Skip to Content

Business Intelligence has evolved over time, and has gained more prominence. Even in the current economy, BI projects are generally alive and well. So what is the path forward? Of course, I do not have a crystal ball to gaze and predict. Nor have I conducted a scientific study of any sort. However, what I have done is that I have discussed this with several people in the field, read what others have written on it and have thought about it in some detail. So here is what I think would unfold – for whatever it is worth 🙂

 

1. Stand alone BI will become a thing of past

 

We are already seeing BI becoming an integrated part of business processes – with CRM applications possibly leading the way. I am very certain that this will be the rule and not the exception in the future.

 

2. Speed of analytics will increase manifold

 

Consider SAP BW and the accelerator appliance. The bottleneck now is BW itself, and I am sure more of the OLAP engine will move to BWA pretty soon. So why have this extra appliance? Can’t we have this high speed analytics available in standard BW itself? I used BW as an example – I am expecting all leading BI products to increase speed many times in next few years. Here, the improvements in hardware related technologies is key – processor speeds have consistently improved, blade servers have become common place, and virtualization is not wishful thinking any more. Hard disk is a lot cheaper than in the past. All of these will continue to improve so I am guessing we should see dramatic improvements in medium term.

 

3. One system that does it all

 

By extension of the above, I suspect that we should start seeing more and more of BI moving into OLTP systems. I am not a computer science major, but my impression is that the database theories that we used over past several years that lead us to develop different models for OLTP and OLAP were built for making best use of the available limited processor power and expensive hard disk space. With the advances in hardware, it should lead to some convergence of OLTP and OLAP into one system. Same should hold true for planning applications – instead of periodic and “after the fact” comparisons of plan vs actual data, it should become integrated into all business processes at real time. Planning was once a part of OLTP systems, including SAP. Then it was moved out of it (like the BW based SEM in SAP’s case). My theory is that it will come back and become part of the combined OLTP/OLAP system.

 

Considering the huge investments that have been done on separate OLTP and OLAP systems around us today, I don’t think that this will be the case any time soon. So what would happen in the interim? That is where the next point could make an impact. 

 

4. Federated datawarehouses 

 

This is certainly not a new idea – we have been talking about this for a few years now. However, it hasn’t quite become the rage yet. Why is that? well, interoperability is still wishful thinking for the most part in the BI world. SOA is a potential way out, but then the marketing of SOA ruined it for us. I expect that we have to wait for the hype to die down and let some early adopters take the plunge and show us what is possible. I had an initial confusion in my mind whether “one system that does it all” and “federated datawarehouses” contradict each other. However, I have concluded that they don’t – for a few reasons. For one, even if OLTP and OLAP systems merge, several such systems could still co-exist in a distributed landscape. And if cloud computing does take off in a big way – big advances will happen in federation techniques.

 

5. More analytics will happen on unstructured data

 

The primary reason that analytics needs to become very good in dealing with unstructured data is the gaining popularity of social networks and media and its impact on business. Pretty soon, executives should be asking for dashboards that tell them what is happening in the social media that impacts them. This will probably start in an “after the fact” mode, like most of traditional BI, but pretty soon – I expect this to become very neal real time. Also, it will become imperative that unstructured data can be combined intelligently with structured data.

 

6. Competition amongst vendors will increase, but may be – so will co-operation

 

If federated data warehouses are the future, then I don’t think there will be many generic data warehousing solutions left in the market. Each vendor should then aim to satisfy highly specialized niches of the BI market. But I also expect a lot of co-operation to go along with the competition – since customers will demand that these disparate solutions can some how talk smoothly wih each other to make the federated system work efficiently.

 

7.Visualization will rule

 

Visualization is  serious tool, but I have a serious gripe on how it gets positioned in BI projects. As a result, a lot of executives get an impression that it is not useful, and that it is nothing more than a pretty marketing picture. In future, I also expect BI tools to be able to suggest intelligent visual representations of the data with an ability to switch to different visuals, and the ability to drill down as needed.

