<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/2.0.3" -->
<rss version="2.0" 
	xmlns:content="http://purl.org/rss/1.0/modules/content/">
<channel>
	<title>Comments on: Overcoming (some of) the BI barriers</title>
	<link>http://andyonenterprisesoftware.com/2005/11/overcoming-some-of-the-bi-barriers/</link>
	<description>Andy Hayler, noted industry expert and founder of Kalido, gives his view on developments in the enterprise software market. Issues covered include data warehousing, master data management, business intelligence and corporate performance management.</description>
	<pubDate>Wed, 23 Jul 2008 22:39:53 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.0.3</generator>

	<item>
		<title>by: Andy Hayler</title>
		<link>http://andyonenterprisesoftware.com/2005/11/overcoming-some-of-the-bi-barriers/#comment-23</link>
		<pubDate>Wed, 30 Nov 2005 08:36:00 +0000</pubDate>
		<guid>http://andyonenterprisesoftware.com/2005/11/overcoming-some-of-the-bi-barriers/#comment-23</guid>
					<description>Thanks for your comment Rajeev.  I think it is a matter of horses for courses.  The analytic functions of the ERP vendors are well suited to operational reporting from their own ERP systems, so I agree that for operational reporting from SAP, BW makes good sense.  However remember that SAP only covers 30-70% of a company's business model (that is an SAP figure) so you have 30-70% in other systems e.g. marketing systems or whatever.  For certain types of analysis (but not operational reporting) you need to combine ERP data with these other systems, and this is where an enterprise data warehouse makes sense.  Just as you probably should not try and shoe-horn an EDW into an operational reporting role (as in your example) similarly you should not try and shoe-horn an ERP reporting system into a role to which it is not suited, that of an enterprise data warehouse. BW is not well suited to dealing with lots of non-SAP data, especially where the business hierarchies are very complex.  Both types of systems have their place, and can work well together.  This is the case in the majority of our Kalido clients, most of whom are huge SAP users.</description>
		<content:encoded><![CDATA[<p>Thanks for your comment Rajeev.  I think it is a matter of horses for courses.  The analytic functions of the ERP vendors are well suited to operational reporting from their own ERP systems, so I agree that for operational reporting from SAP, BW makes good sense.  However remember that SAP only covers 30-70% of a company&#8217;s business model (that is an SAP figure) so you have 30-70% in other systems e.g. marketing systems or whatever.  For certain types of analysis (but not operational reporting) you need to combine ERP data with these other systems, and this is where an enterprise data warehouse makes sense.  Just as you probably should not try and shoe-horn an EDW into an operational reporting role (as in your example) similarly you should not try and shoe-horn an ERP reporting system into a role to which it is not suited, that of an enterprise data warehouse. BW is not well suited to dealing with lots of non-SAP data, especially where the business hierarchies are very complex.  Both types of systems have their place, and can work well together.  This is the case in the majority of our Kalido clients, most of whom are huge SAP users.
</p>
]]></content:encoded>
				</item>
	<item>
		<title>by: Rajeev</title>
		<link>http://andyonenterprisesoftware.com/2005/11/overcoming-some-of-the-bi-barriers/#comment-22</link>
		<pubDate>Tue, 29 Nov 2005 22:26:00 +0000</pubDate>
		<guid>http://andyonenterprisesoftware.com/2005/11/overcoming-some-of-the-bi-barriers/#comment-22</guid>
					<description>I think points 2 and 3 could be minimized if you go for the integrated product i.e. SAP or any other product for that matter. Coming from SAP background as well as having seen some of the blunders, at the end of the day it does make sense to have one software rather than having multitudes of them.

•  difficulty in getting business intelligence out of enterprise applications (e.g. ERP) – Having the integrated system does take a load off from retrieving the data. Nothing is 100% done but at least there are tools and standard programs which could bring the data into the BI system. 
•  perceived high cost of ownership – Cost of ownership also get reduced as now you have one less system/software to maintain. The percentage of cost is absorbed by avoiding the duplication of work and sharing the same resources.

One of the implementation I worked, the client decided to write the custom BI product which was supposed to take over SAP reporting as well as from other systems. The blueprint looked great but when it came to implementing the design, the complexity of getting the data from SAP was tremendous and the problems attached to it. This resulted into reducing the scope of the BI system and instead of doing selective data retrieval they ended up duplicating SAP tables in their database. So at the current time the ratio of reports coming out from SAP to BI is 10:1. And, moreover they have team of few dozen maintaining that system just in case…</description>
		<content:encoded><![CDATA[<p>I think points 2 and 3 could be minimized if you go for the integrated product i.e. SAP or any other product for that matter. Coming from SAP background as well as having seen some of the blunders, at the end of the day it does make sense to have one software rather than having multitudes of them.</p>
<p>•  difficulty in getting business intelligence out of enterprise applications (e.g. ERP) – Having the integrated system does take a load off from retrieving the data. Nothing is 100% done but at least there are tools and standard programs which could bring the data into the BI system.<br />
•  perceived high cost of ownership – Cost of ownership also get reduced as now you have one less system/software to maintain. The percentage of cost is absorbed by avoiding the duplication of work and sharing the same resources.</p>
<p>One of the implementation I worked, the client decided to write the custom BI product which was supposed to take over SAP reporting as well as from other systems. The blueprint looked great but when it came to implementing the design, the complexity of getting the data from SAP was tremendous and the problems attached to it. This resulted into reducing the scope of the BI system and instead of doing selective data retrieval they ended up duplicating SAP tables in their database. So at the current time the ratio of reports coming out from SAP to BI is 10:1. And, moreover they have team of few dozen maintaining that system just in case…
</p>
]]></content:encoded>
				</item>
</channel>
</rss>
