Understanding where to start when it comes to insights from your data can be troubling. Especially for new managers, data scientists and analysts new to data science.
With more organisations now investing heavily in data collection, it is projected that spending on the data infrastructure will reach $100 Billion in the next 5-years.
However, with costs for storage, collection and organisation of data, organisations need results from their investment. This has put more pressure on management to prove some ROI from their data assets.
For data teams, this may lead to a lack of context on requests from management.
An example of this would be a broad request, “What customers are likely to cancel their memberships?”
Where there are multiple answers, and the result will not always offer a solution for the organisation.
Data teams could also be asked to simply mine the data and produce some sort of result. However, without the right context for what they are looking for, the organisation will not benefit from any findings.
So how do you get the right insights from your data?
If we imagine an organisation as a mechanism of constant moving cogs. Each playing its part for the organisation to survive. Its relationship with customers, the environment, functionality, etc. It would be impossible for any one single person to understand the organisation 100%.
There will always be a gap between how you think the organisation works and how it actually works. This gap will always increase, based on how the company evolves, changes and grows.
Data helps to understand the behaviour within any organisation.
Example “How one cog could affect another.”
Insights help us bridge the gap between how we think the system works, and how it really works. The insight helps to understand the behaviour to benefit the organisation.
Andy Grove’s said in his book- High Output Management. Complex systems are black-boxes and an insight is like a window cut into the side of the black box that “Sheds light” on what’s going on inside.
Data helps to understand the behaviour of any organisation, and the insight helps to understand that behaviour to benefit the organisation.
Predictions and insights.
Data teams can make predictions based on their current knowledge of how the mechanism works. Creating predictions and experiments to match the predictions against the data.
With the right context, the data is checked and reviewed to understand if it matches the prediction.
If it does, great. If not, further exploration is required to understand why they got the result, and update their understanding of the mechanism. Creating new predictions and repeating the process.
It’s the understanding of the “If not” and the “further exploration” of why the prediction didn’t match the result that will create the insight. Thereby helping to understand that element of the business in more depth to benefit the business.
While finding the insight from your data is key for any organisation to succeed and for managers to prove ROI. It is also essential that data teams have the right context for what you want to achieve with the information.
Using the same example as before ” What customers are likely to cancel their memberships?”
Data teams will make predictions and will try and prove those predictions. Thereby creating insights that may not be a benefit to the business.
Prediction: Low income based families.
Insights: 23.4% of low-income families cancel their membership within the first year.
But what they actually wanted to know was, “what customers are likely to cancel memberships in city 1 against city 2?”.
This may present different insights from the data if customers cancel their membership for different reasons in each city.
Context with all requests
Making sure your data teams have full context included in the project management of any insight request is key for the success of the project. Brijj project management gives data teams full context on all requests, providing everything that is needed to increase project success, prove ROI and deliver value on your data.
With reports stating that 85% of big data project lead to failure, and with the average company wasting £158,000 on data projects a year. Having the right context with any request is key to get the right insights, and having the right platform will help you deliver more value.