ReadWriteStart

SAP Embraces Real-Time in the Enterprise

Written by Steven Walling / August 20, 2009 1:51 PM / 5 Comments

10_sap_logo.jpgNot long ago John Schwarz, an executive board member at SAP, sat down for a video interview about the current state and future goals of the enterprise software giant's popular BusinessObjects BI and analytics platform, which Schwarz is in charge of.

The discussion was wide-ranging, but a key statement buried in all the talk about BusinessObjects was an admission that real-time access to analytics was vital for the future of the enterprise. If the future of the consumer Web is real-time streams of information and communication channels, then what will keep businesses caught up will be tools to analyze and iterate just as quickly. For an absolute leviathan in the B2B space, this is a huge acknowledgment.

SAP's Top Priorities

Dennis Howlett conducted the original interview, and he brought up some great points about SAP's approach, such as how frustrating an analytics focus on spreadsheets can be for end users. He also kindly posted the video on YouTube, which you can watch in full below.

The first part is largely a recap of what BusinessObjects is all about, so if you're familiar with the platform then it's nothing that special. But later, he outlines the three top priorities for the future of SAP's analytics and optimization group. In general, he recognized that:

Customers have implemented and automated their core processes and have done a very good job in making that happen. They've created enormous amounts of data that is generated by these automated processes.
We now need to help them to use this information to understand how their processes are executing, to do a better job planning and optimizing the processes based on the understanding they have gained from the analysis that they have done.

Real-Time is the Future of Analytics & Optimization

In addition to a general need for stronger analysis of the loads of data being generated in the enterprise, the industry needs to do a better job of providing real-time analysis capabilities. Schwarz outlined the problem:

Customers already do that today with let's say Neilsen data, or IRI data. But these analyses are not real time, and they are only available in massively undifferentiated databases. We've got to bring tools to the customer that allow them to take their product marketing strategies and the retail response to those marketing strategies and make those available real time so that the analysis can be done instantly and that the customer can iterate and do some predictive work on the information that they have.

As Jennifer Martinez at GigaOM has reminded us, real-time isn't actually something new. The hurdle has always been that we don't have what we need to properly filter the constant stream of information.

In the enterprise, that means that big-time software companies like SAP are going to have to ensure that their platforms for analyzing the data are real-time too. Automating business processes is one win, but it's not going to matter a whit if users can't get the data they need to continue innovating at a faster pace.

Thanks to Tim Elliot of SAP for the transcription.


Comments

Subscribe to comments for this post OR Subscribe to comments for all ReadWriteEnterprise posts

  1. John is a smart guy, but SAP's license sales are down 74% Q/Q while Oracle is up 15% Q/Q. So long term you have to wonder who will own SAP.

    Also, while it's great to have "visual dashboards" (I have one for my department) at the end of the day it ends up in Excel and you gotta crunch the numbers. So the eyecandy may sell, but it doesn't do the job - and everyone in the business of business knows that.

    -OT

    Posted by: OT | August 21, 2009 6:06 AM



  2. What I find interesting with real-time analytic capabilities is that the enterprises will need to enforce the discipline for real-time team data input. For example, an in who manually change process status from, say, "not started" to "in progress" to "completed" would need to do so when they actually do the work rather than waiting until Friday to update their statuses. Otherwise, the real-time analysis will not be accurate. I see real-time data opportunities becoming a push for improving technology simplicity. These means technology user experience that doesn't get in the way of daily work.

    Posted by: Paul Dandurand Author Profile Page | August 21, 2009 6:09 AM



  3. In light of real-time analytics, the next logical step is automated decisioning based on real-time predictive analytics.

    Humans cannot keep up with the amount of data collected in real-time today, so our only chance is to turn the data into knowledge and then put a comprehensive enterprise decision management strategy in place to act upon the knowledge. Predictive analytics which focuses on the implicit knowledge hidden in your data will play a major role in the implementation of automated intelligent decisions that can drive almost any business process.

    A recent industry panel on open standards and cloud computing for data mining touches on the real-time integration of predictive analytics.
    http://adapasupport.zementis.com/2009/07/kdd-2009-panel-report-open-standards.html


    Posted by: Michael Zeller | August 21, 2009 10:04 AM



  4. I agree completely here. Analysis in the search arena will require a consideration of context and the implications of the search. I also agree that basing search on social constructs is interesting but as of yet, not proven to be of value in any sense. Great, a link! Is it a valuable link or a dead link?

    Posted by: aa batterien | December 27, 2009 10:13 PM



  5. Thank you for your sharing.!

    Posted by: unutulmaz | February 3, 2010 3:12 AM



Leave a comment

Optional: Sign in with Connect Facebook   Sign in with Twitter Twitter   Sign in with OpenID OpenID  |  
RWW SPONSORS