Data and analytics are becoming increasingly integral to business decisions. Organizations across all sectors are leveraging advanced real-time predictive analytics to balance intuitive decision-making and data-driven insights to monitor brand sentiment on social media, better serve their customers, streamline operations, manage risks and ultimately drive strategic performance.
Traditional business intelligence reporting tools with static charts and graphs are being replaced by interactive data visualization tools that enable business insights to be shared in a format that is easily understood by decision makers, in turn enabling better, faster operational and strategic decision-making.
Given the constantly changing business landscape driven by increased competition, macro and geopolitical risks, intense regulatory demands, complex supply chains, advanced technological changes etc. decision makers are turning to finance teams for actionable insights to help them navigate through this volatility and uncertainty.
As business unit managers increasingly rely on finance decision support for enhanced business performance, it is imperative for CFOs and their teams to ensure the performance insights they are delivering are informed and actionable. However, a 2016 survey report by KPMG revealed that 60% of the survey participants are not very confident in their data and analytics insights.
Data Quality or Quantity?
In today’s world of exponential data growth, the ability for finance to deliver reliable and actionable insights rests not on the quantity of data collected, analyzed and interpreted, but rather on the quality of that data. The starting point for any successful data analytics initiative involves clarifying why the business needs to collect a particular type of data.
One of the challenges facing many businesses today is identifying and selecting which internal and external sources of data to focus on. As a result, these companies end up investing in data warehouses full of massive amounts of useless data. To avoid being data rich and insight poor, CFOs need to understand the role of quality in developing and managing tools, data and analytics.
Before inputting data into any analytical model, it is important to first assess the appropriateness of your data sources. Do they provide relevant and accurate data for analysis and interpretation? Instead of relying on a single source of data for analysis, you need to have the ability to blend and analyze multiple sources of data. This will help make informed decisions and drive business performance.
Further, businesses are operating in a period of rapid market changes. Market and customer data is getting outdated quickly. As a result, being agile and having the ability to quickly respond to changing market conditions has become a critical requirement for survival. The business cannot afford to sit on raw data for longer periods. Capabilities that enable data to be accessed, stored, retrieved and analysed in a timely basis should be enhanced.
Thus, in order to provide business users with access to real-time and near-time operational and financial data, the organization should focus on reducing data latency. Reducing data latency allows finance teams to run ad-hoc reports to answer specific business questions which in turn enables decision makers to make important decisions more quickly.
In the event that finance provides business insights or recommendations based on inaccurate data, analysis and predictions, this will quickly erode, if not extinguish, business trust and shake the confidence of those decision makers who rely on these predictions to make informed decisions.
As data volumes increase and new uses emerge, the challenge will only increase. It is therefore critical for finance to put in place robust data governance structures that assess and evaluate the accuracy of data analytics inputs and outputs.
Work with business partners to set objectives up front
Churning out performance reports that do not influence decision making is a waste of time yet that is what most finance teams are spending their time doing. It is not always the case that the numbers that make sense to finance will also make sense to business partners.
The biggest problem in the majority of these cases is lack of understanding by finance of the business objectives. Instead of collaborating with the business to develop a better understanding of their operations and how performance should be measured and reported, many finance analytics teams are working in their own silos without truly linking their activities back to business outcomes.
To improve finance’s reporting outcomes, the function should take stock of the reports it produces per month, quarter or annually. Then evaluate the nature and purpose of each report produced and what key decisions it helps to drive. It is not about the quantity of reports that matters, but rather the quality of the reports.
Business partners need to be engaged at the start of the process and throughout the analytics process. They need to be involved as finance explores the data and develop insights to ensure that when the modeling is complete, the results make sense from a business perspective
By working with business partners and setting objectives up front, finance will be able to focus its efforts and resources on value-add reports that tell a better business story. Further, the function will be able to assess and monitor the effectiveness of its data models in supporting business decisions
Simplify interconnected analytics
With so many variables impacting business performance the organization cannot continue to rely on gut instinct to make better decisions. The organization has no choice but to use data to drive insights. As a result, organizations are relying on a number of interconnected analytical models to predict and visualize future performance.
However, given that one variable might have multiple effects, it is important for the business to understand how changes in one variable will affect all the models that use that variable, rather than just one individual model. By maintaining a meta-model, the organization would be able to visualize and control how different analytical models are linked.
It also helps ensure consistency in how data is used across different analytical models. Ultimately, decision makers will be able to prioritize projects with the greatest potential of delivering the highest value to the business.
Build a data analytics culture
Advanced data analytics is a growing field and as such competition for talent among organizations is high. Due to their traditional professional training, many accounting and finance professionals lack the necessary data and analytics skills.
More over, decision makers not knowing enough about analytics are reluctant to adopt their use. Because of cognitive bias, it is human nature to subconsciously feel that their successful decisions in the past justify a continued use of old sources of data and insight. What we tend to forget is that what got us here won’t get us there, hence the need to learn, relearn and unlearn old habits.
To move forward, the organization should focus on overcoming cognitive biases developed over the years, and closing this skills gap and develop training and development programs that can help achieve the desired outcomes. Using advanced analytics to generate trusted insights requires an understanding of the various analytics methodologies, their rigor and applicability.
It’s difficult to have people understand if they don’t have the technical capabilities. However, building a data analytics culture does not imply that the organization should focus on developing data science skills alone. You also need to develop analytics translators.
These are individuals who are able apply their domain expertise in your industry or function, using a deep understanding of the drivers, processes, and metrics of the business to better leverage the data and turn it into reliable insights and concrete measurable actions.
Building a data analytics culture that promotes making decisions based on data-driven insights is a continuous endeavour that spans the data analytics life cycle from data through to insights and ultimately to generating value. Successful use cases can be used to further drive the culture across the organization.