 

8.Predictive analytics will turn from half-fiction to reality

 

Traditional analytics presents data to a human decision maker, who then takes some decisions based on it, and some rules and his or her experience. A good part of these decisions can be made by the machine as predictive analysis improves its abilities. Even if we don’t reach sci-fi type scenarios, it is not unrealistic to expect BI to give us options with some pros and cons in near future. This should be of immense use to people like scientists and insurance underwriters and eventually to every process owner. 

 

Before I finish, let me also state that I don’t have unrealistic expectations on how soon all these changes will happen. When I look back at how BI has evolved since I started in the field, technology has improved manifold, but the requirements from business users have not changed that drastically. I still hear requirements from my clients that are very similar to what I heard in my first project in the nineties. People still primarily need summarized information with the ability to slice and dice as needed. They also need canned reports, and very few do their own reports. Charts have not replaced line item data, and most of my users need the ability to use excel spreadsheets in tandem with BI solutions. What has indeed changed is my ability to deliver solutions for these requirements – due to better technology, tool features and the experience I have gained over a period of time.

 

So, what do you folks think? I am eagerly waiting to hear your thoughts and ideas.

To report this post you need to login first.

35 Comments

You must be Logged on to comment or reply to a post.

  1. Tammy Powlas
    I have considered many of your thoughts before, especially BI/BW becoming integrated in daily processes.  It already is using Visual Composer, Information Broadcasting and alerts.  It is a move/trend towards operational BI.  TDWI.org has several white papers on operational BI.

    I hadn’t considered federated warehouses or even moving the appliance to the OLAP engine.  Excellent suggestions.

    It will be interesting to see too how analytics would operate on unstructured data!

    Great job.

    (0) 
    1. Vijay Vijayasankar Post author
      thanks Tammy – glad you liked it.
      I have a friend who is a research guy dealing with unstructured data. I have been begging him to blog on his work – but haven’t quite convinced him yet. I saw some prototype things last month – and gosh it is impressive. 
      (0) 
  2. Witalij Rudnicki
    Hi Vijay,

    Nice summary, indeed. But from my perspective it is already “known next”, i.e. something that is already happening. What is more interesting is what is “next next”, i.e. what is coming that no one is probably working on yet?

    Going back to you original list of things, it will be interesting to see how this will shape our IT work. Adoption of these “next things” might hurt, when business is buying an idea from vendor, and IT is struggling with not mature technology.

    Regards,
    -Vitaliy

    (0) 
    1. Vijay Vijayasankar Post author
      Good point on “next next”, Vitaliy – I don’t think even the known “next” will go easy…as I said at the end of the blog, business requirements have not changed significantly (except maybe the social media analytics), and so “next next” is probably still something that might interest academics at this point. I read some very cool ideas on analytics on genetics and artificial intelligence..but have not seen it being correlated to business scenarios yet.

      In the short term – the only thing I can see gaining momentum is the quest for interoperability gaining momentum.

      (0) 
      1. Witalij Rudnicki
        Last year I thought that Streaming Analytics is “next next”. But last month I read IBM had done already some implementations of own technology delivering Streaming Analytics.
        (0) 
        1. Dirk Herzog
          I’d say that one next next is integrating any kind of AI, fuzzy logic, however you will call it, to find out the reports the user should have a look at combined with a method to generate a proposal of what to do next.

          Dirk

          (0) 
  3. Arun Varadarajan
    Vijay,
    I think the only thing left is educating the customers in terms of what all is possible. I think there is a lot of inertia with regard to customers hearing what is not possible that they have started looking at BI in terms of consolidation first and then let me integrate. This has led to huge systems which consolidate all the data and enable siloed operational reporting … but then the next step is still confusing … both from the capability part and tool part. I guess we need a lot of examples to drive home what is possible and what is not…? Maybe a sort of hosted system where people can collaborate etc and see for themselves what is possible… and also this will gear SIs towards building the skillset within to deploy such a thing where you are looking at a truly connected BI which gives what the business wants and will want in a short period of time.

    Most of the above is something that we all know and some gibberish in between .. but then IMHO it is a matter of time before we get there…?

    My 0.02
    Arun Varadarajan

    (0) 
    1. Vijay Vijayasankar Post author
      Apart from time and education, there is a question of cost too, Arun. Most of this is not very affordable (except the hardware part becoming more affordable) at the moment.

