For a significant amount of time, I believe that the following quote from Voltaire has been applicable to some of the practitioners of business intelligence and data warehousing (BI/DW):

“The perfect is the enemy of the good."

This comes from the experience that often the speed of business moves faster than many practitioners can implement BI/DW methodologies and technologies to supply value to businesses.

This is not to say that those BI/DW teams cannot match the speed of business… That would be an “excuse” for a return to the “wild, wild west” days of silo-ed data and limited value across the enterprise that created the need for enterprise data warehouses in the first place… But BI/DW teams need continually improve their skills and implementation timeframes to offer the value that their ‘customers’ demand.

TDWI – Agile Business Intelligence

Last week, I attended the TDWI conference in San Diego where the focus was on “Creating an Agile BI Environment” to provide timely value to business stakeholders.  Those familiar with Agile development practices could have confused TDWI’s use of “Agile” with the specific Agile methodology ( count me among them… ).

However, the TDWI executive summit sessions, classroom content and product showcase focused more on the “nimble” aspects of providing BI/DW environments as in “rapid prototyping or multiple iterations” rather than focusing specifically on BI/DW implementations using THE Agile methodology.

“Uppercase A” Agile vs “lowercase a” agile (read nimble)

In terms of BI/DW, I view the implementation of “nimble” practices to be more important than “dotting the Is and crossing the Ts” of a specific methodology like Agile.  Which brings us back to Mr Voltaire…

While this may not endear me to the fans of Agile methodology,  I believe that using “nimble” frameworks as a guide is a much better practical implementation solution than following a strict process that puts more emphasis on “perfect” process than “good” results.  Also, I believe that “nimble” enabling technologies can speed the implementation of BI/DW solutions to focus on delivering business value rather than worrying about technical “specifics”.

In Support of “Nimble”

During my three days at TDWI, I had the opportunity to sit down with several vendors who are focused on helping BI/DW teams with offering business value to their business stakeholder/customers.  Here are those that choose to blog about:

• Corda supplies quick implementation data visualization solutions that have helped customers like AT&T, Saleforce.com and Comcast achieve quick results.
• WhereScape has a rapid implementation environment toolset has enabled Telecom New Zealand, Vodafone and First Data to develop BI/DW environments efficiently and nimbly as well as in accordance with processes/methodologies. Specifically, Wherescape had Ted Schill of CoinStar with them as a case study/customer testimonial on how to speed BI/DW implementations.
• SpatialKey takes the approach of being able to implement data visualization of geo-location information without delving into the ‘details’ of geo-coding with utility and government clients.
• 1010data brings “big data” to the cloud with SaaS offerings that provides the ease of use of a spreadsheet with the power of 10b+ row tables/record sets.
  

What “nimble” means to Telecom

With the speed that telecom organizations are looking to:

• Acquire customers/provision services
• Introduce / enhance products
• Analyze / improve network performance

BI/DW teams need to be working at the speed of business or find themselves replaced by ‘shadow’ IT departments or other outsourced options.  This is not an “idle threat”, but rather a reality that many IT departments have failed to understand and/or recognize.

Using these new ( or at least new to BI/DW… ) “nimble” methods and technologies, telecom BI/DW organizations specifically, and IT departments in general, can focus their energies on providing business value from their BI/DW environments AND not cut corners that would limit the overall value for the sake meeting those requirements.

How do you see the use of “Agile” or “nimble” practices in telecom BI/DW?

Post your comments below or email (John.Myers@BlueBuffaloGroup.com) / twitter (@BlueBuffaloGrp) me directly.

With the execution of automated and manual processes becoming more and more important, it is essential to be able to develop KPIs effectively to communicate the status of that performance.  Most importantly is the ability to reduce the time to implementation or the amount of time from the “request” for information to the “delivery” of that information.

One area provides management teams and business stakeholders the most value is to visual process status.  Only understanding one particular aspect of the process can leave potential bottlenecks unidentified.   Seeing the status of the entire process as it is laid out is key to finding those bottlenecks for resolution or assuming an acceptable level of risk

At the recent TDWI Conference, I got to sit down with the folks from Corda to discuss their CenterView platform in regards to process performance management.

Currently on Centerview 4.0, improvements are planned for an early 2011 release of Centerview 5.0.  These will include increasing the velocity of what can be considered already quick implementation timeframes with the Centerview product.

Also, Corda talks about the ability to not just show information from the process being monitored, but to provide ‘metadata’ on how that information is being used by internal teams.  This application usage ‘metadata’ can be used for internal human resource improvement or setting goal targets.

Telecom Applications

As telecom organizations move toward nearly automated processes for operational processes like product/service provisioning; understanding and visualizing those processes will become paramount.  Service Level Agreements (SLA) will drive not only customer experiences, but have an impact on partner agreements as products move from the world of features of the network to IP-based applications that operation on the network

Just as important will be to understand the controls associated with who is using that data/information.  Teams that embrace the monitoring and management of the process KPIs to impact their performance should show associated improvements in their team productivity and KPI goals.

Conclusions

Process management status visualizations will be more important as organizations continue their process automation.  Rapid and effective implementation and iterations of those KPIs could mean the difference between leading and following in a particular industry.

Long time readers know that I don’t like John Wooden… I LOVE John Wooden.  Two of his best know maxims are:

• Flexibility is the key to stability
• Be quick, but don't hurry

Both of these are key to lowering the time to implementation for business intelligence and data warehousing projects and their ongoing support and development.

