Enterprise Architecture: From Incite comes Insight...

I previously blogged on Social CRM and its effect on the enterprise ecosystem. Today, I will share human perspectives and things your IT executives need to noodle regarding social CRM...



OK, so on the chance that your enterprise is a leader and not a follower, you are probably thinking a lot about social media and how to integrate it into your enterprise architecture. One of the troubling aspects of social media is that the enterprise is no longer in control of the conversation for either its customers or even its employees. Does your enterprise have a strategy to empower coworkers by providing them with inspiration to exercise their calling or is it to stifle conversation?

We have all heard about the challenges of IT aligning with the business and how enterprise architecture requires listening to the business. The funny thing about this mantra is that enterprise architects now have a greater ability to directly listen to the conversations of customers and bypass the business and the overdistillation and misdirection that occurs. Sadly, the vast majority of enterprise architects are focusing on anything that matters to the customer.

So, its time for enterprise architecture to show leadership when it comes to social media but the best example of leadership isn't in hiding behind the corporate veneer but to figure out how to remove barriers from allowing outsiders to talk with insiders. Leaders in the social media space have realized that real leadership requires you to acknowledge that you are a human being first and a leader second.

Much of the mantra of leadership within large enterprises occurs through the one-way flow of information to their target audience. This requires not just changes in culture related to awareness of capitalism, but the elimination of senses of class and rank within the enterprise. Does the customer care if you are a manager, director or VP or do they care to simply find someone who can solve their challenge?

On too many occasions, whenever a new concept emerges in the marketplace, enterprises immediately run out and hire the boutique consultancy of the day to have them create a strategy. In the new world, this is probably the biggest mistake you can make. Don't outsource thinking around social media, but instead figure out ways to embrace it. What prevents each and every member of the enterprise architecture team from making social media personal to them? Wouldn't your strategy be much better if you understand social media from the perspective of the user...

Many IT executives have given up on their employees well-being and believe that the only way to save money is to pursue the path of outsourcing without considering other alternatives. Let's look at how their thinking may be flawed...



As an employee of a large enterprise, I run across lots of employees who have a variety of backgrounds within IT. Some have achieved their bachelor's degree while working and others stayed the more traditional path of going to college right after high school. Some have pursued Master's degrees while others have pursued technical certifications. Some were graduates of computer science programs while others have majors in more esoteric topics such as biology and english literature. Collectively speaking, I asked several individuals in various human resource capacities to tell me whether the performance of a particular demographic as measured by the annual review process favors one group over another and the answer was not mathematically provable.

The challenge faced in large enterprises is less about technical aptitude and more about the ability to understand the business and its particular challenges. Each corporation has its own challenges when it comes to methods for communication, cultural attitude, willingness to be leaders or followers within their vertical and even what it truly values above and beyond the stated requirements of a job description. None of this can be learned in a university setting.

When I worked for my employer in the late 80's, my salary was only $16,000 a year. I was fresh out of the United States Coast Guard with a high-school education. To be fair, I did have IT experience when in high-school. In those days, the work day was less than 40 hours. In high-school, I worked at Cigna in a department named Application Field Services which had two responsibilities. The first was to be technical support for when IT programmers got stuck and needed guidance, in essence they were the elite of the company. The second responsibility was to oversight into mainframe infrastructure. In this department, I had the opportunity to learn both hardware and software. In fact, I changed out many terminal controllers and even did minor electrical repairs on them. The interesting thing was I would start my work at 2pm and go all the way to 9pm, which essentially was a full-time job in IT while being in high school.

Anyway, someone recognized at an early age that I had the aptitude to be a good IT employee and gave me the chance. Imagine what would happen if America started harvesting all the IT talent that exists within the nations high-schools and start to create the same experience I had for them. The economic perspective says that the average outsourced developer in India from a major vendor such as Wipro, Cognizant, Infosys, etc is billed to the client in the range of $30 to $50 an hour. One needs to ask can you pay a local high school student $15 an hour to accomplish the same task.

The interesting thing is that on one hand, modern enterprises are looking for advanced degrees but on another hand, there are many high-school students that we all know of that are proficient languages such as Visual Basic and Java that could do a better job than many professionals. So, why are CIOs ignoring the basic economic principles. One perspective may say that it is too difficult to find these types of students. I say that the real challenge is in the fact that the majority of IT executives nowadays don't know true talent when they see it and prefer to remain blissfully ignorant. Maybe, they know that if their CEO even thought for a second along the lines of this blog that they would be immediately fired. After all, what does it say about a CIO who doesn't have the innate ability to recognize talent. Can't we all agree that this is job number one of leadership...

