Previously, four types of analytics were known, namely descriptive, diagnostic, predictive and prescriptive. Descriptive is the simplest type of analytics. This type of analytics provides an overview and summarizes a dataset quantitatively.

Analytical descriptive describes the use of various historical data to draw comparisons. This means that analytics uses data aggregation and data mining to provide insight into the past and answer: “What happened?”

An example is the amount of website user data in one of the websites **Indonesia’s Best Hosting Provider** in every province in Indonesia in 2016. The examples are only meant to illustrate what is going on and do not aim to draw conclusions for a larger population.

Analytical descriptive can be said as a way how data is displayed so that the information displayed can be clearly accepted by others. In descriptive analytic, a data is usually displayed in the form of tables and graphs. The selection of data presentation in the form of tables or graphs is adjusted to the type of data and the objectives to be achieved.

Descriptive analytics uses a series of data to provide an accurate picture of what has happened in the business and how it differs from other comparable periods. These performance metrics can be used to mark areas of strength and weakness to inform management strategy.

The most frequently reported financial metrics are descriptive analytic products — for example, year-over-year price changes, month-on-month sales growth, number of users, or total revenue per customer.

There are several important metrics that are used to provide an overview of the data we have. These metrics have their own (unique) roles and interpretations to describe the characteristics of a dataset. In general, these metrics can be grouped into measures of concentration and size of data spread.

Data

*middle of the data*) is useful for knowing the location of the center of a data. While the size of the data spread (

*range or spread of the data*) serves to determine the distribution of the data, the diversity of the data, and how diverse the data is. Metrics that include measures of data centering are mean, median, and mode.

Meanwhile, the metrics that include the size of the data spread are the minimum and maximum values, range, upper and lower quartiles, standard deviation, and variance. The following is an explanation of important metrics that are often used in descriptive analytics

**Sample size**

Sample size is the number of individuals, subjects, observations, experiments or elements of a sample.

**mean**

Some people refer to this metric as the average or average. This metric is usually the most frequently used to describe the size of the data center. This metric is usually used to answer informational questions, at what value is most of the data at?

This data center measure includes a metric that is sensitive to changes in data (not robust). If a website user data in a** Indonesian hosting provider **is in the range of 200,000 – 400,000 visitors per month and then there is one additional data that is 50 visitors, with this additional data, the resulting average will change drastically.

Another example can be seen from reports on the mean salary and median salary of a profession. Some European and American countries tend to use *median salary* than *mean salary*. Use of this metric (*mean*) needs to be used with caution, especially if a data contains outliers and the data is very heterogeneous.

This metric includes a measure of data centering that *robust*. That is, the size of the center is not sensitive to changes in data (addition or subtraction of data) because the calculation uses the position of a data. The median is also known as the middle quartile or quartile two (Q2). If a data has a median of 100, it means that 50% of the data has a value less than 100 and the other 50% has a value more than 100.

##
id="modus">**mode**

There are differences regarding the meaning of mode in the world of statistics and what is commonly known by young Indonesians. Some people interpret the mode as an attempt to get someone’s attention, but in the world of statistics, the mode is the data that appears most often.

**Minimum**

Is the minimum (smallest) value of a data.

**Maximum**

Is the maximum (largest) value of a data.

**Range**** (range)**

Is the difference between the maximum and minimum values

**Lower Quartile (Q1)**

The lower quartile is usually used to answer the question where 25% of the data are at what value. For example, if the chat time between *customer assistant* and *visitors* in a **Indonesian hosting provider** has a value of Q1 = 130 seconds, meaning that 25% of conversations that occur are less than 130 seconds.

**Upper Quartile**** (Q3)**

The upper quartile is usually used to answer the question 75% of the data is at what value. For example, if the chat time between *customer assistant* and *visitors* in a **Indonesian hosting provider** has a value of Q3 = 200 seconds, meaning that 75% of the conversations that occur are less than 200 seconds.

**Variety**

One of the most frequently used measures of data distribution is variance. By definition, variance is the sum of the squares of the difference between each data to the average. This illustrates that the variance describes the distance of each data to the data center.

**Standard Deviation (Standard Deviation)**

In the definition formula, the standard deviation is the root of the variance. Like variance, standard deviation is also a measure of the spread of data. The difference between variance and standard deviation lies in the resulting units. If a data has units of cm, then the variance of the data will produce units of cm2 while the standard deviation produces units of cm.

Create by Ipadguides in category of Website