The way in which fiber optic networks can be modeled on a map with a Geographic Information System (GIS) is truly fascinating. Thousands of man-hours of mouse clicks and keyboard strokes go into it -- all with the intent to transform drawings and spreadsheets to bits and bytes of data stored in GIS-based Network Documentation Systems (GNDS). And that’s not all. The investment in this proactive effort represents tens, if not hundreds, of millions of dollars of investment in documenting network assets.

The bigger question is this: From the chore of the initial input to the steady struggle of data maintenance, how is that information being used today? And are those who have invested in the technology reaping rewards, or reeling with remorse?

This article is not intended to calculate the return-on-investment for a GNDS, but rather to examine the usefulness of such systems and the data contained within them. Nevertheless, it does help to have an understanding of the investment required to implement and maintain such a system.

To understand the costs associated with a GNDS, let’s examine a company that just a few years ago had their network drawings and related information spread across multiple platforms. This company was managing nearly 30,000 route miles of fiber optic cable with annual revenues over $1 billion. The effort to migrate their disparate data to a common GNDS was demanding and tedious. Overall, the costs of software licenses along with the labor to gather and input the data totaled nearly $3 million, or about $100 per route mile of fiber optic cable. Ongoing annual expenses for software maintenance fees and the labor to maintain the database is nearly $800,000, a figure that is 0.32% of the $250 million construction budget for building entries and network expansion. Though only a very small fraction of the cost of the physical network, the cost of this virtual network is still significant; which makes one ask: Why is a GNDS necessary?

A Picture Is Worth…?
In general, a GNDS is designed to allow a user to draw network elements such as buildings, cables, poles, duct, etc. on a digital street map. Then, have the ability to splice or terminate the contents of the cable, whether they are fiber or copper; to create a virtual representation of a physical transmission circuit. With the data in such a format, the end user can determine the location of a cable break, better plan new builds, and perform a myriad of other functions.

In 2009 our organization sent out a survey to 50 GNDS users across 32 companies. It was designed to determine what the system was ultimately used for. The respondents were asked how often they used the application and the data within it to support various job functions. As can be seen in Figure 1, the majority of these users utilized the software for Network Design and Planning and for Network Maintenance on a daily basis. This is to be expected since the software application they were using is primarily intended to support those functions.

Figure 1.

But, it was interesting to note that nearly two-thirds of those surveyed were using the data from the system to support Sales and Marketing needs. Similarly, it can be seen that nearly half of those surveyed used the data to answer inquiries dealing with Tax and Accounting.

The significance of these results is that they show that the data created by the engineering department is being leveraged across organizational boundaries. This data, ostensibly conceived to manage the most valuable corporate asset, the fiber optic cable, is now becoming a precious asset itself and its use continues to expand.

Another pertinent observation revealed within Figure 1 is the relative number of users that were not aware that the data could be used for Sales and Marketing (52%) and Tax Reports/Accounting (34%). However, when this survey was taken, users were already shifting their perceptions in regard to GIS data, and the impetus of that change came from a familiar company with a strange name: Google.

Way back in 2005, Google released Google Earth. Google Earth brought web maps to the main stream, and it became easy for an ordinary user of GIS to export their data to Google Earth to share it across the Internet. GIS data was making its way out of obscurity and onto the corporate radar. Sales people could now envision selecting a point on a map to quickly determine serviceability and a rough estimate of cost-to-build. Marketing professionals would use maps to design marketing campaigns. CATV sales specialists could have the ability to view neighborhoods and the associated network penetration. The examples go on.

But while the data from the GNDS was starting to be used across the enterprise, a GIS specialist was frequently needed to get the information in a format that everyone could use. The data, though discovered, was still not easily accessible. This situation has become a frustration to users whose awareness of web-based map applications has grown over the years. Such users expect simple, interactive interfaces that make it easy to analyze and visualize the data that they depend on.

Data Mining and the OSP
A few months ago I was summoned to the headquarters of a client of mine. This company had a mature GNDS; the objective of this visit was to help the IT group understand the data well enough in order to mine it for the treasure trove of information that they had been yearning to get their hands on. As I sat across the table and began to describe the spatial components of the database, you would have thought that it was Christmas morning and that the IT professionals were 5-year-old kids. They were excited, and began describing how this data could link to other data warehouses. Their enthusiasm was contagious, and we began to brainstorm.

Within a few hours we had come up with not only the way to disseminate the information, but the platform as well. Our strategy was to use Rich Internet Application (RIA) technology. The main reason for this strategy was to create web-based applications that were familiar to the end users and intuitive. RIA would give us those tools since it is the technology used by many online gaming systems, systems which require impressive graphics in order to attract users.