Cloud computing is the latest hype where every software vendor, industry pundit and IT executive who practices management by magazine praises this new innovation, yet no one talks about what is missing. So, I will take the opportunity to share three things that I have been noodling about cloud computing that are not discussed elsewhere...



Elasticity: Cloud is sold as being able to grow and shrink capacity on demand. The challenge with this model is that much of the activities of shrinking and growing are somewhat manual in nature and involve putting a human in the middle. If my application is under severe load where it could benefit from having a few additional instances being spun up, what standards-based APIs would it use to communicate to the cloud to make the request? More importantly, since most applications are load-balanced, how would the cloud communicate to my load-balancing infrastructure that it should add new servers to the pool dynamically? Most of the conversation regarding cloud is about how to use APIs to reach into the cloud, but not about how the cloud can communicate back to critical infrastructure such as firewalls who could dynamically open access based on new IP addresses or even load balancers for that matter.

Compliance and Discovery: So, what happens if you have a running instance that just happens to be co-located on the same set of servers as an Al-queda application when the FBI decides to seize all the equipment? What happens if your backup is done in the cloud and it is on the same set of tapes as Bin-Laden's information when it is also seized by the FBI? What happens if you needed this backup because you live in Rhode Island where floods were occuring and you needed to recover but couldn't? Multi-tenancy isn't just a strict access-security concern, but could also pose discovery related challenges.

RefactoringOK, if you are in a large enterprise and are sick and tired of the software vendor community coming up with the latest approach to technology that requires you to rewrite enterprise applications from scratch for each new paradigm shift then raise your hand. We have already rewritten applications when we moved from mainframe to mini-computer to client-server to web-based to SOA and so on. The business is sick and tired of having to invest in this type of activity. While technologies such as Amazon Cloud, Azure and so on are cool for new applications, when will the conversation shift to refactoring? Who is noodling how I can incrementally take my mainframe COBOL application and move say just storage to the cloud without having to rewrite in a Java or .NET language? Are there any book authors noodling a manuscript entitled refactoring enterprise applications to the cloud? Can we have a healthy conversation on how to leverage many of the assets we already have invested millions in and figure out ways to harvest it? The cloud can be wildly successful if we put on our economic hats. The cloud needs to become less about technology and more about business value...

Not a day goes by where some IT executive isn't talking about the need to hire and retain the best talent. Sadly, their mantra doesn't match their actions...



Let's start with an analysis of the hiring process which usually is kicked off by writing a buzzword laden job description that is chock-a-block filled with lots of terminology but otherwise lacks substance. This description is then posted on the company website and various job boards ranging from Monster to DICE and then the game of sitting back waiting for people to apply begins.

Ask yourself is this truly the way to find the best and brightest? Of course not! Top talent traditionally doesn't spend a lot of time searching job boards and instead prefer to be recruited. Now ask yourself does the recruiting arm of your company actually have the ability to properly find talent especially in situations where they aren't visiting job boards?

As an IT executive, should your enterprise architecture team have a strategy around social media that goes above and beyond using channels such as blogs and twitter to post one-way heavily sanitized public relations communications that no one reads? Should enterprise architecture figure out ways to help out your recruiting staff to writing compelling 140-character twits with the right hashtags in the community which top talent resides.

Taking this theme one step further, should recruiters actually be recruiting? Let's acknowledge that top talent is frequently contacted by intermediaries better known as recruiters who know nothing about the position other than what the job description says. Wouldn't it be better if executives didn't outsource the requirement to find talent to others and took this on for themselves?

So, let's take a hypothetical example of recruiting in the world of industry analysts. Imagine you work for Gartner and you wanted to recruit top analysts such as Brenda Michelson of Elemental Links, James Governor of Redmonk, Richard Veryard or even Richard Mogull and you sent some no name recruiter to establish a conversation. What to you think the odds would be of success? Now, if you could say have Gideon Gartner perform the same activity, do you think many more would be willing to at least listen?



The funny thing about enterprise architecture and those who are enterprise architects is that they tend to participate in multiple communities and have insight into ways to interact with them that aren't shared by many other demographics in the enterprise. What could recruiters learn from their observation of how enterprise architects network? For example, in terms of my background, you will find me deeply participating in communities related to not only enterprise architecture, but also security & privacy, BPM, SOA, cloud and networking yet how often am I ever asked how to best reach these communities by IT executives when it comes to recruiting from them? For those who will get it twisted, me isn't really all about me but the fact that there are dozens of people just like me within your own enterprise whose body of knowledge isn't being leveraged to find top talent.

