Diagnostic analytics uses analytics to describe why something happened. Diagnostic analytics takes the development of descriptive analytics a step further and is generating a root cause analysis to see:
- What caused the company’s sales last month, last quarter, last year to increase or decrease?
- Why were there changes in the company’s largest customer?
- Why were there changes in customer profitability
- What caused changes in the company’s largest vendors last month, last quarter, last year?
Using descriptive analytics to know what happened is important, but it is usually not enough to make certain decisions on future events, thus the use of diagnostic analytics. Diagnostic analytics generally look to:
- Identify outliers and anomalies – Upon review of descriptive analytics, a company must look for events that transpired without a reason or large transactions that are skewing the analytics. An example would be a customer creating a large purchase as they got additional debt or grant funding. This purchase might be outside the norm for the company and could be considered an outlier/anomaly.
- Drill-down into the data – Next the team will drill-down into the data and the analytics to look for patterns and trends that would explain why an event happened.
- Establish causal relationship – The team will begin to look for relationships that are directly caused by the occurrence of other events. An example would be the increase in website traffic could be correlated to additional ad traffic or a change in the company’s SEO methodology. Identifying causal relationships can be done with multiple methods but can include time series and regression analysis and probability theories.
Prior to the implementation of technology and usage of Big Data, diagnostic analytics were a manual process but now a company can introduce machine learning with software tools to assist the analytic team. Software can complete the steps above quicker and more efficient and have more capabilities in identifying patterns, anomalies, or trends. The implementation of software to complement a team can reduce correlation as causation and unintentional bias of the team. The use of software and specific methodologies allows for a company to answer Why Did it Happen?
We’re here to help you get the most from your data. Please contact us with any questions or to get the ball rolling on building your own data analytics dashboard.
Kevin Bach, CPA