In previous episodes we explored how the role of the Finance team is transforming into an insight generating engine that fuels business performance management and the decision-making process. We also looked at different types of analytics (Descriptive, Diagnostic, Predictive, Prescriptive) based on use cases relating to revenue optimization strategies. In this episode, we tap into what it takes to build a Business Performance Analytics capability, and why some organizations are more successful than others.
Business Performance Analytics is a cross-functional discipline aimed at improving a company’s performance towards achieving its objectives. It includes activities like:
- Analysis of past performance
- Identification of relationships and patterns in data sets and performance indicators
- Developing a hypothesis about data interrelation, cause-and-consequence linkages, and generating a view of future performance
- Proving the hypothesis right or wrong, and iteratively adjusting the hypothesis as needed
- Finding the optimal values or parameter relations to drive the required outcomes
- Formulating actions that will enable the achievement of these outcomes
On the surface this may sound quite similar to what most Financial Planning & Analysis (FP&A) teams have been designed to do for several decades. Because at the end of the day, the aim of every business is to increase its value, be that through revenue growth, cost optimization, cash flow improvement, Mergers & Acquisitions, sustainability strategy, etc. The key differentiators that distinguish Business Performance Analytics from traditional financial analysis are the volumes of available data and the ability to capture, store and process these data using the latest technologies. As a result, even though they might be aimed at answering the same business questions as before, the generated insights go much deeper, thus becoming more:
- Detailed – due to a larger number of data points that were not previously available,
- Accurate – thanks to more sophisticated algorithms and the possibility to exploit different types of data: financial, operational and even unstructured data,
- Fast – due to the greater processing power of modern technologies, and
- Reliable – due to an ability to iteratively adjust the logic, perform multiple tests, and leverage the scale of digital solutions.
'More detailed, accurate, fast, and reliable insights: the game-changers of today’s business performance'
Take a common FP&A exercise: cost control. The traditional approach would be to break down costs into categories, analyze the period-on-period evolution, run a comparison of actual costs with budget and/or forecasted values, and calculate some ratios. More advanced teams would also interact with the cost category owners to understand the dynamics, key drivers and expected trends. However, this rarely goes beyond an analysis of cost data that are available in the company’s Enterprise Resource Planning (ERP) system, the quality and complexity of which may vary significantly depending on a number of factors, e.g. sophistication of the cost accounting methodology, granularity of reporting, transactional data quality, etc.
Analytics-driven approach to performance improvement
Business Performance Analytics offers so much more. Beyond technical advancements that allow better visibility on costs, the ability to analyze in much more detail at speed, and the possibility to connect financial data with operational and even unstructured data (text or images) mean that an entirely new approach to performance management can be employed. Take for example recurring equipment maintenance expenses. Instead of shaving off costs across the process, what if you could channel costs into areas of highest risk, so that you only need to spend maintenance effort on equipment about to break or fail. The state of equipment could be identified via sensors that indicate under which operational conditions it is working and hence the probability of an upcoming failure. This will result in more precise resource allocation and overall cost reduction, while at the same time reducing the risk of interruption and downtimes.
Discover how a global oil and gas company was able to generate value from data by implementing predictive analytics for maintenance and process-control in its production operations.
Other (non-exhaustive) examples of using analytics to optimize costs and improve business outcomes include:
- Reducing transportation costs by optimizing the transportation route network and making decisions about what is shipped from where
- Reducing inventory levels by improving sales forecasts and supply planning
- Reducing inventory write-offs by identifying potentially obsolete items
- Improving customer service levels by proactively spotting potential issues in the supply chain and taking action before customers are impacted
- Reducing marketing spend (while simultaneously improving marketing campaign effectiveness) by targeting campaigns only at customers who are ‘persuadable’ (and skipping those who are either likely to buy a product without marketing, or unlikely to buy even when approached).
5 crucial elements to succeed with Business Performance Analytics
There is no doubt that Business Performance Analytics can only be built on a strong foundation of robust data governance, an integrated approach to data management and analysis, and an insight-led decision-making culture. Besides, before thinking about the HOW of implementing analytics, organizations should first be clear on WHY they are implementing it. The importance of linking the expected analytics outcomes to business objectives can’t be overstressed.
Let’s take a closer look at five elements that are crucial to unlocking the full potential of Business Performance Analytics.
Link with business objectives
Which business objectives and business challenges need to be addressed by analytics? Without making this link, analytics become a pure ‘data juggling’ exercise that won’t support decision making on how to achieve a company’s goals. When these goals change, the analytics strategy and methods must be scrutinized to ensure that they are still fit for purpose. In order to generate meaningful insights, data used by analytics must always be designed and generated with the business objectives in mind.
Data are the ‘bread and butter’ of analytics. This means getting the right data (data availability), ensuring the data are relevant to business needs (as explained above) and reliable (data quality). As our recent CFO research shows, seventy six percent of CFOs agree that without ‘one version of the truth’ across business units, their organization will struggle to meet its objectives.
Ensuring the right data are in place can require changes in underlying processes. In the example given above of predictive maintenance analytics, these changes could include adjusting processes to generate and capture data (e.g. sensors to measure equipment performance and a system to record and store the values), defining thresholds and alert criteria to trigger an action, and even adapting workforce planning procedure to factor in the insights generated by the model. This means that analytics should be considered as part of the broader business context, involving all the processes that will provide inputs, as well as processes that will use its outputs.
When it comes to drawing conclusions from the analysis of large volumes of cross-functional data, stakeholder alignment and end-to-end integration are of the essence. This includes the alignment of data definitions, assumptions and hypotheses, interpretation of outcomes, and their impact on business goals. Since many business decisions are steered by the positivity of the underlying value case, Finance will often find itself the best placed team to then translate the outcomes into financial value terms and link them with the strategic objectives.
Insight-led decision-making culture
With the right data in place, the analytics model built and the team ready to drill into the business challenges, how do you ensure that the insights generated will actually be used to steer decision making? Traditionally, long-term and strategic business decisions are taken based on the knowledge and experience of management, and more often than not, on the ‘gut feeling’ of business experts. Changing this to a data-driven approach requires management to understand the assumptions used in the analytics model, the logic used to draw a conclusion, and the connection between the results and business outcomes. Lack of awareness and understanding often results in a lack of buy-in from the leadership. A data-driven decision-making culture is not built overnight. It requires commitment, investment in skills and people, and an experimental mindset. The best way is to start small, build up knowledge, make sure a common language is established between business stakeholders and data scientists, and then expand the approach to more complex business situations.
Feel free to get in touch if you want to learn more about Business Performance Analytics and to discuss how it could support your organization’s objectives.
Analytics represent just one component of Finance’s digitalization journey. Is your Finance team on its way to becoming a #DigitalLeader? Find out by using our Finance Digital indicator.