The discipline of finding top talent doesn't just extend to employees and is equally applicable to the world of outsourcing. How many times have you interacted with an outsourcing firm (e.g. Wipro, Cognizant, TCS, Infosys, etc) or even a global consultancy (e.g. Accenture, CapGemini, BearingPoint, etc) where you were sold an individual yet over time you didn't get exactly what you expected? While you have outsourced responsibility for delivery, does this relieve you from ensuring that your business operates at maximum efficiency? Have you considered that the best way to ensure a successful outcome is to manage talent regardless of where it resides?

In my day job, when working with outsourcing vendors I take personal responsibility for ensuring that I get the best talent available from the firm. Instead of just sitting back and reviewing resumes I prefer to instead proactively figure out who is the best talent within these firms and ask for them by name.

While most firms do their best to hide their talent in this regard, there are many ways to discover them if you are tech-savvy. Consider the scenario where one is seeking a developer knowledgable in web application security. In this scenario, I may use the community to help guide my decision and have been known to ask members of the Open Web Application Security Project who attends their local chapter meeting from these firms. Wouldn't it make good business sense or at least is a predictor of success when comparing candidates to choose the one who will spend their own time learning about their craft over ones who don't?



Another bad practice in recruiting top talent is to discuss salary way too early in the conversation. Too many recruiters arbitrarily filter out good candidates simply because of their current salary. If I make $150K a year and am applying for a position that pays $250K a year, does that automatically make me unqualified for the position. Instead of focusing on what the candidate currently makes, why not focus on whether the candidate can do the job and what the position is worth? Having a first conversation that involves uncomfortable conversations around salary simply are getting started on the wrong foot. We need to focus on passion more than compensation.

Isnt it fascinating that a large enterprise can outsource work around the planet but can't hire top talent and allow them to work from home? If you have to be with a company for a significant period before you are allowed to work from home sends a bad message. Minimally it indicates that you are immature in measuring productivity and rely on soft factors. It also indicates that you truly aren't serious about finding top talent and simply desire occupationalists who will resort to clock punching once the face time clock has elapsed. Sadly, by not considering people in other geographic areas, you are doing a disservice to your enterprise by not letting it transition from good to great...

This is Part One of a two part blog entry on Enterprise Architecture and NoSQL. Part Two will cover the technical aspects of NoSQL while this entry will focus on the human aspects...



If hype is the plague on the house of software, the NoSQL movement most certainly capitalizes the investment. NoSQL has managed to gain lots of media attention for what should be viewed as an incremental change in how we leverage technology.

Social networking sites such as MySpace, Google, Twitter and others have reached the limits of traditional relational database technology. Within most enterprises, a wildly scalable application may have a population of tens of thousands of users with a user concurrency of a few thousand. Internet social networking sites tend to have tens of millions if not hundreds of millions of users who have larger expectations when it comes to response time and therefore their architecture needs to be thought about much differently.

Anyway, independent of the technology, enterprise architects can learn a lot from the NoSQL movement in terms of creating hype. First and foremost, the community came up with a term not to describe what NoSQL is but what it is not! More importantly, NoSQL doesn't really mean that you can't use SQL. In fact, many of the current NoSQL implementations leverage SQL as its query model.

This tactic most certainly brings people out of the woodwork who will go out of their way to prove you wrong via blogs, twitter and other social platforms increasing the visibility of the effort. As a blogger, I have noticed the trend within my own blog where if I post a 100% accurate blog, there is no dialog as it shuts down the ability for others to participate. However, if I leave others an opening by stating something that others can attack, then participation increases.

From Incite comes Insight and within large enterprises, NoSQL leverages another collaboration pattern that is time tested. Given a scenario where two people have taken opposite stances, there is always someone within an enterprise setting that believes their value is to moderate discussions and drive things to a middle ground. This is analogous to neither being democratic nor republican but merely a moderate and NoSQL leverages this mantra well.

I believe that moderation in of itself is neither good nor bad. Tempered responses need to be made in context and based on doing the right thing and not just want is most popular. We have all heard the phrase, bleep or get off the pot. I don't think you want to take the meet in the middle scenario here as you are hogging the pot and might have something run down your leg.

