Re: SAP BW compared to Essbase/SQLServer/Oracle as a Enterprise Data

"Belinda" <belindacur_at_yahoo.com> wrote in message news:41af5e48.0402190341.5dfae7f4_at_posting.google.com <snip>

>
> Nigel Pendse as published a very interesting article on SAP BW a
> worthwhile article to read if you are already being bled dry by SAP.
>
> Belinda

SAP BW did even worse in The OLAP Survey 3 (http://www.survey.com/olap) than in the previous edition, and much worse than any other product. The strictly-independent OLAP Survey 3 recently surveyed 1047 OLAP-using sites from around the world, so the results have more statistical validity than most surveys.

Compared to nine other BI products with significant representation in the survey sample:

• BW customers were less likely to report that they had derived business benefits (cutting costs, improving reporting, enhancing customer satisfaction, better decision-making, etc) from the project than users of any other product.
• BW users were less likely to achieve their original goals than users of any other product. Uniquely, SAP BW sites were more likely to have achieved no goals at all than to have fully achieved their original goals.
• BW customers were less likely to have carried out a competitive product evaluation before purchasing than any others.
• BW sites had incurred the second-highest license fees; the product with slightly higher license fees had much larger numbers of deployed seats.
• BW sites incurred by far the highest ratio of consulting to license fees.
• BW projects took longer to go live than any others.
• BW sites had the highest shelfware rate (that is, seats purchased but not deployed).
• BW sites have the second-highest rate of reporting technical problems (almost twice as bad as Microsoft Analysis Services sites, who had the lowest problem rate).
• BW sites were the least likely to report that they saw no deterrents to wider deployment.
• BW sites were the second-most likely to complain of poor query performance (two to three times higher than Oracle Express, Microsoft AS and Essbase sit es). Actual query times were about four times longer than these products, despite having comparable data volumes.