This article is courtesy of TechRepublic Premium. For more content like this, as well as a full library of ebooks and whitepapers, sign up for Premium today. Read more about it here.
Vendors of business analytics tools are keen to widen the appeal of their products, and TIBCO is no exception with its latest Spotfire release. Top of a long list of enhancements is a best-practice Recommendations engine.
New 'Try before you buy' Recommendations engine
Customisable visual themes
Improved data recognition and matching
Appreciable learning curve beyond new Recommendations option
Separate desktop tool required to edit Spotfire Cloud visual themes
From $200/month for TIBCO Spotfire Cloud (1 authoring + 5 consumer seats)
Business analytics tools are still very much the domain of data analysts and other 'expert' users, but vendors are understandably keen to make their products more widely accessible. Salesforce.com, for example, recently released a search-based analytics service called Wave for use by regular knowledge workers. Now it's the turn of TIBCO, which has put usability top of the list of enhancements in the latest version of its flagship analytics tool, Spotfire 7.
Enjoying this article?
Download this article and thousands of whitepapers and ebooks from our Premium library. Enjoy expert IT analyst briefings and access to the top IT professionals, all in an ad-free experience.Join Premium Today
Despite being recognised as one of the more straightforward analytics products, Spotfire has nevertheless followed the standard model of equipping specialists with the tools to build custom analytics 'dashboards' for other, less expert, users. This approach is fine as far as it goes, but it can take days for new dashboards to be crafted and assumes that the specialists building them are skilled in both the analytics platform involved and the business processes and data being analysed.
The two rarely go hand in hand, so TIBCO's answer is to encapsulate the expertise assumed by previous releases and make it available to regular knowledge workers in the form of an option known as Spotfire Recommendations.
A key component of Spotfire 7, the Recommendations engine is, in effect a wizard that can be used to step users through the process of building a dashboard from scratch in just a few minutes. It does this by turning the traditional approach on its head so that, instead of of designing visualisations in abstract and then trying them out, it employs best-practice rules to let you pick from suggested visualisations based on an analysis of live data.
A simple example is the best way of understanding how this works in practice, so we signed up for a trial of the Spotfire Cloud service that delivers browser-based access to the TIBCO analytics tools. This has the advantage of platform independence plus the ability to call upon cloud server processing power to handle demanding workloads. It also delivers workgroup sharing facilities, although a standalone version called Spotfire Desktop is also available for Windows workstations, which can be linked to another tool called Spotfire Platform to enable sharing and collaboration in an enterprise setting.
We then needed some data, and Spotfire is well equipped to work with a wide range of sources -- from local files to shared databases, ERP systems and other enterprise platforms. For our evaluation, however, we opted for a fairly basic Excel spreadsheet, listing UK interest rates since 2000.
From data to visualisation in minutes
Creating our first visualisation was very easy. We logged onto Spotfire Cloud, created a folder for our analysis and then dragged our spreadsheet onto it. The data was immediately imported ready for use and the Recommendations engine started, as shown here:
Column headings in the spreadsheet were displayed ready for selection and by clicking on Rate and Date in the Data Panel a number of potential visualisations were immediately presented for us to choose from (below) enabling us to quickly add both the line and bar charts shown to the dashboard we were creating.
That done, we also added the Movement category (that is, the change in interest rate) to our analysis. The graphs were immediately refreshed with the new data and we opted to add the block diagram displaying coloured rectangles sized to suit the values represented (bottom left in the screenshot):
Click 'Close' and finally here's our finished interactive data dashboard after roughly two minutes' work -- not bad for analytics newbies:
You also get to see another nice feature of Spotfire in this screenshot - the ability to dynamically filter data, where we've opted to only display interest rates from 2007 onwards. Moreover, with more complex data sources you can drag columns onto a visualisation to have them quickly added to the graphs they're dropped onto.
The Recommendations engine isn't the only new feature in Spotfire 7. Another is the ability to customise the look of the application and dashboards using visual themes. Basic light and dark themes are available by default, plus you can also create and customise themes of your own and copy themes from one visualisation to another.
All of this really helps both enforce corporate style and highlight important information -- but it does seem like a fundamental option that should have been in the product long ago. Moreover, while the ability to edit visual themes is easily available in the desktop product, the cloud version when accessed via browser only lets you apply ready-made themes. You can still create custom themes for use with Spotfire Cloud, but this requires a download of a specially-adapted version of the Spotfire Desktop that can connect to the Cloud service.
Unfortunately this employs a somewhat different interface and switching in and out of the browser interface is a cumbersome workaround, clearly aimed at specialists rather than general business users.
Other improvements are also targeted more at the traditional specialist user market, such as grouping of categories to, for example, compare the top-10 performing countries against the others in a sales visualisation, or where naming conventions in the data might otherwise cause too many categories to be generated.
Support for nested aggregations has also been added, removing the need to create intermediate calculated columns. Spotfire 7 is also a lot better at recognising common data series, such as days and months, and sorting them automatically.
Elsewhere there are plenty of other expert tweaks and enhancements, although the overall emphasis is very much on making the application more accessible and easier to use. TIBCO manages this quite well and we were very impressed with how quickly we were able to progress from raw data to professional-looking visualisations.
Of course we were testing with a very simple dataset, and considerable expertise is still needed when connecting Spotfire to more complex data sources. A great deal of functionality hidden away behind the new Recommendations engine and visual themes options, but Spotfire 7 still has an appreciable learning curve. Still, it's a step in the right direction and we look forward to even bigger usability strides in the next release.