I recently met with NetEvidence, a company that provides insights into network and application performance. It provides the service through a cloud offering, and allows areas such as the root cause of poor performance to be rapidly identified; it can then integrate into help desk systems such as Remedy to kick off remedial action.
NetEvidence provides a role-based dashboard view that can be tuned to the requirements of different people in an organization — for example, a network engineer can focus in on what is happening at the network level, whereas a data center person can look more at the servers. Business people can also be included, having views that are more targeted at them — maybe showing how the ecommerce site is performing, or the ERP system.
The dashboard works on a traffic light system — green when things are OK, amber when there may be a problem, and red is, “oops, better do something here”. This red/amber/green (or RAG) is used by many vendors in different areas and gives a quick and easy visual means to see how things are going. But let’s take an example where it may not be the best way to do things:
Imagine a scenario where the monitoring service is cloud-based and the main data center is connected to the business manager’s office via a leased line. The leased line goes down, but the general access to the internet from the business person’s office is still there. Therefore, the monitoring service is still active, but the data center is not available. In this situation, the business person’s dashboard is a sea of red — no ecommerce availability, no ERP, no SFA, no CRM — all of these are provided via the data center and they are not accessible.
Picture for yourself what the considered response from said business person will be. A case of, “Oh, never mind. I’m sure it will all get sorted”? More like a very close imitation of a headless chicken as panic sets in, as pictures of money going down drains flits through his or her mind. Without feedback loops, what else can be expected from a person faced with a red mist?
How about if there were a capability for the techie people to input comments against what the real problem is and what is being done? So, against the red indicators would be a single textual box (maybe yellow to make it stand out, but this is possibly just a little bit too much detail for here) with, “This is due to a fault with connectivity to the data center. Expected time to resolution 15 minutes. Entered by Engineer X, extension 123”.
End of flap. Head back in place. The business person can now toddle off and get a cup of coffee and a quick doughnut knowing that everything is under control — or at least that it is in the process of being fixed.
The problem is that many of these RAG systems assume too much. They expect the person looking at the system to be able to drill down through multiple levels until they get to the actual problem — and then understand it — and this just won’t happen. If the business person drilled down here and got to something that said, “Problem with Leased Line. No response to Ping. Packet loss 100%”, it is unlikely that he would have any greater understanding than when he first looked at the dashboard, and he may well be even more confused than he was before.
Systems that are meant to be multifaceted in dealing with different roles within an organization have to be able to talk consistently to the people in their own language. This may mean that they should be actively stopped from drilling down too far.
Image credit: WikiMedia Commons