How to decide which database?

[ceh4702]

I work for an educational institution and we have to pick one vendor out of three and decide what platform, database, and application to use for our community college. We service an average of 6,000 to 8,000 students, and manage all the budget, purchase orders, Human Resources, bills, fees, etc.

We have 3 vendors of products we are looking at:

1. Peoplesoft
2. Datatel with a Unidata Database (IBM)
3. SCT Banner

All provide the same type of products. Has anyone had to deal with any of these products?

If you do not like my post feel free to point out your opinion or my errors.

 

[harebrain]

You're not evaluating databases: you have requirements for a specific set of packaged ERP applications. This is really not the appropriate forum for your question.

Not that you won't get any opinions.

 

[SQLSister]

Let me take a go at this.

First, you need to determine exactly what tasks your program must do and what are nice to haves.
Decide what reports yo need to have.
Make this up into a checklist with each of the three programs listed. Assign a weighting factor to each item. (1 not important- 5 cannot live without). Make sure you add some items like ease of use that are not sepcific to the features, but to the usuability of the product.

Next do you need to be able to access the data in conjunction with data from other sources or query it independently, then make sure the database either has the same backend as the database you want to use it with or that the backend has a an obdc driver, so the other database can easily read or import, export records.

Invite each vendor to come give a demo of the product. Include the people who will be doing data entry and the managers who will want to analyze data inthe meeting. Give every one a copy of the checklist and have them mark it up based on the demo. (I usually evaluate numerically, 0 for the feature doesn't exist and up to 5 for the feature is really well done.) If the program has features you didn't think of add them to the check list and give them a rating.
ANy features which you want, that the demo doesn;t mention, bring up at the end of the demo and ask if they do it or how it could be accomplished. Rank the answers. Allow the raters to revise earlier demo scors based n what they saw at laater demos (that feature that looked great inthe first demo may be blown out the water by how the third demo does it.)

After all the demos, do the math. Average the ratings of all evaluators for each itam and program. Mutiply the average rating by the weighting factor for each item and each product. Then add up all these totals for each product to get the total. The one with the highest total is the one which best suits your needs.

So for example you might have:

Item Wt Program 1 Program 2 PRogram 3
ease of use 4 2 3 5
MOnthy summary report 5 1 0 3
Ability to add fields 3 5 4 2
Track employee time 2 5 4 3

Totals Program 1 Program 2 Program 3
Ease 8 12 20
Summary 5 0 15
fields 15 12 6
Time 10 8 6
Total 38 32 47

Here program 3 is the clear winner. You might also have not bothered to calculate program 2 because it didn;t have one of your must have features. (Of course it may turn out that none of the programs have all your must have features, then you need to go back and decide which ones you can live without or get a differnt program.)

 

[ceh4702]

Thanks for the input I have sat through several presentations of these products. I am kind of leaning toward the Datatel System with the IBM Unidata Database. I was looking for a general Application forum for Database Systems but they did not seem to have one. I guess my main reason is they have been in business for about 30 years and they only focus on Education Application Systems.

I think my biggest problem is I did not see enough of the actual products to base my decision on true logic. I will try to investigate further. I did find out that according to IBM that their database is used by 75% or all larger businesses that use databases. Since I only make recommendations on the technical side, and I am a junior programmer/Application Developer they may just look at what I say with a grain of salt and move on to the next one.


I will look at some other forums. Sorry for bugging you too much.

Thanks a lot for your input. I will try to suggest we talk to some of their clients and see what they dont show you at the presentations.

If you do not like my post feel free to point out your opinion or my errors.

 

[SQLSister]

A lot of vendors will give you a version of their software to play with that is limited to only a hundred records or so or have a demo set up on the web that you can play with.

You might ask to see what you can get. Being able to talk to existing customers would be a huge plus too.

And if their demos don't show you everything you need to know, ask for another one. Especially ask to have a technical guy not just the sales guy there. Often the sales people don't really know the software well enough to demonstrate it well. They just know their canned pitch which is designed generally to de-emphasize whatever weaknesses the program has. So if they don't demo it, likely the feature isn't there or is poorly implemented.

You might be surprised at how much influence you can have on this process. Usually everybody is stumped as to how to evaluate the product and they are glad to have someone offer a process. When I was a very junior civil servant I got whole Navy policies changed by the simple expedient of giving them a detailed alternative and pushing it up the line til I got to the decision maker. I ended up writing the whole policy manual (in two different organizations!) If everybody knows something needs to be done and no one else has stepped up to take responsibility for doing it, very junior people can get their ideas implemented. Plus they put you on this evaluation for a reason, likely they will want to listen to you.

 

[ceh4702]

This is a follow-up on this thread. After talking with my boss I found out that he too felt that we do not know enough information about these vendors. In fact he plans on hiring someone to evealuate the three vendors who has had prior experience with them. They know how to right proper requirements documentation and they have had experience installing these vendors products in the past. It seems that he was a little unimpressed with the lack of information that they were forthcoming with.

I think one of the biggest problems is that a salesman can show you what he wants you to see and knows what not to show you. Unless you know beforehand the problems others have had with their product you do not know what to ask questions about. I love it when salesmen say something like "Yes our product can do that."

If you do not like my post feel free to point out your opinion or my errors.

 

[ChrisHunt]

IBM Unidata Database ... is used by 75% of all larger businesses that use databases"

Really? I've never heard of it. I doubt whether any database has that degree of market dominance, still less do I believe that IBM make one that does. Out of interest

http://www.google.com/search?q=unidata+database

returns 17,500 hits

http://www.google.com/search?q=oracle+database

returns 3,730,000

Deeply unscientific, I know, but I know which one I would pick out as the major player in database land.

-- Chris Hunt

http://www.mcgonagall-online.org.uk

http://www.napitalia.org.uk

http://www.leicesteryha.org.uk

 

[ceh4702]

When I combined DB2, Universal, and Unidata together it comes up with about 3,000,000 Hits. I know that doesnt tell the whole story. For instance how many hits represent a small standalone company and how many are Hits from large corporations representing the Larger database installations.
The other thing is how many of these hits are for Problems with the product. This could just mean Oracle is just a bigger headache with more problems.

So we hired someone from the outside to investigate this further to give us an independent recommendation. You can never tell what the sales people are misleading you on.

If you do not like my post feel free to point out your opinion or my errors.