The A to Z of Analytics: The Complete List of Analytics Terms

The A to Z of Analytics

The world of analytics is vast and ever expanding. With new technologies, tools, and techniques cropping up regularly, the glossary to keep up with becomes a bit of a task. Worry not, we got your back. We’ve compiled an A to Z list of terms in analytics that are commonly used and need to be known to everyone in the field.

The terms are from across the analytics spectrum of data analytics, data science, big data, machine learning. So, let’s get started, shall we?


A is for Artificial Intelligence

You know, the same thing that is apparently set to wipe out humans. Ok maybe not, but Artificial Intelligence (AI) is more popular and mainstream now than it has ever been. AI is being implemented across sectors and is finding various applications such as in speech recognition, visual perception, language translation, chatbots, self-driving automobiles, military and space travel simulations, strategic game competitions (Chess, Go). AI is the next technological revolution that will alter our regular way of life.

B is for Business Analytics

Business Analytics encompasses the skillset, tools, technologies, and practices of being able to explore business data and derive insights for future business planning.

Business Analytics puts an organization on track to becoming data driven. Business analysis uses complex quantitative methods to build business models, also making use of evidence based data.

C is for Cloud Analytics

Cloud Analytics is the application of analytics and statistical tools and techniques to derive insights from data stored in a public or private cloud. Due to the nature of cloud storage, custom tools and techniques are needed to provide analytics processes to the cloud.

Some examples of Cloud Analytics solutions are SaaS Business Intelligence, social media analytics, hosted data warehouses.

D is for Deep Learning

Deep Learning is the cutting edge of machine learning that uses complex algorithms to enable machines to learn by themselves and be able to solve complex and complicated problems across business, data science, mathematics among other things. It is inspired by the functionality of the human brain and is based on artificial neural networks.

Deep learning is currently the gateway to Artificial Intelligence and of the more popular methods of achieving it.

E is for E-commerce Analytics

Ecommerce Analytics is the use of analytics practices as applied to the field of ecommerce retail. Being an online mode of operation, ecommerce generates huge amounts of data across hundreds of data points. This provides the business with information and insights which were earlier not tapped into.

Ecommerce Analytics can help online businesses by giving information on user behaviour, user activity, purchase patterns, user retention, user acquisition etc. among others. These insights help the organization in providing a better experience to its users, thereby driving customer loyalty and purchases.

F is for Financial Analytics

Financial Analytics is used to answer the questions that are specific to businesses operating in the BFSI space. It analyses the company’s financial data to provide insights. It involves working on the methods of fraud analysis/detection, credit scoring, forecasting, risk analysis etc.

Financial Analytics can be used for managing company investments, tracking the performance of the business, market predictions and forecasts, evaluate the assets of an organization among other things.

G is for Game Theory

Though a concept finding origins in mathematics, Game Theory is incredibly useful for data scientists. Essentially, Game Theory is used to analyse situations that are strategic in nature. It is used to gauge the logical and analytical actions and outcomes of a given situation.

The knowledge and understanding of Game Theory fits perfectly with the strategic business problems that Data Science deals with.

H is for HR Analytics

HR analytics is the application of analytic processes by human resource department of a company to help improve the employee performance by tracking employee data. This helps in the areas of employee attrition, retention, satisfaction, and performance.

HR Analytics is fast graining traction and organizations are increasingly implementing it. It deals with the methods/concepts of HR software, compensation analytics, strategic HR management, workforce analytics, talent management etc.

I is for IoT Analytics

IoT is the network of connected devices that constantly interact with each, exchanging information, and executing functions. Currently there are expected to be about 5.7 billion connected devices, with this number set to grow to 20.4 billion by 2020. This massive number of devices generate huge amounts of data and this is where IoT Analytics steps in.

The analysis of IoT data gives insights on both the device and the consumer behaviour. GE predicts that the convergence of machines, data and analytics will become a $200 billion global industry over the next three years.

J is for Journey Analytics

The customer journey and lifecycle with respect to the product and brand is a crucial aspect of running a successful business. Journey Analytics deals with the combination of advanced analytics practices, big data technology, and domain expertise to map out a customer journey that is flawless.

Utilizing several million data points across different buyers, channels of interaction, and touchpoints, Journey Analytics identifies the bottlenecks and highlights the top line. It also helps in optimizing the customer journey for best possible results.

K is for KNN

KNN (k nearest neighbour algorithm) is a simple algorithm that requires a minimal training phase and stores all the data to classify new data based on similarities. As it does not make any sort of underlying assumption about the data and does not focus on generalization of the data. This makes it extremely useful as practical data usually does not follow the theoretical assumptions made about it.

It is one of the most popular and widely used classification algorithms out there.

L is for Location Analytics

Location Analytics is used to analyse geographical data to gain insights for business operations. By adding a geographic component to business data, it provides a brand-new context for understanding transactional and customer data. This kind of analysis provides a way of looking at data through interactive maps which gives more clarity and context as compared to graphs and charts.

