Data analytics has gone through significant change lately. Industry professionals are predicting that this year will see some radical changes, but for these to be beneficial and effective, business professionals need to address the fact that data analytics is not as quick as it needs to be.
According to recent industry research*:
- 72% of business leaders feel that the time it takes to gain insights from data could be improved
- 9 out of every 10 of those who stated the above, feel that this is down to the limitations of effectively combining data from multiple sources.
Integrating data from multiple sources or ‘data blending’ is vital for gaining business insights. Despite this, analysts are limited by current methods and tools. Dan Mills, Head of Data Warehousing and Business Intelligence at Corecom Consulting said:
“Feedback we have received from many Business Intelligence professionals is that they spend a significant amount of time in data blending and preparation rather than analysis.”
Ed Thewlis, Director of Technology at The Data Shed said:
“Data analysts require tools and skills to allow them to integrate data from disparate systems easily. If effective systems are in place, insight specialists can focus on the business problem, rather than performing the job of a developer.
In addition, the tools used must also enable analysts to easily publish their model. Currently, a lot of effort is wasted in taking a predictive model constructed in one tool, and translating it into another language for production use. Tools such as Azure ML and prediction.io overcome this by allowing analysts to immediately release their model as web service – simple to integrate into any production system.”
Experts predict that 2015 is going to see some radical changes in the analytics arena. Here’s what we can expect to see:
1. A significant decrease in coding-based predictive modelling
Data analysis and predictive modelling are responsibilities that are increasingly being used less and less by IT departments and more so by non-technical people who have a deeper understanding of business goals and objectives. Technical experts and data scientists are no longer solely required to perform such duties and the days of writing code and producing masses of SQL are on the decrease.
In the coming year, raw code-based predictive modelling will be increasingly replaced by drag-and-drop user interfaces which can be extended by using analysis-specific languages such as R. A reduction in data wrangling in code will enable analysts to focus more time on the analysis, and allow them to extend their models more efficiently
2. An increase in analytics and data hobbyists
2015 will see the emergence of an increasing number of people who thrive on unearthing and understanding self-taught data insights. A ‘data hobbyist’ is a great asset to any company – their enthusiasm inspires others and leads to an increased adoption of data analysis methods.
‘Data democratization’ is an interesting concept – perhaps not one for this article though.
3. Deeper insights leading to better informed business decisions
An increased amount of data to understand and analyse requires analytical methods to be quicker and more productive. Insights need to be shared in a timely manner; within minutes rather than hours or days.
If quicker, real-time data blending and analytics can be achieved, more data can be processed and more time can be dedicated to data exploration and gaining deeper and more intelligent business intelligence. This will lead to better informed business decisions and strategic planning.
Want to join the discussion?
If you are a Senior Business Intelligence Professional in Yorkshire and you would like to get involved in debating and discussing hot topics such as this, as well as networking with and learning from reputable and established speakers, then BIboss is for you.
Our next event will take place at 6.00pm on 11 June 2015. To register your interest for this exclusive invitation-only networking event in Leeds, please contact Dan Mills from Corecom Consulting at email@example.com or 0113 394 4188.