Tuesday, September 24, 2002

What I Do


I know a lot of you have NO clue what I really do for a living. So I thought I'd share this big project I'm working on.

Our company develops, sells and supports Human Resources and Payroll for mid-sized companies. This is no QuickBooks. Anyway, we use CRM (customer resource management) and help desk software to keep everything in line. We used to keep all this data stored and accessed from an IBM AS/400 mid-frame system. A mid-frame isn't quite a main-frame, but it's more powerful than your basic Windows server. These machines are specifically good with databases.

They system was fast and strong, but the software had not been supported for a couple years and custom support was costing us a great deal. We had big ideas for our data, so we decided to upgrade. I say we, but basically the CEO and VP of Sales made this choice on their own.

I knew something was going on, but at the time I was still pretty new to the company and didn't have the clout to push my way into it. I wish I could have.

The sales person/vendor REALLY over sold the products. He also over sold his abilities. He did the data conversion for us and screwed up a number of times.

Data conversion is necessary because one application will treat data in different ways. For instance. Lets say the original application stored phone numbers with no special punctuation, ie. 5551234567, but the new application wants all phone numbers to be stored as (555) 123-4567. Those extra characters have to be inserted at the time of import. It only gets more complicated!

So we lost data here and there as this guy converted. I can't honestly say I would have done any better, but it was poorly planned and executed. This was the case through out the whole project.

I was not brought on-board until it was about 95% done and then I was expected to maintain the system. Well since it was such a poorly designed system I was busy for nearly 6-9 months! Things have gotten to the point where no new bugs crop up, but we have a slew of bugs that we cannot fix and the developer can't address.

Why cant the address them? Well that's the other problem. Because of our “needs” a great deal of customization was done to the system. Most of those customizations were done to attempt to emulate the old system as much as possible.

That brings us to now. The developer of the product we are using now, was bought out shortly after we came on board with them. They continued to develop and support the product. However, the parent company also bought another competing product that is much much better.

So far this product, out of the box looks like it'll give us about 90-95% of our desired functionality without any customizations! But I'm getting ahead of myself.

I have been looking at our systems and processes and seeing a great number of inefficiencies. It's my job as a systems engineer to find solutions to problems. This was a significant problem I saw. Not only was data being entered in half a dozen locations, but important information was not being shared. This has caused some anger because some individuals think others are hording info. Now I will not say they haven't been hording, but the current system doesn't have the necessary checks and balances to make certain responsible parties are responsible. The systems currently in place are perfect for pointing fingers and shifting blame. Obviously a problem.

So I started researching applications that would give us the tools to fix some of these problems. Granted you cannot fix bad processes with technology. It just doesn't work. However, I have a plan of attack that will change those processes.

Because the current product had so many problems because of customizations, I have decided that we should implement this new CRM with as few custom mods as possible. The new company President agrees. We also have a new Sales Manager. Fresh blood, new ideas and flexibility. They bring some powerful tools to the table. Are the managers who have been with the company awhile going to like some of these changes? Probably not, but if they are for the good of the company, then it's the right thing to do. Sometimes its the right thing to stick with how it's always been done, but in this case it isn't.

The next phase is to work on the data conversion. Needless to say this is THE crucial point. There can be no screw-ups here. But I have plenty of time. I have forcasted 9-10 months for this project.

Through the Fall I will work on the data conversion.

Through the Winter I will work with departmental “super-users” who will be key in creating the new processes and data flows. They will also help design the interface to fit our needs. (note: this isn't considered a customization as the application is designed to be modified in this manner.)

Through the Spring I will work on training and any final testing. This would be the time we may or may not decide to purchase a new server for the system.

Finally sometime in the early part of next Summer we'll go live with the new system. The old system will remain alive for a short period of time, but in a read only mode. This is only a precaution, but should not be needed at all.

I'm taking my time with this because of the problems with the last implementation. It went live during the 4th quarter. For our sales team that was a big problem. Prospects they had been working with for months were difficult to find (because they were unfamiliar with the system) and important notes or information was lost or wrong.

If I can pull this off seamlessly, with no data loss and good training, then I'll be the hero who fixed a MAJOR part of the way our company runs.

It wasn't my intention to put myself in this position, but this is a career make or break project here. This is the biggest responsibility that I've ever undertaken, from a business point of view. I'm scared, but more excited. I'm maturing not only as a person but as a business person.

I didn't mention much about the support team. They like this system because they will be sharing a database with the sales team, so communication will be tremendous. Currently there are a lot of phone calls and e-mails sent back and forth for information. Stuff falls through the cracks. I'm gonna fill those cracks.

I covered this quickly, but without going into a 100 page dissertation on the project, I couldn't cover everything. There are many little branches to this project. It will touch every department and almost every employee.

Wow!

Specific information was left out incase the big bosses would prefer this info not be too public.

No comments: