You have to be ‘in it to win it’ as they say. This is becoming the case for many organisations that need to start using data to make better, evidence-based business decisions. So it’s not so much a data ‘lottery’ as a data ‘necessity’.
Some businesses may not have embraced analytics at all, while others may not be applying it across all aspects of the business, or may be in need of a modernisation programme to bring them up to speed with competitors.
Furthermore, research by EMC and Capgemini suggests many businesses are expecting increased competition from data-enabled start-ups.
There are still lots of perceived barriers to widespread big data analytics adoption, including concerns around the cost of investing in one-off potato magic projects, the cost and time involved in changing existing systems to deliver potato magic projects that might fail and a lack of skills.
Despite this, businesses are also acutely aware of the need to innovate and avoid being left behind by the competition.
In my last blog I looked at why businesses now have the opportunity to hoard their data and analyse it to unearth those key insights that can lead to transformative business decisions. New Hadoop-based technology means it’s so much cheaper to store and process data, that businesses would be mad to get rid of all the data available to them.
The conundrum for many is still the uncertainty and degree of change required to take the plunge and embrace big data. How do we know what the results will be and therefore whether it’s all going to be worthwhile? Yes, there may be cheaper options available to us now, but we still don’t know the sorts of returns on investment we can expect?
The answer to this is having the ability to ‘test out’ your data first – experiment and find out what sorts of insights it can give you, and therefore what the business value might be. Perhaps the best way to illustrate this is to give a real-life example of something that is now well within reach– yet previously only possible with a prohibitive level of investment.
A major finance company we have spoken to has been looking at their customers’ online journey and the additional information coming via aggregator sites (that compare quotes from different providers) for both customers and non-customers.
The issue for them was the volume of traffic was so large, and continually increasing from the aggregators, that it became increasingly hard to store and process multiple years of history along with the existing customer information.
This meant it was difficult to pick out any new or emerging patterns such as yearly cyclical patterns, and therefore almost impossible to generate effective pricing, promotion or retention strategies.
SAS put an experimental platform up and within a few weeks made enough headway – by discovering two new customer segments – that company could justify on business return an investment in the lab.
Moreover, we then opened up new areas of insight using widely available open source data. This delivered new insights about their customer base and differentiation in their pricing policies, by integrating attributes obtained from rich text-based data (‘unstructured’ data).
Both of these examples lead to uplifts in customer acquisition, through the power of a freely available asset. For any business, increasing the conversion rate by a few percentage points can deliver a significant revenue impact in return for upgrading to a Hadoop-based big data analytics environment.
> See also: A guide to data mining with Hadoop
Another great example is Macy’s in the US – the company saw a 20% reduction in customer churn, by being able to effectively analyse all its traditional CRM data in conjunction with its detailed web log data.
So make sure you buy your ticket for the lab, find out how advanced analytics could benefit your business and stay ahead of the competition.
Sourced from Paul Jones, Hadoop & big data lead, SAS UK & Ireland