Business Intelligence Data Mashups with Jay-Z and The Beatles

What do The Beatles and Jay-Z have to do with business intelligence dashboards?

Jay-Z, The Beatles and Enterprise Data Mashups

Welcome to the world of the data mashup!

As we have seen in the last few business intelligence dashboard examples, there is definitely a trend towards the use of data mashups on enterprise dashboards. Components such as Google Maps are valuable contributions on BI dashboards that add to the value of the user experience.

I’d like to point out an MP3 on that gives some great background on data mashups. It’s an interview titled: Audio Report: Mashups, Data and BI.

Byron Igoe starts off by explaining the general mashup phenomenon with a look at The Grey Album – a mashup of The White Album by the Beatles and The Black Album by Jay-Z. He explains that the idea of mashups in general has become quite popular.

Mashups in music were one of the first ways that mashups rose in popularity. People have been combining different songs that were not created originally with the intent to be used together.

The example given in the interview is that of music mashups with the famous example of “The Grey Album” by DJ Danger Mouse that mashes together the Beatles’ White Album and rapper Jay-Z’s Black Album.

An aside: The Grey Album is not just a simple mashup where the vocals from one song are laid over the beat from the other song, but a deconstruction of both and a merging into something new. The background story is interesting. EMI sent a cease and desist letter to DJ Danger Mouse and to several websites and sellers of the album. That triggered an organized “Grey Tuesday” in which over 170 web sites made the Grey Album freely available for 24 hours. The 100,000 downloads made it the number 1 album that day. If you are interested, you can download the Grey Album here.

To explain data mashups, Igoe starts off with the idea of formalized ETL and data federations and adding into that the informal requests on the front end by the user. He feels that the data mashup is really the next iteration of the ad-hoc query.

The use of Excel as a business intelligence front end and its weaknesses is explained. The mashup is compared to that and addresses the weaknesses.

Another topic discussed is “Why isn’t the data mashup more prevalent in today’s BI tools?” Is the concept too new to be embraced? Issues of trust and security can be difficult to address.

Take a listen to that very interesting interview. From the tdwi description:

Microsoft Excel is a wildly popular data manipulation tool for users, but it’s not always the best tool. It continues to be popular because it lets users access and manipulate data themselves. Another way to achieve that, while still maintaining IT controls over data, is through a relatively new concept that InetSoft’s Byron Igoe calls data mashups. Data mashups, according to Igoe, are a logical next step from ad hoc data queries. He explains how through a data mashup, data from different sources, including both formal data warehouses maintained by IT, and informal user data, can be linked together with a BI front end.

2 thoughts on “Business Intelligence Data Mashups with Jay-Z and The Beatles

  1. So… where is the product that does this well?

    My longest task each month (and the one least relevant to my core role) is manually mashing a huge number of things in Excel into one dashboard.

    Something like QlikView could be good because you can just point it at e.g. somebody’s spreadsheet – the sort of stuff you’d never dare stick into a data warehouse, but then QlikView when challenged by me to create a really data rich dashboard (and given an example with about 40 microcharts on a page came back with a many page dashboard that has about 3 charts on it, so it mashes, but falls at the visualisation hurdle.

    I have high hopes for Tableau, which I am taking a much closer look at.

  2. We use Microsoft’s SQL Server Integration Server to pull data from various sources and ‘mash-up’ into fact tables for SQL Server Analysis Services cubes. A job is scheduled on the server to run each morning before the execs come in to run the SSIS package to pull the data, then another SSIS package builds the cubes. SSIS works for us since we have our data on different platforms – SQL Server and Oracle.

Leave a Reply