A Call for Analytics Literacy

Michael Vizard
Slide Show

Nine Predictions for the Analytics Industry in 2011

The competitive gap between analytical innovators and those who do not invest in analytics will widen over the coming 12 months.

There's a lot of interest in predictive analytics these days, but not everybody is on the same page in terms of what to expect from an investment in analytics. The simple fact of the matter is that expectations of what can be accomplished using analytics might be too high, while at the same time obsessions with data quality can easily result in nothing ever really being accomplished.


On the one hand, business executives hear the term "predictive analytics" and they start to think about magic bullets. While predictive analytics can be a boon, businesses need to be very selective about where they apply the technology. Andy Fano, head of analytics research and development at Accenture Tech Labs says analytics should not only be applied to high-value business processes that are reasonably well-defined to justify the return on investment required, as opposed to trying to predict the company's stock price. The people using the analytics also need to understand the values that have been applied to the data in order to know if the assumptions ingrained in the analytic application are valid. Otherwise, you can easily wind up with analytics applications that make no sense to the business.


On the IT side of the house, Fano says that IT organizations need to get used to the idea that business data has a lot of "noise." The data being analyzed will never be pristine and yet IT organizations obsess about how pristine the data is to the point where they never get around to applying any analytics. Analytics applications are designed to give users confidence in business decisions by providing additional insight within a range of probabilities. The analytics models make assumptions about the validity of data from multiple sources, so as long as the data has a reasonable level of quality the analysis will stand on its own.


Of course, Fano says we need to spend a lot more time educating business users in terms of becoming more literate about how to read into analytics. With the advent of huge amounts of unstructured text and video, there is more data than ever to analyze, but there's not enough people schooled in the ways of analytics to make use of it. Fano says this means we should expect to see a proliferation of new businesses based on the concept of offering "analytics-as-a-service" to a particular vertical industry.

 

But in the meantime, business leaders and IT professionals would do well to sit down and establish some realistic expectations for the value of analytics in their business and what it will take to derive value from those investments. Otherwise, analytics will simply become yet another over-hyped application that nobody gets the real value out of because they don't understand how to really use it.



Add Comment      Leave a comment on this blog post
Apr 1, 2011 6:53 AM Ilyas Iyoob Ilyas Iyoob  says:

Some of these gaps can be filled by what we call 'preprocessing steps' in Analytics.  For example, data filling (for missing data) and anomaly detection (to remove noise) are key preprocessing steps.

Reply
Apr 4, 2011 6:32 AM Meri Gruber Meri Gruber  says:

Hi Mike

You raise a good point. We find many companies struggle with their analytics projects when they start with the data instead of with the business decision they are looking to improve. Companies that collected and integrated the data they needed for a specific decision like 'is this product out of stock' or 'what is the best retention offer for this customer' had shorter implementation times that those with a more general approach. By starting with the decision in mind, they also know what data they need and at what level of granularity.  Too often analytics projects that start first with data quality projects find that the data they have isn't what they need.

Reply

Post a comment

 

 

 

 


(Maximum characters: 1200). You have 1200 characters left.

 

 

Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.


 
Resource centers

Business Intelligence

Business performance information for strategic and operational decision-making

SOA

SOA uses interoperable services grouped around business processes to ease data integration

Data Warehousing

Data warehousing helps companies make sense of their operational data


Close
Thanks for your registration, follow us on our social networks to keep up-to-date