LinkedIn Twitter Facebook

Ben Kepes is a technology evangelist, an investor, a commentator and a business adviser. His business interests include a diverse range of industries from manufacturing to property to technology. As a technology commentator he has a broad presence both in the traditional media and extensively online. Ben covers the convergence of technology, mobile, ubiquity and agility, all enabled by the Cloud. His areas of interest extend to enterprise software, software integration, financial/accounting software, platforms and infrastructure as well as articulating technology simply for everyday users.

More about Ben here.

6 responses to “GoodData – More of the Same, Dressed Up Like New?”

  1. Sam Boonin

    Thank you for the writeup, Ben. It’s a good discussion. I agree with at least half of what you’ve written :-).

    Yes, GoodData is a hosted data warehouse, which has a set of advantages (historical trending, faster pivoting & drill-down, working with multiple data sets) as well as its disadvantages (not real time, introduces some complexity). One BI disadvantage we’ve removed is the cost – you can see our pricing on our website. Not sure I agree that a data warehousing approach inevitably leads to obsolete data, but that’s an endless debate.

    Yes, GoodData takes what you call a traditional approach to BI. Just like salesforce.com takes a traditional approach to CRM and Amazon Web Services takes a traditional approach to hosting.

    However we are not aimed at “huge enterprises with terabytes of data” (not yet, at least), but instead at the hoards of companies that do not have any viable BI solution today. Two BI statistics stand out (sorry I don’t have the source, but they are generally accepted as the norm) – 9% adoption and 65% project failure rates. You can see there’s plenty of room for improvement there, regardless of the approach.

    You’re welcome to try GoodData, you can even use it for free forever for simple use cases. Or take a look at what some of our customers are saying at http://www.gooddata.com/customers.

    Thanks
    -Sam

  2. Ben Kepes

    Sam – cheers for the reply. More than happy to have a chat for a more in-depth view of what you’re doing

  3. Roma Stanek

    It is very naive to assume that arbitrary data source supports direct queries. I guess Google wouldn’t need to build a search index. They could directly query the web every time someone searches for a word: on fly search. It is obvious that it would be not efficient and extremely slow.

    Clickstreams, logs, transactional databases and other data sources are not designed for complex ad hoc queries and that is why we need to frequently copy data and store them in a form that supports fast response time. Real time analysis will always be only “near real time”…

    -Roman Stanek, Founder, Good Data

  4. Ben Kepes

    @Roman – Thanks for your comment – I’d love to moderate a shoot out between yourselves and YouCalc to really test both sides hypotheses about this… you keen?

  5. Rasmus

    Seems there is already a shoot out taking place in here. (But I’m always willing to participate in further discussions if anyone is interested).

    I obviously tend to agree with Ben, although the truth is probably, as almost always, somewhere in between.

    I agree that what GoodData offers is not new compared to what on-premise BI tools offer – the only clear difference is that is takes place in the cloud.

    Why move your data to the cloud to analyze it? According to GoodData because you then only pay for computing power when you use it. And I guess making it easier to share the job of designing the analysis and sharing the result (although I believe most on-premise BI platforms offer that as well)

    One of the disadvantages is that you have to upload all your data to GoodData before you can analyze it, and then obviously you need to keep it synchronized. Once its uploaded, you can analyze you data fast and easily (as you could do with an on-premise BI tool).

    The fact that GoodData is not radically new/different compared to existing on-premise BI tools does not means that is does not offer value.

    youcalc is not only cloud based (i.e you only pay for the computing you use, easy to share, etc.). We also have a radically different and new approach to analytics. We don’t ask users to upload their data to us, instead we query the data source that stores the data (incl. querying multiple data sources) and then analyze in-memory – this is all done real-time.

    We save users the hassle of having to upload and synchronize data. Especially for users of SaaS systems (salesforce, Omniture, etc.) this gives a lot of meaning. Why move your data from one cloud system to another when you can analyze it where it is?

    Also, we offer an extreme amount of analytical flexibility, offering our users a combination of not only BI tools (pivots, sorting, filters, queries, etc., but also custom analytics and KPI modeling using a spreadsheet type modeling environment. GoodData and the other SaaS BI vendors offer the classical pivoting, sorting, filtering, etc. but no custom modeling.

    The value of our approach as opposed to that of GoodData (and all the other SaaS BI vendors who all have the same approach) will depend on the customer needs and situation. I am sure there is a market for both, but I am confident that our low tech, easy adoption, high flexibility approach is currently what meets most SaaS user needs.

    Rasmus
    CEO/founder, youcalc

  6. Elad

    What you call “real-time” or “direct” queries is usually just a tool for vendors who support it to feature-bash vendors who don’t. In reality, in most cases the ability to execute queries over ‘operational’ sources directly is highly overrated, not required and in fact introduces more problems than it solves. Cloud BI vendors sometimes often forget that most sources to be queries are on-premises operational sources, and no one in his right mind would execute direct ad hoc queries over them anyway.

    We are not are not a cloud BI vendor (in fact, I’m not a big fan of this type of solution at all) but I have to side with Roman on this one. Basing a decision on which BI solution is better on whether it executes queries directly over the source or not is silly. YouCalc and Gooddata need to be compared at the solution level, not at this particular aspect that in most cases is completely irrelevant.

    I do agree that placing a data warehouse in the cloud is nothing spectacular and hardly a paradigm shift in the space. However, I don’t see how executing direct multi-source queries on premises, as YouCalc does, change anything either. These types of solutions were available way beyond Cloud BI was conceived and were only able to address very simple scenarios. As soon as amounts of data become non-trivial or if you have sources that cannot be queried directly – a data warehouse will be required no matter what you say – so what’s changed here? I think YouCalc’s solution is even more niche than Gooddata’s in respect to the variety of solutions it can support.

    Again, I am not a big fan of cloud BI but if you’re going to be such a player, doing both cloud and on-premises really beats the purpose. Whatever pitch you use to promote one option, reflects badly on the other.

    Elad Israeli
    CEO, Founder
    SiSense