Consider the following situation. The store manager of a large retail store checks an app in the smartphone, first thing in the morning. The app knows which reports are likely to be analysed at this hour and provides a ‘default’ view. Also, based on the historical data and some external data on prices, the app ‘recommends’ decisions for merchandise around pricing, volume and stocks. The store manager reviews the data and then ‘enters’ his decision in the app which gets actioned by the team. Subsequently, the app also provides a report on the accuracy of these decisions as a feedback to the store manager. Welcome to the world of contextual mindful apps which are powered by analytics for effective decision making!
Analytics is no longer just about taking strategic one time decisions. It is getting embedded in the work-flow and helping in improving the quality of everyday decision making around core business processes like ‘procure to pay’, ‘order to cash’, ‘hire to retire’ etc. Such decisions usually require micro-level decision making by a large number of teams on a daily basis. In the store manager’s case, the organisation derives significant competitive advantage by way of either improving the top line (effective promotional campaigns for select products/formats), bottom-line (appropriate pricing for improving margins) or risk management (ensure reliability/quality of suppliers).
As micro level decision making becomes important day by day and the frequency of such analyses increases significantly, there is a need for a sustainable solution that will be embedded in the workflow. Embedding in the workflow is crucial from two perspectives —the app can get valuable data points about the usage of the application and hence the screens/output can be customised for the user and more importantly, the user is ‘encouraged’ to use the app since it is part of the workflow. This can drive the adoption of such initiatives in the organisation.
Such analytical apps help in augmenting the human thought process rather than replacing the need for it. Rather than looking at ten reports on a daily basis and figuring out which ones are relevant, such apps store the usage pattern of these reports and highlight the ones which are most likely to be used. This reduces the information overload. Moreover, the decision making process after the analysis of the data can be seamlessly achieved in the same application which also reduces the complexity for the user. But, the algorithms which are used for predictive/prescriptive analytics have to be provided based on the requirements of the situation and the accuracy of these predictions needs to be constantly monitored. The user may still have the ability to over-ride the system generated decision and such manual “over-rides” will also provide feedback to the application (improve accuracy) or to the user (for improving the decision making).
Some of the examples of such products are dynamically auto customised reports based on the CXO’s review pattern, estimating the demand of a product/offering by using external and internal data sources and in turn decide new territories for growth, predicting the employee churn using HR analytics or measuring the brand health index using social media analytics.
One such example is PwC’s SocialMind, where the solution uses natural language processing, industry specific taxonomy along with propriety scoring techniques to deliver consumer insights and actionable recommendation. Historically, the focus has been on improving workflows by automating and integrating the transactions and providing the information using business intelligence (BI) platforms or social collaboration platforms. However, the next frontier in business value creation is guiding employees do the right thing with the focus on reducing the time lag between ‘knowing’ and ‘doing. Contextual mindful apps driven by analytics are a significant step in that direction.
The writer is executive director —Technology Consulting, PwC India.
With inputs from Rajat Mathur, managing consultant, PwC India