      In my mind – starting small is key here. If we over do the pitch, we will end up similar to where we are with SOA. A certain lack of imagination by IT is also to be blamed I suppose – a large part of IT is stuck in “old world” and won’t look for opportunities to take a step forward and try out new things.

      Business is generally risk averse, and I think they are justified in remaining sceptical till some one shows them without marketing spin that it is not all smoke and mirrors.

      (0) 
  4. Srini Tanikella
    Vijay,

    Great blog and a fantastic topic to discuss. I agree on all the points mentioned – Visualization, for some reason, is not given much importance in BI. I feel it’s probably one of the most important aspects and should not be ignored at all. Visualization plus Usability of the app will go a long way towards improving User Productivity. I will probably sum up my thoughts in another blog, but my question for you is – Most often, you will see the ‘Core Analysis’ still being done in Excel and Access (needless to say that over time, these excel and access databases becomde monster applications). Will the ‘Next’ or ‘Next Next’ Generation BI tools take us away from this pattern?

    Regards,
    Srini

    (0) 
    1. Vijay Vijayasankar Post author
      As much as I would like to say that excel and access based adhoc work is going to go away, I must admit that I don’t see that happening any time soon.

      Consider workbooks in BW. Despite spending lots of time and money on education, most power users still like to keep local copies and work with that.

      I have also encountered power users who have told me “I want every one to come to me for trusted info, and not to some system”.

      A big aspect of reporting is the offline capability. It is not realistic to expect that you will be connected to internet at all times. So most people who have travelling jobs – like in sales and marketing, prefer something like Cognos Powerplay (or something similar) where they can do offline work at high speed, and synchronize with server data periodically.

      (0) 
  5. Srivijaya Gutala
    Regarding OLTP vs. OLAP: there is strong tendency in literature and on the market to emphasize OLTP analytics. Data replication to a warehouse is no longer regarded as sole approach and the focus is shifting towards ‘main memory analytics’. ByDesign for example will not use any OLAP warehouse at all.
    In the Business Suite we also expect a decline of BW importance in future years simply for TCO reasons.
    It is questionable to what extent huge data volume is well handable by OLTP reporting. We are trying to tackle this issue by leveraging ‘operational BIA’ capabilities (following the byDesign example).
    (0) 
    1. Vijay Vijayasankar Post author
      main memory analytics is indeed what most players seem to be focussing for both reporting and planning. About data volumes – that is a valid point. But if disk space is getting cheaper and computing power is increasing, can’t OLTP systems run using something similar to a denormalized OLAP type schema and negate the volume problem? or Maybe additional abstract layers can be built between database and application logic to support OLTP and OLAP distinctly from a development perspective.
      (0) 
      1. Dirk Herzog
        Hello Vijay,
        first of all a great blog. But I disagree with your OLTP/OLAP Thesis. While OLTP will include more and more analytics functions the OLAP engine will do a shift to the unstructured data you mentioned which still will not be integrated into the OLTP. So while the OLTP takes some of the old OLAP functions OLAP takes some new functions and still remains separate.

        Best regards
        Dirk

        (0) 
        1. Vijay Vijayasankar Post author
          Hi Dirk,

          Very interesting point on OLTP and OLAP.
          Let me explain what I had in mind, and pick your brains a little on whether this could happen.

          1. At the lowest level – I expect a common data layer that can store data efficiently irrespective of its usage in OLTP, OLAP, Stream etc.
          2. Above this, I expect a mapping layer that can link several abstract data models to the lowest layer.
          3. Above the mapping layer, I expect a set of abstract data models that suit the application – for example a third normal representation for OLTP(which will save the effort to re-engineer existing code), a star-schema like model for OLAP, and so on.
          4. Above this, I expect everything to continue as it exists. Specifically  – I expect something similar to the SAP CRM BOL/GENIL model (to support MVC and SOA style development) wherever it makes sense.

          I am very interested in finding out what you think.

