project’s launch date had been delayed due to various complications with the implementation (that’s a whole other story). Any ideas what one of those might have entailed? If you guessed “performance”, congratulations! You’ve won! Download your free copy AppDynamics Lite.

Every week new customizations were being released from R&D back in the states, PS in Korea and SI’s sitting on the other side of the room. You could call it Agile development’s nemesis, frAgile development. The dynamic nature of our java-based environment only introduced more challenges to the performance team who were heads-down trying to reverse engineer someone else’s code and refactor it using APM tools that just didn’t provide us with the full visibility we needed to comprehensively profile and diagnose application performance issues (using JenniferSoft). In fact, one of the consultants on our team ended up creating his own profiler to expose these blind spots, but what we really needed was a next-generation APM tool that would visually map and connect the dots for us like the one below.

Then we ran into another stumbling block after we completed migrating legacy data to a new “production” environment. When the time came to retest the entire set of performance use cases in this new environment we experienced all kinds of performance regressions. Since everyone was collaborating so well with each other for over the past two years, we all cheerfully marched forward without any finger pointing as to what the root cause was. Ok, so it wasn’t that utopian. Fortunately, because of everyone’s undying commitment and personal sacrifices, the project went live successfully in mid 2010 with over 2,000 users visiting the system per day. In hindsight, we could have easily saved a month’s worth had we used a better tool thereby eliminating the usual suspects.

From that experience I’ve come to appreciate and understand how business-critical managing application performance is for any company. Now I am on a mission to spread the word of AppDynamics to help companies manage rapidly evolving, distributed environments.

Buckle up 2012, we’re just getting started.