In today’s connected world with data overload, it is becoming increasingly difficult to differentiate your product in the eyes of the customer. With the customers having a large number of options available literally on their fingertips with their mobile screens, the choices are endless. The race is going to be won by companies who can capture the mind space of the consumer and when the consumer feels such a personalized connection with a product that he doesn’t make a small effort required to find other options.
So how does an organization go about creating this personalized connection with each unique customer caring for their unique preferences? A marketing drive however closely targeted to your customer segment cannot target each customer’s preferences uniquely. The future is going to be decided by hyper-personalization and better targeting that most large organizations are trying to do today.
The other part of the puzzle is employee satisfaction. The old adage that happy employees make happy customers cannot be underestimated. Organisations need to balance hyper-personalized customer marketing and ensure employee satisfaction at the same time.
The effectiveness of the future strategy is based on the decisions taken by the organization today. The decisions are based on the information available to the management and the foresight of the decision makers. The better the quality of information, lesser is the need of the intelligence of the management team. This is where the difference lies – Better the information, better the decisions. But, how do we assess the quality of information? Is there a yardstick that helps us decide when the information is good enough to take a decision that impacts the company’s fortunes in the future? The answer is clear and present – data Analytics.
Taken to an extreme, if the information is analyzed to such an extent that it can provide a clear course of action, the requirement of someone taking an intelligent decision reduces and the decision can be automated. This is where artificial intelligence and machine learning comes in. At the heart of it, however, is better data analytics and higher processing power to allow the data to be analyzed to such an extent that the human decision-making process is no longer required.
AI, Machine learning, NLP, data modeling, statistical testing, and predictive analytics work together to allow the computers to achieve a reliability of data that can be used to make automated decisions. In the case of data science and data analytics, by using the right combination of these technologies, companies can automate sophisticated aspects of their responsibilities that previously had to be completed manually. This clearly leads to better decision making and therefore better future outcomes for the organization.
In my opinion, I would outline the decision-making structure in the following steps:
- Get valid data– ensure data is accurate and with as little measurement error as possible. Big data, social media data, system generated data for customer usage patterns and other data are all put together and summarised. This could be achieved by survey applications that support statistical tests and sample sizes. This is especially useful in getting customer preference data – in the absence of asking customers directly, all your data would be inward focused and assumptions built on assumptions. Be sure to conduct surveys and find out what the customer or employee is telling you directly, in their own words.
- Data Analytics– text analytics and advanced modeling of this data could throw up interesting observations. These observations need to be validated for statistical significance and sample size. While not a recent buzzword, ‘statistically significant difference’ is possibly at the heart of most data-driven analytics. This enables decision-making as the preference difference is established. To take an example, if we are measuring customer preference for product A Vs B, if the data shows that customers have a statistically valid preference for Product A, it is a no-brainer to go with the development of product A. It would be superfluous and unnecessary to have a human make this decision if the data is clearly showing the preference.
Similarly, on the employee’s front, Companies around the world are adopting people analytics. In today’s world, any company can replicate your product, marketing strategy, business model or the services you offer. While customer data science can keep you ahead of the curve on the customer side, people analytics can ensure that your employees are satisfied and helping build a great successful company –people could be your greatest competitive advantage.
Organizations today have lots of data about their employees including educational qualification, age, demographics, salary, tenure, satisfaction, ratings, performance, and much more. However, few organizations have a structured way of putting all of this information together to understand their employees. People analytics helps answer the following critical questions – what is the ideal profile for recruitment? What makes the employees motivated to perform well? What will help retain the employees best? There are of course a host of other data points like what qualities in managers make the organization more effective, how to get teams to collaborate better and how work-life balance is affecting employee performance. Without going into training and Organisation development side of it, even though data analytics helps in both of those as well, the role of data science in people analytics cannot be underestimated.
Organizations must do their own internal research to find our their own reality and truth. It’s easy for us to read research reports and studies from consulting firms and assume that what we read applies to us. If a study says customers prefer more email communication, then it means we should do more communication for our product too. External research is broad-based and sometimes questionable in the first place. Recommendations from external research should not be applied directly to your organization without a validation. Each organization and product is unique and therefore each organization needs to find out the preferences of their stakeholders in their unique environment. It is imperative to conduct your own research, do your own data analytics and derive your own conclusion to take the right decisions.
Fortune favors the prepared – be prepared with data analytics! Good luck.
About the Author
Ashutosh is the Founder of datapotential.com (an advanced statistical online survey platform) and Aselector.com (an Artificial Intelligence based knowledge management platform). He has over two decades of experience with multinational companies in leadership roles in quality, data analytics, transformation and program management. He is a speaker and thought leader on AI and statistical methods.