          Cheers
          Vijay

          (0) 
  6. Anonymous
    Can I suggest 2 more points about the future of BI:

    9) Data Quality and Metadata lineage. This is a recurring pain point for customers running complex DW landscapes, especially where federation is used. While SAP have inherited tools to manage those issues from the BOBJ acquisition, we can anticipate further product developments in the domain of data traceability and reconciliation.

    10) Vertical collaboration. Nothing really new but a trend that will accelerate with cloud computing whereby customers will share BI data from each other. Today content Vendors such as IRI provide market share data to large FCMGs. We can easily extrapolate that large companies will not only consume data and reports from BI content vendors, but also share their BI content between themselves or sell it via collaborative marketplaces.

    (0) 
    1. Vijay Vijayasankar Post author
      on point 9 – most customers already use a data quality tool (some IBM infoserver tools and trillium are what I have encountered the most personally) even when there is no federation. That is why I did not put it in the list. However, from a strict SAP perspective – the BOBJ tools should make a big impact.

      on point 10 – I agree, and this should be a part of “next next” that Vitaliy was talking about. Selling BI data should become common place if clouds become commonplace (maybe even otherwise).

      (0) 
  7. Anonymous
    Hi Vijay!

    Great blog. I have a question about unstructured data. Is it possible to make some kind of analysis over this data without trying to structured it before? I have heard about BO tool Data Integrator so I think maybe in the future this will be a BI specialized niche to be covered. How important do you thinks this will be?

    (0) 
  8. Anonymous
    Hi Vijay!

    Great blog. I have a question about unstructured data. Is it possible to make some kind of analysis over this data without trying to structured it before? I have heard about BO tool Data Integrator so I think maybe in the future this will be a BI specialized niche to be covered. How important do you thinks this will be?

    (0) 
    1. Vijay Vijayasankar Post author
      Hi Karim,

      I did not fully understand the algorithm that my friend in research explained on how this is done – but it certainly sounded like data was not held in DSO/Cube type structures to get useful information out of it. I am still trying to get that dude to write on this topic himself.

      Cheers
      Vijay

      (0) 
  9. Joerg Rohde
    Hi Vijay,

    I enjoyed reading your blog – a good summary on trends and developments.
    As we as consultants (on behalf of our clients) are always interested how to make these things become more of a reality – and I agree with you on your closing comment regarding mostly still basic requirements from customers – it would be interesting to discuss how SAP’s Business Objects Solution will change the process and enable faster / deeper adoption as opposed to BW. I.e. the visualization capabilities are so much better and starting small is easier. Perhaps a subject for another blog….?

    (0) 
    1. Vijay Vijayasankar Post author
      Hi Joerg,

      BOBJ has a fine array of tools that improve visualization compared to old BW tools. Other vendrs like Cognos etc have good tools too. So my belief is that it is not a lack of tools that is preventing visualization from taking a leap – it is a lack of education/awarenss on users, consultants etc that is slowing down the adoption. Once people see tangible benefits, I am sure they will jump in.

      And it might very well be a good topic for my next blog 🙂

      Cheers
      Vijay

      (0) 
    2. Witalij Rudnicki
      It is an interesting topic, but have you ever had a feeling, that people do not see the DATA behind nice visual effects?

      Another aspect is that visualization has a component of data manilulation. My favorite example is that by depending on length of X-axis “Time” comparing to Y e.g. “Revenue” you can present message of “steady growth”, when X-axis on the graph is longer then Y, or “rapid aggresive development” when X is shorter 🙂

      Still, waiting for your blog on this topic.

      Regards,
      -Vitaliy

      (0) 
      1. Vijay Vijayasankar Post author
        if you overdo visuals – the message gets lost or wrongly imlemented. Switching the axesscale is a great example of this – thanks for pointing it out nicely 🙂

        It also depends on your user base. Certain domains like marketing love charts, and certain domains like finance and engineering love numbers. My wish is for a tool to intelligently suggest the best representation of input data, based on roles and data itself.

        (0) 
  10. Rick Bullotta
    Hi, Vijay.

    I must respectfully disagree that most of the items you submit as “future” BI are indeed forward-looking.  In fact, most, if not all, already exist today, some of them in quite mature ways.  They may not be embedded within the “traditional” BI vendors’ offerings, but these approaches are absolutely being implemented to broad success in the market today.

