Validation: Determining the value of ICT methodology.
By Steven Jordan on November 26th, 2013.
Abstract:
ICT methodology is used to research a BGP routing problem. The research found that interdepartmental communication breakdowns failed to immediately identify and resolve the problem.
Problem:
Customers, all from the same organization, complained of poor web site usability. Symptoms of the problem were described as slow loading web pages, limited functionality, and unreasonable delay of data.
Multiple people, all from a single customer location, called to complain about access to our web service. The Customers depended on the data and they were unable to complete their work because of reliability problems.
I usually have limited contact with our customers, however one of our sales representatives asked for my help with this issue. We were reasonably sure there were no network or server problems but we felt it was best to investigate the problem further.
After I reviewed internal systems I reached out to the customers. I spoke with 2 managers, and 3 different staff members, from their organization. Each person had a similar experience. The Internet appeared to work OK, but everyone was of the opinion that our site was slow and unresponsive.
The company in question is a large organization located on the other side of the country. Simple troubleshooting was a challenge because all workstations were securely locked down. After some persistence I was able to get a hold of someone from their help desk department.
The help desk associate stated that he firmly believed that there was a problem with our service. I politely explained that while it was a possibility, I hadn't found evidence to confirm his position. The help desk associate sounded annoyed with the situation, but I was happy he had agreed to troubleshoot the issue.
Before I could offer a suggestion, the associate told me, "See? I just pulled up Google quickly. Yahoo is fast too.."
I politely countered, "Yes, but there may be some other underlying problem. It would be very helpful if we could run some latency tests."
The help desk associate shot back, "Google and Yahoo work fine. Our Internet works fine. Clearly your company has problems. I suggest you fix it." Click. The help desk associate had hung up the phone.
That hadn't gone as well as I had hoped. People still needed help so I decided to try again. I recalled a manager that I had spoken with earlier. Kathy, had been frustrated, but she was also very friendly. I called Kathy and explained I was unable to determine the problem based on my initial conversation with help desk. I politely added that the phone call must have been accidentally disconnected.
I think Kathy got the hint. She managed to track down another help desk associate that had helped her with other issues in the past. The second help desk associate, Bob, was very helpful. Bob was forthcoming as he explained his access to network resources was limited. He was however, able to open a command prompt from which we ran some basic tests.
Basic ICMP tests, indeed, suggested there may be an Internet related problem. We were also able to determine the organization's public IP information. I was then able to determine the organization's Autonomous System (AS) number, and identify detailed information about the root of the problem.
Even after our tests, Bob was uneasy with approaching the organization's network administrators. Bob sounded reluctant as he promised he would escalate the issue. I assured him it was the right thing to do and let him know he was a big help with identifying the problem. I also I told him I was going to immediately write an email that detailed the problem, and planned to forward it to the organization's management. *Click here to view the email.*
ICT is the study of technology and organizations. The hierarchical divisions between departments, including interdepartmental, were nearly as disruptive as the BGP routing problem. The end-users lost a full day of productivity because of poor communication. The help desk staff provided limited assistance to the end-users. The problem was additionally frustrating because the network administrators had a logical, or rather illogical, escalation buffer. During the troubleshooting, both help desk associates were apprehensive about approaching the network administrators. I suspect their hesitation was founded on previous experiences. Nobody wanted to take ownership of the problem.
The communication problems may be worse than they appeared. In the course of my work, I have limited (no) contact with external customers. Yet, I was able to identify the communicative dysfunction after only an hour of working remotely. The IT department was in desperate need of a cultural shake-up. From the outside, it appeared the network administrators sat atop of the IT silo. The organization is large however, so the help desk department, and the network administrators, may have different department managers. It's also possible that the separate IT fiefdoms operate from different buildings. Whatever the case, increased face-time will help establish real communication. It may also be beneficial to allow certain help desk staff to job shadow network administrators on occasion. The experience may be educational as well as build camaraderie. Conversely, it may be worthwhile for network administrators to occasionally man the help desk. This may provide a greater appreciation for each others' contributions. Basic workflow changes within the IT department may have positive reverberations throughout the entire organization.
Communication is the exchange of information.
