How to turn insights into action
One of the challenges that organisations most commonly face when conducting data science and advanced analytics projects is how to take the findings of their analysis and use them to generate value for the business. In short, how to turn insights into action. It’s not unusual for such projects to be viewed as failures because the organisation wasn’t able to demonstrate how the insights generated action. In this blog I’m going to suggest some ways in which you can ensure this isn’t the case for your data science projects based on our many years of helping organisations not only conduct the analytics but also deploy the findings effectively.
What are actionable insights?
Actionable insights are conclusions drawn from the refinement of the raw data gathered by your analytics tool that can be turned into an action/response.
The data gathered by your analytics tool is pure information, for example the location and age of consumers visiting your webstore.
Insights are the value that can be drawn from this data by identifying key information and acting upon it.
What should an actionable insight look like?
Relatable – the insight should determine what aspect of the company should be addressed and how the insight itself relates to that area of performance.
Relevant – an actionable insight must show meaningful information that can be used to make the best decision for your company.
Clear – the actionable insight should be easily interpretable and communicated so that departments across the company understand how it could help the business and how organisation behavoiur will need to change to effect positive outcomes.
Intuitive – the best actionable insights display a pattern or trend that the business can be made aware of, make sense of and act upon.
Tips for turning insights into action
Set out your goals – approaching the insights and data with specific targets ensures that turning the insights into action will be as efficient as possible. Asking the right questions early on stops you getting bogged down in data that has little merit. By having end goals planned out, you know what to look for when analysing the information and can respond to the insights much more effectively.
Consider the context – data only become valuable when it is contextualised, without context data is a list of numbers and figures with no meaning. To add context to your data try asking the questions of, who, what, when, why and where, so that the interpretation of the numbers is prioritised and the insight becomes actionable.
Look at the wider time frame – make sure the data being analysed is not solely from one small time frame. Referencing historical data while analysing present information can help make more sense of it, as looking at the little picture can give a false reading and skew a business’s response to the insight.
Communicate the insight effectively – verbally discussing the insight or using visual guides such as charts ensure that the information gained from the insight can be efficiently shared between departments within a company. Good communication provides a solid foundation that action can be built upon as it makes sure all aspects of the business understand the information and subsequent response.
Look for patterns – recognising patterns means that the data will be analysed further. Identifying and understanding trends turns data into knowledge that can be acted upon or shared.
Formulate a hypothesis – after analysing your data and drawing insights from it, you must create a hypothesis that can be tested and experimented with.
To formulate a beneficial hypothesis
- Define the problem and validate with the data gathered
- Describe the proposed solution and explain why it will solve the problem
- Suggest metrics to measure results and set the criteria for success and failure