    I think we should look out a bit further and try to start a dialog as to what the real “future” BI solutions will look like!

    I’ll start. 😉

    For one, I think these solutions will be significantly more collaborative.  The ability for people to discuss, search, communicate, comment, and share knowledge around traditional BI representations of information will most definitely increase dramatically.

    Next, I think stronger linkages will exist between BI solutions and the underlying applications that would make the analytics more “actionable”.  This work is starting but is still quite challenging to do with some legacy applications.  Non-actionable or high-latency BI yields the “we had a bad month last month” or “we performed poorly last shift” view of the world, instead of being able to act on deviations as they occur.

    I envision significant improvements in stream-based analytics (BAM++) that function upon the data (raw data or intermediate data) as it is generated by the sources, not necessarily as it is brought into a data warehouse.  These events could be driven from RFID/AutoID, point-of-sale, web activity streams and feeds, real-world event streams and sensors, and many other sources.

    I think we will see BI and related data being used in a much more “closed loop” manner to refine and update enterprise master data (such as assumptions related to costing, resources, time, and so on that are used by planning, logistics and costing algorithms).

    We will also see more model-based analysis that provides a foundation for better predictive analytics.  By better, I mean more accurate, but also, MUCH easier to implement.  Many current predictive analytic approaches require a “rocket scientist” to help in design and deployment.  In a related expansion of functionality, we should see an increase in “proactive” analytics so that individuals need not be at their computer, but can be alerted or notified when tasks specific to their role/function require attention.

    There will be a continued expansion to the modalities by which BI data can be delivered to the consumers, including increased mobile capabilities, 2D and 3D immersive and highly interactive UI’s with gestural drill-down and manipulation, and many others.

    I also think we’ll start to see (finally) some standardization and interoperability, or at least better openness, so that BI solutions and their underlying capabilities (as provider and consumer) can be better composed/utilized with other applications.

    That’s my list – let’s see what you all can come up with!

    Cheers,

    Rick

    (0) 
    1. Vijay Vijayasankar Post author
      hi Rick, most of what you said seem to be the same ideas that are mentioned in the blog – just in different words. Or am I reading your reply wrong? Please help me understand.

      Let me start by comparing your statements in the comment to the statements in the blog.

      You said > these solutions will be significantly more collaborative
      I said > Stand alone BI will become a thing of past, Competition amongst vendors will increase, but may be – so will co-operation

      You said > I think stronger linkages will exist between BI solutions and the underlying applications that would make the analytics more “actionable”
      I said > OLAP and OLTP systems will converge (although I used hardware advances as primary reason for my comment). Plus I said> Even if we don’t reach sci-fi type scenarios, it is not unrealistic to expect BI to give us options with some pros and cons in near future.

      You said> significant improvements in stream-based analytics.
      I said >  More analytics will happen on unstructured data (I used social media as an example, and you used RFID etc)

      You said> We will also see more model-based analysis that provides a foundation for better predictive analytics
      I said> Predictive analytics will turn from half-fiction to reality

      You said> we’ll start to see (finally) some standardization and interoperability.
      I said>  customers will demand that these disparate solutions can some how talk smoothly wih each other to make the federated system work efficiently

      You said> continued expansion to the modalities by which BI data can be delivered to the consumers, including increased mobile capabilities, 2D and 3D immersive and highly interactive UI’s
      I said > Visualization will rule. I also expect BI tools to be able to suggest intelligent visual representations of the data with an ability to switch to different visuals, and the ability to drill down as needed

      So what is the “futuristic” idea in the comment that you are talking about that I missed out on?
      Or are you saying that I did not explain the ideas clearly or give enough scenarios as examples?

      Cheers
      Vijay

      (0) 
      1. Rick Bullotta
        Hi, Vijay.