By employing thematic mapping and spatial analysis to business analytics, businesses, especially those in the retail, supply chain, and logistics sectors can gain actionable data and insights.

M is for Machine Learning

Machine Learning provides a way for machines to learn without explicit programming. Algorithms are employed to provide machines guidelines. The machines are fed data and use the algorithms to make sense of the data and produce an output that is concurrent with the data and the algorithm. The machines learn and adapt through experience i.e. more data they analyse, the more robust they become over time.

We use machine learning almost every day, though we may not know it. When you book a cab through Uber or Ola, make online purchases, interact with the digital assistant on your phone (Siri, Google Now, Cortana), it is machine learning at work.

N is for Neural Networks

A neural network is a computer system that is modelled on the functionality and operation of a human brain. It enables a computer or machine to learn through observation of data. The reactions, responses, actions a machine takes is learnt by it through iterative learning. It contains many interconnected processing components which work together to solve a problem.

Neural networks can be used to derive patterns and detect trends through imprecise and complex data.

O is for Operational Analytics

Operational analytics is a type of business analytics which focuses on improving existing operations. This type of business analytics involves the use of various data mining and data aggregation tools to get more transparent information for business planning.

Operational Analytics is applied to front line business systems to provide insights to take decisions in real-time operations. This enables businesses to optimize for efficiency and cost, and also to streamline their operations.

P is for Predictive Analytics

Predictive Analytics employs practices from modelling, machine learning, data mining, AI, data mining to perform analysis on a data set to predict future actions based on it. For example, it is used widely in the finance and retail sectors to predict consumer behaviour and business growth based on historical data across thousands of data points.

In this method, existing data governs future actions based on the probability and plausibility of specific actions being taken.

Q is for Q-Learning

It is a type of reinforcement learning, which is a method to deploy machine learning. In this, the machine learns the optimal policy from its history of interaction with its environment. Based on the actions done and rewards received, it learns how to make the most optimal move to get the best reward.

R is for Retail Analytics

Retail Analytics corresponds to data analytics practices as applied to the field of retail. A buyer or customer has various touchpoints with a retail business, and retail analytics analyses all the actions at different touchpoints to give insights into purchase behaviour, spending capacity, demography, product loyalty, brand loyalty, brand recognition, customer retention, customer satisfaction, sales among other things.

With retail chains dominating across fields like fashion, FMCG, appliances, electronics etc., the retail sector most widely uses analytics for business decision making.

S is for Supply Chain Analytics

Supply Chains are domains which are complex in nature owing to the touchpoints and information exchange involved in the process. Application of analytics to these systems helps in the better management and design the supply chains for operational, cost, and functional efficiency.

Analysis of the available data to gauge factors such as fuel, storage costs, delivery times, sourcing costs, logistic costs etc. gives an organization a better picture of areas where improvement is needed and how to optimize for maximum output.

T is for Traffic Analysis

Traffic analysis is the process of intercepting and analysing messages to draw information from patterns in communication. It can be performed even when the messages are encrypted and cannot be decrypted. This is done to check the performance, security, and network operations.

It checks factors such as network speeds, network utilization, type and size of data being transmitted, and can be used to monitor and trace for malicious data streams and packets.

U is for Unsupervised Learning

Unsupervised learning is a method of machine learning where the data provided is not labelled and the machine learns to analyse this data to find a hidden pattern or structure to it. Here the machine learns without explicitly knowing what it needs to do, by focusing on analysing the data and trying to find the commonalities in it, and deduce an overarching pattern to the data set.

It finds application in image recognition, genetic clustering, pattern mining, object recognition etc. Unsupervised learning works to uncover deep lying patterns that are not instantly noticeable or are not visible.

V is for Visualization

Visualization is the representation of data analysis in a graphic or visual format where it is more readily understandable and consumable. Analytics doesn’t end at number crunching and insight gathering, these have to be represented in a way that caters to a business context.

In recent times, visualization has become key skills for analysts as at a business decision making level, a pure number based insight will not make much sense or headway. Visual representation aids in making the data more comprehensible.

W is for Web Analytics

The internet superhighway is the biggest contributor of data, and is in fact the biggest factor behind the rise of Big Data. Web Analytics is applied to internet data for analysis of different behaviours and detection of certain patterns based on demography, usage, location etc.

This type of analysis helps in understanding and optimizing of web usage. With the entire world online, analysis of website data is crucial to how a business can improve its performance.

X is for XML Databases

An XML database is a data persistence software system that allows data to be specified, and sometimes stored, in XML format. As the use of XML is increasing in every field, it is required to have a secured place to store the XML documents.

Y is for YARN

Apache Yarn is an open source cluster management technology that is employed in the analysis of Big Data. It can be seen as a as a large-scale, distributed operating system for big data applications.

Z is for Z-score

It is one of the most important statistical measures in analytics. It measures the score’s relationship to the mean in a group of scores. It essentially states how many standard deviations an element is from the mean.

So, there you go, this is the A to Z of analytics terms. Read up, and you are already one step further in your analytics journey.

Related Articles

Please wait while your application is being created.
Request Callback