In my last analysis of WhereScape’s RED rapid application development tool, it was apparent that WhereScape had done an excellent job of providing the “flexibility” of implementation to maintain the “stability” of value for a data warehouse environment.

The WhereScape RED allowed for either the rapid prototyping of a data warehouse environment or the iterative support of that environment.

This week at the TDWI Conference in San Diego WhereScape announced the next edition of RED.  With this release, the Agile Development community will find many features that directly implement Agile practices for: 

“continuous integration testing, improved ability to back out changes, and better integration with source control and testing systems.”

These improvements allow for the continued flexibility of the WhereScape platform to be “quick”, but now integrate established process to make sure that DW implementation teams don’t “hurry” the process of iterations.

Telecom Applications

In telecommunications, the increasing standards of just in time (JIT) product delivery has put pressure on telecom BI implementation teams to provide analytical value to their business customers in marketing and product management on a similar time line.   The flexibility of WhereScape will enable that time to implementation AND will provide controls over how those systems are implemented.

Conclusions

Whether a team has adopted Agile methodologies or are just looking to be nimble due to time pressures, having quick iterations will enable a telecom BI implementation team to provide the stability to provide ongoing value to business customers.

In the world of telecommunications, location based services (LBS) are going to be the next level of products/services that provide value to consumers and those interested in targeting those consumers.  This trend is starting to take hold as LBS attributes / applications are starting to go “mainstream” via Twitter, Facebook and others.

However with the LBS applications, geocode data is attached to the event records associated with those services.  If there is something that might be more difficult to “read” than trunk group information, it is probably geocode data.  The following is an example:

For the initiated, it is probably clear that these events come from Columbia.  However for most management teams or marketing decision makers, seeing the patterns associated with that data set is probably impossible.  NOTE – I think I found errors in that dataset since I doubt any part of Columbia was located at 74 degrees “EAST” (-74.2343) rather than 74 degrees “WEST”….

“If you don't know where you are going, you might wind up someplace else.”

Ahhh the immortal words of Yogi Berra…. If you don’t know where you are and/or are going, you could end up somewhere else.  For telecom geocoding information, you need to present the geospatial information in a format that the business stakeholders can understand and interpret.

At the recent TDWI conference in San Diego, I was impressed with the capabilities of the software as a service (SaaS) “location intelligence” provider SpatialKey.

Their presentation layer was clean and visually appealing.  But the aspect that I enjoyed over most was the SaaS model that allowed for this relatively complex geocoding presentation layer to implemented without the need to deploy internal servers.

Telecom Applications

Clearly being able to either monitor the “status” of LBS services or the location of customers using the LBS services is an application that telecom business intelligence organizations would find with SaaS geo data visualization.  The next level of analysis/information value is communicating to external stakeholders who may be:

• Purchasing targeted adverts
• Monitoring effectiveness of their campaigns

Conclusions

With the issues associated with IT Departments in telecoms, I see a growing need to move toward a ‘value based’ approach rather than an infrastructure approach for systems delivery.  And with LBS being the next step in applications, telecoms BI teams would be well served to look at SaaS implementations like SpatialKey for their initial projects and possibly long-term internal or SaaS systems.

"Hide not your talents. They for use were made. What's a sundial in the shade?"

Wonderful words from Ben Franklin.  They are too often forgotten when it comes to business intelligence and data warehousing.  Many times the design and implementation of an analytical environment takes away an analysts ability to look at the “raw data” and replaces it with a view into a single aspect of that data.

To be fair, that encapsulation of the “raw data” in analytical environments was often a result of a lack of functionality/power to present millions of records at one time.  However, currently the power of analytical environments can allow for more of the “raw data” to be presented to analysts in a format in which they are familiar.

At the recent TDWI conference in San Diego, 1010data presented their Transparent Database concept which matches a “spreadsheet” interface with a powerful software as a service (SaaS) DBMS in the cloud engine to support the following:

• Data Warehouse as a Service
• Business Intelligence as a Service
• Data Integration as a Service

Telecom Applications

For telecommunications organizations, the transparent database from 1010data offers opportunities for both business stakeholders and IT departments.

Business stakeholders will find the “spreadsheet” interface dovetails nicely with the most widely available analysis tool (MS Excel) in telecom.  This lowers the learning curve and enables Franklin’s call for not hiding the talents of analysts to use the skills they have developed over years.  1010data, however, provides the ability to see not just 1m rows/records as with Excel 2007/2010.  The range is closer to 10b+ per “table” or record set.  This allows the analysis of a much wider range of events rather than simply what can fit into a spreadsheet.

For IT departments, 1010data offers the ability to speed data availability for business analysts by leveraging the SaaS model to allow for ad-hoc datasets to be quickly placed in that transparent database environment either for short-term/ad hoc environments or long-term standardized analysis.

Move that Sundial

While I probably would have preferred to see a greater level of analysis tools being available with 1010data’s offerings, I like the focus on the quick easy presentation of data associated with the heavy loads in telecom usage anlaysis – be it for marketing, billing or network.  This presentation style provides analysts of all types the ability to get their fingers in the data instead of “hiding their talents” behind encapsulating interfaces whether they be SQL to get the data or an analysis wizard to manipulate the data.