        Let me elaborate a bit:

        – When I say “collaborative”, I don’t imply collaboration between vendors or products, I am referring to collaboration between people in the context of analytics.  Thus, being able to blog/comment/discuss about analytic information “in context”, being able to view user and role-relevant analytic-related business events in an activity stream (e.g. Facebook, Twitter), and enabling “instant collaboration” (IM, VOIP, video) with relevant experts in the context of the combined BI/line of business applications

        – Linking analytics to actionability is more than simply showing “embedded analytics” in the OLTP system’s UI (or vice versa).  It requires the ability to retain (or dynamically create) contextual links from an exception/deviation back to the underlying processes that could/should be affected by a corrective action.  There’s a lot of R&D to be done to work on the semantic linking (it could perhaps evolve from some of the work in BI linking it with KM).  Though embedded analytics is a good start! 😉

        – Stream processing is very distinct from analysis on unstructured data.  The primary difference is in the temporal nature of the data and the query/in-line analysis mechanisms utilized (often referred to as CEP/complex event processing).   Also, I’d make the case that most social media is actually quite reasonably well structured (e.g. RSS, Atom).  If you want to see a pretty interesting example of applying analytic/search technology to unstructured data, take a look at what Autonomy is doing with it’s IDOL server (others are of course doing similar things).

        – Regarding predictive analytics, I think they are very much reality today.  There are lots of vendors doing this, usually in “niche” markets (forecasting, manufacturing processes, etc.).  The problem is that there needs to be an order-of-magnitude shift in the approaches to modeling and implementation.  I have yet to see the right metaphor for allowing a business user to create those models (or automatic model creation) yet, but am hopeful that it can be realized.  Of course, I’ve been hopeful of this for about 15 years. 😉

        – Also, there is a subtle but important semantic difference in ‘interoperability’ and ‘standards’.  Interoperability is usually driven by vendors/integrators with point solutions and often meets individual user requirements.  Standardization is a painful process (as we’ve seen with the mess known as XML/A) but can ultimately yield a much lower cost of interoperability if consensus can be reached.  Customers really need to lead the call this if there is demand for it, as there’s no real incentive for vendors to do it.

        – Agree totally on visualization.  That’s definitely a cool area for innovation in BI, but I hope we see functional innovation, not just cool “cartoons” on the screen. 😉

        I think on balance we’re in violent agreement – but as I mentioned earlier, we should explore and focus on those areas that aren’t yet as “mainstream” or perhaps even have no implementations yet. 

        You and your colleagues at IBM are doing some VERY interesting things as of late in this area, and I’m very much looking forward to seeing the innovations that result.

        Best regards,

        Rick

        (0) 
        1. Vijay Vijayasankar Post author
          Hi Rick

          Your point on actionability reminds me of an idea I played with, in a project couple of years ago. When a certain CRM transaction went into error, we figured out from history what happened to previous such transactions (presented in graphical format)- and tried to create a followup transaction set(options as a radio button group that a user had to choose) that best had the chance of correcting the error.

          It was an interesting concept and I had some business buy-in too.  But as soon as I tried to go beyond a prototype – I realized that this is a huge effort, and there are many many possibilities that need to be analyzed, and that it is beyond the processing power of BW and ECC as they are designed today.  

          One lesson I did learn from that exercise is to keep a very open mind on tool selection. If I had thought outside the world of SAP tools I am familiar with – I would have probably found a nicer way to solve and implement it.

          Next time I have a use case that calls for some such actionable BI – I will make a stronger attempt to get it past prototype 🙂 Or maybe we can try this out as a community project too. I need to think through this in some more detail.

          Cheers
          Vijay

          (0) 
          1. Witalij Rudnicki
            While discussion about coolnest of software of the future is exiting, I’ll be a broken tape and go back to my blog:
            What should be the Next Big Thing for SAP in BW/BI?
            Some aspects are usualy lost in the pace to produce the fancy stuff or soft that follows current buzz. I won’t stop repeating that besides all the coolnest, tools require stability, completeness of functionality and maintenability (good documentation suppose to be part of deliverables as well). Otherwise customers keep overpaying in implementation and support costs (what I – as a consultant – should not complain about 😉
            Just look at BW Accelerator. Everything is fantastic, when it works. But please try undertand what went wrong and how to fix it when you get error code 5528 (Unable to reorganize BIA with return code 5528 error).

