Making Big Data Useful

Companies see the value of big data, but are still struggling with how to use it effectively. These 3 stages are crucial in making big data useful.

A recent report by NewVantage (2019) has found that 92% of Fortune 1000 companies are increasing their investment in big data; however 77% also report that it continues to be a challenge for them. To become more data-driven in decision making it isn’t enough just to have the tools and technology, businesses must develop a data culture.

Applying an effective strategy to the development of your data culture will undeniably increase your overall competitive advantage in your market. Working within our data team here, these are my three key areas to focus on when aiming to become more data-driven.

Data Strategy

Without strategy, there is no purpose and this rule can be accurately applied to data gathering and analytics. Data and strategy teams must create an effective bridge in communication to work together effectively.

Firstly, consider what your hypothesis or objectives should be to match the appropriate data to the situation. This will help you discover what opportunities there are with the data that is available.

Turnali (2019 Forbes) stresses that “data is useless when we do not ask the right questions”. Through critical thinking, some questions you can consider before collecting data are:

“What is the problem you need to find a solution for? What is the purpose of gathering certain data? Where can you pull first-party or third-party data from?”

Once you have your hypothesis and objectives, consider their feasibility such as whether the data you would like to use is accessible and where you would gather such information from. You may also find some correlation with your data which can allow you to improve your objectives and understanding of the situation.

I find that more data doesn’t necessarily mean more insights as this can also skew your findings and confuse the interpreter as Mayhew, Saleh and Williams (2016 McKinsey Quarterly) discuss that “it’s the data points that help meet your specific purpose that have the most value.”

Organising your data will also reduce the time spent trying to interpret it if you wish to share your schema with other people or report on the findings later. This is an extremely important consideration as it makes everyone’s lives easier in the long-term if you have an efficient and consistent extraction, transforming and loading data process.

 

Insights

So, you have all this data and though you have your objectives, you’re staring at this wealth of information like it doesn’t mean anything. One singular point of data may rarely show something that you didn’t already know; however, when you join or compare data it can provide some greater understanding and even open up unexplored areas of interest.

For example, when I have been interpreting some data I found a correlation between the bounce rates of particular pages and the reduction in conversions of a product that was being sold on one of these pages. An insight made from this would suggest that their website is missing a level of optimisation for engaging with their users.

I find that delving into data to find all the best insights will take time and it comes with some very satisfying aha! moments as the data starts to take form.

Visualisation

62% of surveyed Fortune 1000 companies have seen measurable results from their investments in data; however, I believe this isn’t as high as it should be.

Turning data into something more visually pleasing is most likely the area where this limits companies’ perception of measurable results. You may have the greatest data and insights team though many forget that this also has to be something that everyone can understand or else it will be cast aside. When creating these dashboards, I consider what visuals would have the most impact than choosing to overload it with information which can confuse and turn away the reader.

It may take some tweaking though considering how your data should be visualised and what tables are best to do this with will make it easier to go back to see if the data has matched your fundamental strategy.

I believe these three key areas should be in the mind of those who would like to become more data-driven in their decisions. From the surveyed Fortune 1000 companies, investment in machine learning has increased from 68.9% in 2017 to 96.4% in 2019 for the automation and faster processing of their data and technologies. Having these systems in place would provide any company with an absolute advantage in their market.