By working the data through the complete business analytics cycle<\/a>, the data\u2019s applications will naturally fall into four types or categories of analytics, depending on the question it helps to answer. These four types of data analytics can equip organizational strategist and decision makers to:<\/p>\n\n\n\n Armed with deeper insights and informed recommendations, you can gain a better understanding of your business\u2019s performance as a whole and make better decisions as a result. <\/p>\n\n\n\n This type of analytics is rooted in facts, so think of descriptive analytics as the \u201cjust the facts, ma\u2019am\u201d basics: the reports that outline measures, like performance metrics, and provide the background for most business intelligence (BI) tools<\/a> and dashboards. <\/p>\n\n\n\n For example, a canned (fixed) report provided by your\nsales team may include information on new clients, win rates and goals. You may\nalso use an ad-hoc report to customize the data view to look at only specific\ndata to answer a specific question, such as \u201cHow many customers do we have in\nFlorida?\u201d and \u201cHow much volume did we sell last month in the Florida territory?\u201d<\/p>\n\n\n\n Descriptive analytics can serve as that starting point or base for building your analytics strategy, with some professionals estimating that 80% of business analytics<\/a> involve descriptions based on aggregations of past performance. Techniques like data aggregation, data mining<\/a>, clustering and\/or summary statistics all serve to provide analytics that describe a past state\u2014descriptive analytics. <\/p>\n\n\n\n Like\ndescriptive analytics, diagnostic analytics also focus on the past. However,\nthese types of analyses look for cause and effect to illustrate why something occurred.\nThe objective is to compare past occurrences to determine causes. Of course,\nthis is seldom a black-and-white answer, but instead provided in the context of\nprobability, likelihood or a distributed outcome. <\/p>\n\n\n\n Diagnostic\nanalytics can provide guidance by helping to: <\/p>\n\n\n\n To determine a \u201cwhy\u201d from your data, applying such techniques as principle components analysis, sensitivity analysis <\/a>and training algorithms for classification and regression analysis can all provide a diagnosis\u2014diagnostic analytics.<\/p>\n\n\n\n Predictive analytics is where you start turning the outcomes of your descriptive and diagnostic analytics into actionable insights for decision making by forecasting \u201cpotential future outcomes.\u201d<\/a><\/p>\n\n\n\n In\npredicting future outcomes, predictive analytics isn\u2019t making a judgement on whether\nor not an event is likely to happen, such as, \u201cWill a new competitor join the\nmarketplace?\u201d but instead is describing what will happen if these conditions\nare met, i.e. \u201cBased on what happened in 2016 when a disruptor joined the\nmarketplace, our sales could decline by 20%.\u201d The goal is determining a trend,\ncorrelation, causation or probability. <\/p>\n\n\n\n With predictive analytics, a data analyst uses quantitative analysis<\/a> of the dataset and establishes predictions, often called predictive modeling. This is a broader approach, which aims to characterize predictions, examining models for their accuracy. Confidence intervals, T statistics, K-S statistics and P values are all part of the umbrella of predictive analytics, applied in machine learning algorithms<\/a>, classification models and regression models, to name some examples. <\/p>\n\n\n\n Building these models requires expertise, typically from data scientists, machine learning experts or other data analytics specialists working under a data analytics manager<\/a>. <\/p>\n\n\n\n Prescriptive\nanalytics builds on predictive analytics by helping determine recommended (prescribed)\nactions based on desired potential (predicted) outcomes, helping organizations\nachieve their business objectives. <\/p>\n\n\n\n Prescriptive\nanalytics models are constantly \u201clearning\u201d through feedback mechanisms to\ncontinuously analyze action and event relationships and recommend the optimal\nsolution. By simulating the solution, prescriptive analytics can examine all\nthe key performance criteria to ensure the outcome would achieve the correct\nmetric goals before anything is implemented. <\/p>\n\n\n\n Artificial intelligence<\/a>, machine learning<\/a> and neural network algorithms are often employed to support prescriptive analytics by helping to make specific suggestions based on nuanced patterns and perceptions of organizational goals, limitations and influencing factors. <\/p>\n\n\n\n In the wider view of applying business analytics for organizational success, prescriptive analytics \u201cdelivers business value through recommendations\u201d<\/a> built on the data of results.<\/p>\n\n\n\n These four types of analytics help organizations understand and learn from past patterns and performance in order to improve predictions and actions for the future. Understanding and knowing when to apply the appropriate form of analytics to answer the appropriate business questions can help you to develop needed business solutions and gain a competitive advantage.<\/p>\n\n\n\n Master these complex analytical approaches by advancing your business analytics knowledge and skill sets with MSU\u2019s online Business Analytics Certificates, featuring two program tracks. <\/em><\/strong> <\/p>\n\n\n\n![\"Infographic](\"https:\/\/bisk-preprod.go-vip.co\/msu\/wp-content\/uploads\/sites\/3\/2019\/08\/Four-Types-Data-Analytics-Infographic.png\")
<\/a><\/figure><\/div>\n\n\n\nWhat <\/em>Happened: Descriptive Analytics <\/strong><\/h2>\n\n\n\n
Why<\/em> Did it Happen: Diagnostic Analytics <\/strong><\/h2>\n\n\n\n
What Happens If<\/em>: Predictive Analytics <\/strong><\/h2>\n\n\n\n
How<\/em> to Make it Happen: Prescriptive Analytics <\/strong><\/h2>\n\n\n\n