            Regards,
            -Vitaliy

            (0) 
            1. Vijay Vijayasankar Post author
              I am with you, Vitaliy. Today – the primary BI vendors like Cognos, Business objects, microstrategy etc are all very similar in capabilities (agreed – they have a handful of differentiators). In this market, how do the vendors differentiate themselves from the pack? They have to keep on “innovating” and this sometimes leads to fluff. A lot of buying decisions are made on either price or on familiarity.Not always will companies conduct a thorough tool evaluation based on actual requirements. This plays to the vendors’ strength too. As always – the prevailing thought it “buyer beware” .
              (0) 
  11. Atul Kulkarni
    Vijay,
    One of the ways I judge success of what you have written in way of a blog is the number and wide variety (Pro and Con) responses.

    You have been successful in grabbing a LOT of attention!! Full credit to your thought stimulating ideas all of which make sense.

    I think embedded BI and collapsing of OLTP/OLAP engines are two areas that are becoming a reality in very near time due to the new technology. However anybody who has worked on BI and backend ERP can tell how difficult it is to embed “Useful” BI analytics in the business context.

    This is partly due to the business users really not knowing what to ask for and that is a big problem.

    I would like to see a much easier way to get data from ERP systems to a BI system (Whether BI is separate from ERP or in the same box). Currently just getting the data there is so exhausting that we lose steam by the time we get to actually deriving any value of the BI data.

    regards,

    Atul

    (0) 
    1. Vijay Vijayasankar Post author
      Thanks Atul.
      I completely agree with your comment – I have seen several times that people get all caught up in datamodels and transformations, that by the time they finish – there is very little good BI left for users.

      On the second point – the sheer difficulty of having data synched between 2 systems is a good reason why this practice should stop and have a
      common system for both. Atleast that is what i wish for 🙂

      (0) 
  12. Rajesh K
    Hi Vijay,

       Very engaging blog.. I enjoyed your earlier blog on ‘Titles’ as well(being an IBM’er myself helped. Plus you articulated things that most would have experienced at some point of time themselves but didn’t think at length on).

       Now coming to the current blog.  You’ve been spot-on with your insights into the future of BI. However, it would have been nice if you also covered the tools that will help achieve these. For example, can Business Objects be ignored when the BI future is being discussed?? Whats the future of BO??. Just a thought..I understand that discussing the ‘tools’ was not the purpose of the blog anyway.

      I also couln’t help think that the BI future is interesting AND rather expensive. That the blade servers(used in BWA) can be like a ‘blade’ on ones pocket is not news for most:). Of course, hard disk, RAM prices fall by the day but they still are a privilege say for the SMB’s, aren’t they ?
    One system that does it all – Business objects with its ‘Universe’ and all that mumbo-jumbo concepts seems to be doing/trying to do that job just fine even though BO licenses come with the premium tag!!

      What I am coming to is that even though most companies do have their wish list of nice-to haves, and functionality that they would like to see working that is otherwise seen at the moment as ‘restrictions’ in BI/BW, but they could stop short of getting the tools because of pricing implications( at least the smaller companies) and learn to live with the basic products .
      
    What could delay the full blown usage of cutting edge BI products is the cost-benefit advantage and compatibility with existing tools (i.e stableness of the product). There may be quite a few ‘ifs’ and ‘buts’ in our way till then.

    BR,
    Rajesh

    (0) 
    1. Vijay Vijayasankar Post author
      Thanks Rajesh – glad you liked the blog.

      Now on tools – as you correctly interpreted, this was not my key focus here. it is a very important topic though, and if my crazy project lets up at some point – I would love to blog on my thoughts on tools.

      I agree on the cost factor – blades are not cheap at the moment. But I will try to put forward an argument that if OLTP and OLAP functionality converges, this should hopefully reduce storage space considerably than having 2 separate systems. Plus, cost has been almost steadily decreasing over time.

      Cost of licensing on software side is an excellent point. And cost can influence the tool choice even if this is not the best tool for solving the problem. Open source and cloud computing might some day help reduce this. On the other hand, I have a fear that cloud computing will also get people stuck with a vendor and hence the vendor’s tools for a long time.

      Cheers
      Vijay

      (0) 

Leave a Reply