Web-based alarm monitoring system referencing geospatical data in GIS-based network documentation system (GNDS).

Most important, the design had to be simple. There had to be a very short learning curve; we were dealing with people who expected the application to be as easy to use as the mainstream Internet maps they were already dealing with.

Furthermore, the RIA environment would make development significantly cheaper and faster because we were working with a technology that ordinary developers already understood; they did not have to be specialists in GIS. From that point, we were able to roll out prototypes in order to meet demand, and tweak those designs to promote them to production-ready applications.

RIA technology, along with other technologies like HTML 5, has ignited the demand for single-task-centered web applications that now serve up the historically secluded GIS data from the engineering underground. The data is becoming more widespread than ever before, on the web and mobile devices, and end users have direct access to it. But, is there a cloud on the horizon?

Since the engineer-sponsored spatial intelligence is integrated with existing business intelligence systems, caution and discipline with the data is essential. If companies don’t recognize and treat this data as a business asset, then it will not be managed as such -- and therefore will not be controlled and maintained with the care and attention required.

Most engineering organizations view the data simply as a tool to aid them in designing and maintaining a network. That is, the maintenance technician in muddy work boots, recording sheath markings from a cable in a pull box, may not realize that those numbers will be used to calculate route lengths for investor relations reports that will be analyzed by a Wall Street broker in Bruno Magli slip-ons. Thus, as the data becomes more critical, the maintenance of the data is more vital.

This subject requires much more attention than what is mentioned here. Suffice it to say that a casual approach to data maintenance processes and procedures will ultimately result in missing or erroneous data. That will infect down-stream reporting and analysis --  and lead to poor decisions that can potentially cause customer resentment and investor disappointment.

It Hurts So Good
Customer attrition, or churn rate, is a metric used by companies to gauge the satisfaction of their products or services. For GNDS vendors, the churn rate is around 15% annually. Interestingly, the number is not the significant finding; rather, it is the question of what happens to those companies that are a part of the churn rate.

It appears that those companies are divided into 2 groups:
1. Involuntary Churn -- They were no longer in business or were involved in a merger.
2.  Voluntary Churn -- They decided their current GNDS was not a good fit for them.

What is striking is that the group that had the opportunity to make a choice, Voluntary Churn, did not go back to relying on spreadsheets and drawings. Instead, they searched for another GNDS vendor or they set out to develop their own. That is, even though they rejected a particular GNDS, they did not reject the need for a GNDS. Thus it may be that even those who may not have had a good experience with a GNDS still determine that such a system is essential to run their operations.

Such a finding fits well with the current direction of the GIS market in general. A report released by Daratech, (www.daratech.com) an industry analyst group specializing in GIS market dynamics, in January 2011 shows that sales of GIS software, services and data grew at 10.3% in 2010 to $4.4 billion. This same study shows the market continuing to grow at a rate of 8.3% to become an industry that generates nearly $5 billion in sales annually. This growth is what continues to drive GIS to becoming a necessity for any company required to manage outside plant assets.

Because GIS has become so ubiquitous, in not only the engineering and planning aspects of everything to do with outside plant, but also related functions, it is now becoming a requirement. In that regard, outside plant drawings without a spatial projection are quickly becoming obsolete, and their ultimate demise appears imminent.

Knowledge Is Power
During the course of conducting interviews on this topic, it is noteworthy that the prevailing comment made by officials attempting to rationalize expenditures for their GNDS was "Knowledge is power." Though nebulous and subjective, that seemed to be the ultimate metric used to explain the necessity of the GNDS expenditures.

It's fitting that such a rationale makes sense, because most of what a telecommunication provider deals with happens somewhere. Indeed, their outside plant assets carry the services that they sell; they are the core of their business, their life's blood. Those assets span cities, counties, countries, and oceans. Thus, when it comes to everything from finding broken cables, to selling services, to leasing assets and paying taxes or mitigating lawsuits, the one prevailing phrase is: "Where is it?" The implications of answering that question, either accurately or otherwise, will directly result in either the reaping of rewards or reeling in remorse.

 Randy Anderson is co-founder and CEO of Advance Fiber Optics. Since 1993, Randy has spent thousands of hours not only documenting fiber optic networks, but working with clients to determine how to best report on the data obtained for a fiber optic network. He manages some of the most talented people in the fiber optic industry whose sole purpose is to develop software tools to make documenting fiber optic networks easier and more reliable. For more information, email him at: randya@advancefiber.com or visit: www.ospinsight.com.

What’s your take on this subject? Leave a comment and get the conversation going.