A Work In Progress

Wednesday, March 27, 2002

Time to build a case against Crystal Reports. My current client seems to think that they need a reporting tool, even though we've built a completely interactive reporting solution. They've seen cool Crystal demos, have been promised a number of things, and, actually, already own the product. On the surface, then, it would make complete sense to try to integrate it into our web-based EIS system.

Dig a bit deeper, though, and it turns out that they (or at least one key individual) almost see Crystal as a silver bullet. Just plug it in and it works. Here are a few key points I'll need to make:

If you're building a web-based EIS system, you'll still need to wrap code around your Crystal report solution (either .ASP or .CSP)
To maintain a hybrid solution, you need to locate and hire Crystal consultants. This client has already tried and failed with two different consultants, in a matter of weeks. What happens if their third try doesn't work out?
Because this application needs to be FAST, the reports do not connect directly against the data warehouse. For offline reporting, there will either be a ton of little files for the different combinations of drilling and filtering, or a handful of huge files. Either way, it's a maintenance and/or performance nightmare.
Crystal is great at nicely formatted reports, which isn't a priority for this system. As far as I understand, it's tough to freeze column headings in Crystal (you wouldn't even need to, since it's mostly a print-preview type of deal anyway)

Basically, it's a bird in hand is worth two in the bush discussion. We promised a rapid, low-cost, highly scalable solution. We delivered on that promise, and have executive buy-in. Now it's just a matter of convincing an important technical resource that they've got a working solution, not a "Crystal prototype."

comments