Departments
Next
Up With Mashups!
Disparate digital data combines into the latest technology hit.
by Lisa Campbell
Mashups are making a splash in the technology world. A mashup is a digital media file that combines data such as text, maps, graphics, audio, video and animation to create a whole new file. Mashups got their name from the disc jockey practice of merging two or more tunes. Similarly, mashup developers combine data from two or more sources to create something more valuable than the sum of its parts.
In its May 2008 report "The Mashup Opportunity," Forrester Research writes: "Mashups—custom applications that combine multiple, disparate data sources into something new and unique—hit the Web in a big way starting in 2005. … Forrester projects that the enterprise mashup market will reach nearly $700 million by 2013."
The trend is gaining momentum. Platforms have emerged that allow people with little or no development experience to create mashups. These platforms include Yahoo Pipes, Dapper and Microsoft Popfly.
A lot of the existing popular mashups are based on adding data to mapping information. One example is Flickrmap.com, which combines Google Maps with galleries of photos from Flickr.com taken at specific locations. Similarly, Housingmaps.com takes real-estate listings from Craigslist.com and merges them with Google Maps to show properties for sale or rent in North America in a graphical format.
Enterprise mashups, however, integrate data from internal and external sources and are used by organizations for their business needs. They are often developed to solve a particular problem, are implemented in days rather than months and use standards like extensible markup language (XML) or really simple syndication (RSS) to make sharing and subscribing to them easy.
Better together
To leverage the myriad data sources modern organizations have access to, enterprise mashups combine the more traditional structured data with the limitless possibilities found in unstructured data forms. When united in a mashup, these two types of data offer businesses an easy-to-read and, therefore, more useful presentation of information.
The structured data in the data warehouse could include information on an organization's sales, salary, operations, human resources and inventory. It could also include external data to provide simulations and analysis of all the internal data.
Additional valuable information lurks in unstructured data—such as text from the body of an e-mail or word processor document, notes taken during a phone call, or text from blogs or chat rooms. It could also be audio or video data that is posted on a Web site.
As this type of data grows with people spending more time communicating on the Internet, this source will contain more and more business-critical information that must be made accessible both for business intelligence (BI) and operational needs.
Enterprise mashups enable organizations to combine structured data, unstructured data and visual BI tools such as dashboards and Web-based portals to let customers, employees, partners and other businesses get information on a particular topic in an integrated and visual way. The information in these cases can be displayed in a variety of formats, such as charts, graphs, tables and maps.
“To leverage the myriad data sources modern organizations have access to, enterprise mashups combine the more traditional structured data with the limitless possibilities found in unstructured data forms.”
Industry uses
Within many industries, the information necessary to make informed business decisions changes on a daily, hourly or even real-time basis. Executives need to be able to pull data from multiple sources, update that information and analyze it in a way that makes sense to them. Enterprise mashups offer solutions for this critical business need in virtually any industry. For example:
- To better enable predictive modeling in the relational database of an insurance company, unstructured text appearing in claims can be captured. An enterprise mashup can then combine this information with its existing customer- and claims-focused structured data. This combination helps the company determine its risk factors and improve its underwriting processes and rules.
- A real-estate agency could create an enterprise mashup by combining a list of all houses sold in the last week by the agencies from local communities with internal data about which houses the agency itself sold. Then, it could plot all of this on an interactive map. This would help the agency track its progress against competitors and save time by having an organized list of available properties.
- For a retail chain, enterprise mashups could combine unstructured information gathered by the company's call centers and in-store associates with structured inventory information to determine whether there is a consistent problem with a given product or manufacturer. The retailer could then reduce return costs by eliminating these products or requesting that the manufacturer take corrective action.
- In the financial services industry, a customer insight mashup could combine consumer information from all sources into a single portal in an effort to up-sell and cross-sell its products and services. A bank could enhance existing credit-risk modeling with external data, such as BI or demographic information from third parties. For online banking, it could allow customers to bring in dynamic new information for a custom portal.
An innovative approach
Enterprise mashups are a fairly new approach to getting the right information into the hands of the right people quickly and in an understandable, visual way. With this technology, new data sources can be added much more efficiently and easily than with traditional technologies. And once a source is made available, it can be mashed with other existing sources into an almost limitless number of information combinations.
Companies from all sectors will surely benefit from enterprise mashups. Will yours?
Lisa Campbell is a freelance marketing and technology writer in New York.