Anyway, I think the NoSQL movement did the right thing by coining an otherwise inaccurate term for their needs and has been successful in getting the community at large to pay attention to their message in a timely and scalable way. This is a technique that I would encourage enterprise architects to add to their bag of tricks...

I am an avid follower of Charlene Li, Jeremiah Oywang, Ray Wang and others from Altimeter in terms of their wisdom regarding social CRM. The conversations to date tend to appeal to those who have public relations roles and primarily focus social media ideas on sales, recruiting, service and support. I have always felt that something was missing from the conversation and a recent social CRM interaction I had filled in the gap. Let me share an enterprise architecture perspective...



Another industry analyst whom I follow, sent me a direct message via Twitter inquiring whether I could help him resolve a challenge he had with acquiring a business owners insurance policy. He knew that I was well connected within my organization that enterprise architects tend to understand who to contact since it is very much a social role and more importantly that since I have an affinity to the security community, I might have a clue as to how things truly work.

Anyway, upon receiving his DM, I first ascertained something I never really thought about in terms of connecting with customers. He described his business as security consulting to an underwriter which was accurate, but this has a totally different meaning within the insurance ecosystem than it does to customers. The insurance vertical uses standard classification codes and security consulting would feel more like what retired FBI agents do in the private sector than advising enterprise clients on best practices for software security. So, the first lesson is to think about how to use social networking to educate consumers on terminology. This simple interpretation could cause a business or consumer to pay more for insurance than they should.

Upon receiving the information, I of course had to figure out what particular interactions he had with our organization. The business community at large always talks about the notion of a 360-degree view of the customer and social media makes this particularly challenging. Most architecture in this regard has the built-in assumption that you know something that feels like a primary key. Since this was twitter, I dare not ask for publicly identifiable information over a non-secure channel and therefore had to resort to other tactics.



Many enterprise systems will allow for one to search for a person based on identifiers such as phone numbers, addresses, social security numbers and so on, but to date I have yet to find one where you can search based on a twitter handle. So this begs the question of what other types of attributes above and beyond simplistic inetOrgPerson should an enterprise store based on its consumer-base. This has turned into homework for me to not only capture attributes but made me think about user-oriented registration and other factors.

Of course, I am not in the customer service business and needed to route this request to someone more qualified. Do I use more traditional email? Since I am security-focused, do I inject something into the current workflow from a CRM perspective so it will follow the defined path even though this would be a violation of controls?

Well, of course I found an answer that was timely and compliant. The actual approach doesn’t matter much as it was company-specific but this does beg the question of how does an employee, any employee become an ambassador of customer service in a world of traditional controls.

With a few exceptions, the vast majority of enterprise architects I know spend an awful lot of time focused on internal issues whether it is rationalization, the cloud, storage governance, data center consolidation, creation of reference architectures, portfolio management and other considerations that aren’t even visible to customers. One should ask whether IT can be truly successful if we are busy listening to the business but otherwise are blissfully ignorant towards the customers they serve.

Maybe a discussion in the blogosphere on how enterprise architects should think about social CRM is in order. We need to move this away from a constrained discussion related to public relations towards a conversation that all can participate in…

Several weeks ago, I had a great conversation with a person in charge of marketing for a large vendor (and former Forrester analyst) regarding a whitepaper on Federated Identity for the Insurance Vertical I authored. The conversation was productive, but in hindsight, there were several missed opportunities that analyst relations professionals should noodle...



During the call, I gained insight into how vendors prioritize industry analyst briefings. The important but otherwise missed perspective was in how she was keenly aware of not only her coverage area but which firms were short-staffed either due to departures or the creation of new positions.

It is fascinating how many software vendors steal great resources from their end-customer base but few ever to think about figuring out how to get them installed as analysts within analyst firms. Wouldn't the absolute best way for a vendor to be certain that their product offerings are seen in a positive light would be to advocate for users of their technologies to get hired as an analyst?

Why would an analyst relations professional sit back and wait for a "problem" to emerge such as the challenges of working with a "difficult analyst" as described by Carter Lusher? A selfish posture would encourage proactive behavior. A quick browse through websites such as Gartner, IDC, Yankee Group and other analyst firms show that there are lots of positions that could be filled by end-customers who would make great analysts.

For those who are reading this post and thinking it is all about me; nothing could be further from the truth. This post is all about taking your customers and leveraging them for competitive advantage where I am simply introducing another technique to your playbook...