Big Data Reality Check
The promise of business gems from big data may be more of a long-range goal as enterprises work through their own issues with BI systems and unstructured content uses, according to a new report from non-profit information management professionals group, AIIM.
The promise of business gems from big data may be more of a long-range goal as enterprises work through their own issues with BI systems and unstructured content uses, according to a new report from non-profit information management professionals group, AIIM.
The survey, entitled “Big Data – Extracting Value from Digital Landfills” consisted of more than 400 responses to an AIIM member survey conducted in March and April. Less than 10 percent reported that it is already an “essential capability” for business and about 15 percent expect that it will be essential. On the flip side, nearly 15 percent reported they weren’t really aware of what big data is yet and about 20 percent were merely more than curious about its potential. As far as data analysis, reporting and BI across structured data sets, only 4 percent are developing advanced analytics and big data tools, while most still opt for spreadsheets and ad hoc reports (30 percent) and reports in individual business systems (29 percent).
With half of the survey’s respondents answering, big data and unstructured content tools are more of a long-range project; about one-quarter expect to move forward on some project in the next two years, and about one-third are expected to invest in something in the next two-to-five years. About 7 percent already have invested in big data and unstructured content tools, but 10 percent stated it is “unlikely” they will do anything.
Doug Miles, author of the report and head of the AIIM Market Intelligence Division, wrote that there has been an “almost sci-fi appeal” with big data from industry analysts. But Miles said there is a “reality check” at the enterprise level as information managers match data analysis techniques new and old to business insights and other IT initiatives.
“One of the confusions amongst the user base is how ‘big data’ analytics differs from normal reporting or BI, particularly in the context of structured data or transactional databases,” Miles wrote in the report. “For these extensions of conventional reporting, complexity and scope play a part, as well as the size of the data to be analyzed. The ability to scale the data yet produce query results within reasonable times may lead on from technical differences between conventional data warehouses, and some of the more recent database technologies.”
When it comes to the existing management of unstructured content, 35 percent of respondents considered their efforts organized but not well indexed or controlled; 26 percent registered it as “somewhat chaotic”; 19 percent noted that management was in place across some enterprise areas. Nine percent stated their unstructured content management was in place across most areas and another 10 percent stated that unstructured content was well managed in its own repositories.
Top uses or considerations of an unstructured content tool included recognition of keywords and phrases in text, pattern recognition, content categorization and migration, and Web tracking. In all, about half of the enterprises in the survey were using or considering a solution for content analytics, and about one-third were using or interested in offerings for unified information access and semantic analysis. There was far less interest in or awareness of emerging database and query technologies, with use and considerations in the single digit percentages for Hadoop, NoSQL and MapReduce.
Entry-level recommendations on big data from Miles and AIIM for those enterprises include considering what business benefit these data streams could provide prior to crunching costs, as well as assessing the data you have in-house and how you interact with customers.
Click here to download the report. The report was underwritten by Actuate, Attivio, EMC and IBM.
Justin Kern is senior editor at Information Management and can be reached at justin.kern@sourcemedia.com. Follow him on Twitter at @IMJustinKern.
The survey, entitled “Big Data – Extracting Value from Digital Landfills” consisted of more than 400 responses to an AIIM member survey conducted in March and April. Less than 10 percent reported that it is already an “essential capability” for business and about 15 percent expect that it will be essential. On the flip side, nearly 15 percent reported they weren’t really aware of what big data is yet and about 20 percent were merely more than curious about its potential. As far as data analysis, reporting and BI across structured data sets, only 4 percent are developing advanced analytics and big data tools, while most still opt for spreadsheets and ad hoc reports (30 percent) and reports in individual business systems (29 percent).
With half of the survey’s respondents answering, big data and unstructured content tools are more of a long-range project; about one-quarter expect to move forward on some project in the next two years, and about one-third are expected to invest in something in the next two-to-five years. About 7 percent already have invested in big data and unstructured content tools, but 10 percent stated it is “unlikely” they will do anything.
Doug Miles, author of the report and head of the AIIM Market Intelligence Division, wrote that there has been an “almost sci-fi appeal” with big data from industry analysts. But Miles said there is a “reality check” at the enterprise level as information managers match data analysis techniques new and old to business insights and other IT initiatives.
“One of the confusions amongst the user base is how ‘big data’ analytics differs from normal reporting or BI, particularly in the context of structured data or transactional databases,” Miles wrote in the report. “For these extensions of conventional reporting, complexity and scope play a part, as well as the size of the data to be analyzed. The ability to scale the data yet produce query results within reasonable times may lead on from technical differences between conventional data warehouses, and some of the more recent database technologies.”
When it comes to the existing management of unstructured content, 35 percent of respondents considered their efforts organized but not well indexed or controlled; 26 percent registered it as “somewhat chaotic”; 19 percent noted that management was in place across some enterprise areas. Nine percent stated their unstructured content management was in place across most areas and another 10 percent stated that unstructured content was well managed in its own repositories.
Top uses or considerations of an unstructured content tool included recognition of keywords and phrases in text, pattern recognition, content categorization and migration, and Web tracking. In all, about half of the enterprises in the survey were using or considering a solution for content analytics, and about one-third were using or interested in offerings for unified information access and semantic analysis. There was far less interest in or awareness of emerging database and query technologies, with use and considerations in the single digit percentages for Hadoop, NoSQL and MapReduce.
Entry-level recommendations on big data from Miles and AIIM for those enterprises include considering what business benefit these data streams could provide prior to crunching costs, as well as assessing the data you have in-house and how you interact with customers.
Click here to download the report. The report was underwritten by Actuate, Attivio, EMC and IBM.
Justin Kern is senior editor at Information Management and can be reached at justin.kern@sourcemedia.com. Follow him on Twitter at @IMJustinKern.
More for you
Loading data for hdm_tax_topic #better-outcomes...