Sunday, August 31, 2008

Asset registers

I received a wonderfully snarky comment last week to a post where I mentioned "spoon feeding" customer service reps at Energy Australia.

It's not spoon feeding moron! Their website asks for the address details. You've read it haven't you?

Ever thought the additional information is asked incase someone gives the wrong asset number. Having a asset register is all good, but not much help when the wrong number is provided. The address provides a means of checking the asset number provided matches the physical location. 

When I said that I would have to spoon feed the customer service people, I wasn't trying to put them down.  Rather, my comment was directed at their management, because those reps are operating without having information at their fingertips that could make their life a lot easier.

I ran a help desk, or service desk (call it what you like - "call centre" if you have to) for a few years, so I know what it is like to be on the end of a telephone, taking a call from someone that might be angry, disgruntled, hard to understand or just plain nuts.  Yes, I took calls myself, usually taking the first calls of the day at 7am for up to an hour or so, and dipping in again during the day if my team was being overwhelmed with calls.  

The whole idea of a call centre or help desk is for it to be a conduit for information to flow into the organisation from the outside.  The trick to keeping call times down (and thus costs down) is to gather the required information in the shortest time possible.  You work out what information you need to process a particular task, then work out ways to get that information from the caller and into your computer system as quickly as possible.

I put CTI into our help desk system as quickly as I could.  Energy Australia have done that as well - for many types of calls or web queries, you start by entering your customer number.  Since that customer number is linked to lots of useful information that they have about the caller (such as your name and address and billing history), it saves time.  The agent answering the phone does not even have to commence the call by saying, "Can I start with your name please?", and then go through the sometimes laborious process of having it spelt out over a crackling phone line.  

The system I put in was an internal one, so we set it up based on employee number - when you called our help desk, the first thing it asked you for was your employee number, and when the agent answered the call, a screen had already popped up in front of them with basic information from the HR system - like their name and location.

Because we were doing IT support, we also added a feature where another system was queried (based on the callers employee number) and it told the agent what computer they were sitting in front of.  This greatly cut call times, since the agent didn't have to ask the user to find their computer name or IP address - something that could take minutes in some cases.  

In short, we did our best to put as much information in front of the call centre operators so that all the "grunt work" of call taking was eliminated.  They could then concentrate on dealing with whatever issue the caller was calling about.  It also makes it as painless as possible for the person making the call, as they don't have to go through the rigmarole of identifying themselves and their location and the computer they are using etc etc.

I also tried to integrate another system into the mix so that if necessary, the agent could call up a floor plan and see exactly what desk the caller was sitting at.  This was really useful information if a technician had to be sent out to deal with a hardware problem - we had thousands of desks in dozens of buildings, and it sped things up a lot if the technician didn't have to spend 10 minutes trying to find the caller.  The software to do that wasn't quite up to scratch, so we had to drop that function, but again the idea was to surround the person taking the call with as much information as possible - and to present it to the agent without them having to open up this program or that program to search for it.  The system used one key bit of data - the employee number - and then did a whole bunch of queries on a number of systems and just plopped lots of useful stuff in front of the agent.  It also plopped that information in front of the agent before they even answered the call.

Now I also worked with a team that implemented a fairly large GIS system (based on Smallworld), and that team included engineers from various disciplines who wanted the GIS to be a portal to all sorts of information about the assets they were managing.

For instance, if you went into the GIS and clicked on a bridge, the GIS would give you the exact location of that bridge.  We had overhead imagery, so you could see what it looked like from above.  The GIS was connected to the asset management and maintenance management modules of our ERP system, so by clicking on various links, it could provide you with information about what the bridge cost to build, all the maintenance orders that had been raised on it for the last 10 years, inspection reports and so on.  It was also linked into a document management system, so you could pull up plans and diagrams and drawings of the bridge, and any reports or memos written about it.  

The reason I have been trying to geocode photos of vandalised assets is that at one point, we were trying to put together a system where engineers could easily upload photos that they had taken of assets and link them into the GIS.  We never got that far, but it's interesting that five years later, the functionality we were trying to achieve is now available via Picasa.  The idea was that you could click on the bridge, and then pull up photos taken from various angles, plus photos showing the bridge over time.  This idea was based on a GIS that Defence had put together for northern Australia.  The Army had a slightly different perspective - when you clicked on a bridge, it might tell you whether it was strong enough to support tanks, or where to place explosive charges if you needed to blow up the bridge.

Everything in our GIS was driven by asset numbers, as every asset that the company managed had a unique asset number.  If someone told you they wanted to check on bridge XB741, you'd type that into the asset query window, and it would return everything you ever wanted to know about that bridge, and more.  

I'm used to working in a world of asset numbers, which is why I thought Energy Australia would be the same.  I understand that they use the same ERP system as the company I worked at, and also had all their assets mapped in a Smallworld GIS.  Given that Smallworld came out with a web interface a few years ago, I wouldn't have thought it was too difficult for the customer service reps to be given a 10 minute demo in how to use it.  Our technicians used it from time to time to find sites that weren't listed on a road map.

So if you ever take 5 minutes to browse through all the photos that I have taken of RTA and Energy Australia assets, you'll see that wherever possible, I have included a shot of the asset tag.  With substations, the tag is only applied to one surface, so if I have taken front and back shots, only one will show the tag.

From my perspective, given the work that I've done with ERP and GIS systems, I assume that the asset tag is the best way to describe an asset.  You use the tag to retrieve the street address, or a latitude and longitude if the asset is not located near a public road (I also worked with a lot of assets that were way off the beaten track - a street address was useless when the instructions on reaching it were "drive west from Bathurst, and 2 kilometres past the 50km marker, turn right onto a gravel track at the large rock.")

But if Energy Australia don't want to do it that way, that's fine.  I'll take the extra time to record the street address of every vandalised substation.  I figure that I am doing them a favour by giving them information on their assets, so they should make it as easy as possible for me to do so, but if that's too hard, I am prepared to put in a bit more effort.  It just means that I have to photograph the damage, then record a video (for my own use) describing the location, such as the street number, street name and cross street (if applicable).  

Given that I have uploaded the photos of the vandalised assets, and taken the trouble to include the asset tag of each in the photos, I would have thought that would be sufficient validation for an Energy Australia employee to work out the address of the asset.  

I'm going to stop here before this post gets any longer.

No comments: