EN – Web analytics and Predictive analysis

EN – Web analytics and Predictive analysis

Analytics helps business managers and executives to improve the business and attract more and retain existing customers. We illustrate the value of web and predictive analytics.
Successful organizations today are using the power of web analytics to realize the full potential of their web sites, developing and maintaining deeper client relationships that create measurable value to the business. In this document, the detailed step by step approach to web analytics is discussed. And, analytics tool Google Analytics Report is clearly explained.

Web analytics and Predictive analysisWhat is Analytics?

Analytics is the systematic computational analysis of data. Analytics will analyze the past historical data with the current data to create meaningful patterns.

What is Web analytics?

Web analytics is a procedure of measuring, collecting, analyzing and reporting of Internet data to optimize the business processes and maximize their revenue. It focuses on in-depth comparison of available visitor data, referral data and site navigation pattern.

Web analytics Data capturing methods:

A web analytics data is captured in either of the following categories

Web Logs

The data here is captured in the form of a log file which gets created when a visitor request for a web page. When we type the URL in the browser, the server provides us that page and simultaneously creates a log of it.

Web Beacon

This measures the hit by a user on an advertisement, banner or any graphics that pops up while opening a web page. The usual form of such data is the image on which the click is made.

Java Script Tags

This is the most popular way of capturing the web data. When an URL is requested, the server furnishes the information along with a separate java script code embedded in it. The moment the page loads, the script gets executed and automatically sends the visitors information to a third-party host server.

Click Data Stream

Click stream data analysis can be defined as ‘a process of collecting, storing, processing and analyzing the data based on the pattern and behavior of a website visitor.’ Such data tracks the path of the visitor while navigating a web page. It includes record of every site visited and every page seen, the amount of time spent on each of them, the order in which the visits were made, the bounce rate of the site and much more. Usually, such data points are captured through in-built click stream tools like Google Analytics, Web trends, Analog, Net Tracker etc.

What we can do with collected data?

Any sphere of business has variety of questions to be resolved. A proper analytical approach provides an appropriate answer to those questions. Let’s see the major questions in the world of e-business which are well answered by web analytics

  • Who are the visitors and how frequently they visit?
  • What do they do they do and is there any pattern in their visit?
  • Are the visitors satisfied with the site? What more they expect and what is not attracting them?
  • Finally, how can more and more visitors be attracted?

Process to collect the Data from a webpage

Step 1: The user requests for a browsing page.

Step 2: The request, in form of an HTML page, is sent to the host server.

Step 3: The server sends back the requested file to the web page & it is displayed.

Step 4: Simultaneously a JavaScript, which is embedded in the page, also get executed.

Step 5: The tracking system, on receiving the script, sends back a pixel tag to the page.

Step 6: The page in turn sends the page view details to the tracking server.

Step 7: In the meantime, a cookie is sent to the user through which his information is collected.

Step 8: Finally, based on the data collected, the in-built tool generates the report

To generate the report, we need built in tools called as “Web Analytics tools”, we illustrate the Web analytics tools below

Web analytics tool

To generate the report, we need built in tools called as “Web Analytics tools”, we illustrate the Web analytics tools below.

Stakeholders are interested to use web analytics for various reasons, prime being its ability to draw insight from the numbers, the organizational structure that is offering such services and the ease-of use of the tool.

The web analytics tools can be broadly classified in two parts Conversion and usability based on the features:

Conversion

It measures the ratio of visitors who convert casual content views or website visits into desired actions based on direct requests from marketers, advertisers and content creators. Some eminent players helping in assessing the conversion rate are as follows:

  1. Google Analytics – Provides market based analysis with an excellent graphical representation. Gives a live view of what visitors are searching for and allow filtering click count from any specific domain. It is flexible in choosing the parameters and generates customized reports
  2. Web Trends- It uses navigating behavior between pages to generate report. It helps to generate a comparative study between new and repeat visitors and generates reports based on demographic traits.
  3. Compete- Its utility is mainly based on competitor’s analysis by comparing one site’s usability to other. It can generate report for maximum of 5 website competitors and rank them based on the KPI’s. It also generates alerts based on site’s popularity and discount offered in the site.
  4. Stat Counter- It offers charts and graphs for a better visualization of the statistics. Provides drill down analysis and its magnifying tool helps in zooming into individual visitors. It helps to set up multiple projects at a time.
  5. Quant cast- It accurately measures the visitors’ information and helps in attracting advertisers. Generates graph for demographic variables. It also offers a free service that measures the usage of video, widgets and games. Finally, it can generate one sentence summary.

Usability

Certain web site tools focus on calculating the usability of the site. There are two main tools that do such a job.

Click Tale

It gives the analyzer to watch movie of user’s individual browsing session it records every movement of the mouse, click and keystroke are tracked for better viewing. With such a tool the webmaster can improve usability and effectiveness of the site.

Crazy egg

It mainly focuses on capturing the page visited and its frequency along with volume of the click. The in-built feature ‘overlay’ count the number of click on every link or URL that is visited. Another feature ‘Heat map’ compares the intensity of site visited by giving different color coding to them based on the volume.

Apart from the one mentioned above, there also exist more tools in the market which are free and have an excellent application to gather and analyze web data. Piwik, Firestats, Snoop, Yahoo Web Analytics, BBClone, Woopra, 4Q and Grape Web Statistics to name a few.

What is Google Analytics?

Google Analytics Report is the web analytics solution that generates detailed statistics about the visitors to a website. It has flexible and easy-to-use features which will let you see and analyze your traffic data. Google Analytics can track visitors from all referrers, including search engines, display advertising, pay-per-click networks, e-mail marketing and digital collateral such as links within PDF documents. We can extract many reports based on our needs/objectives. To get the access to GA reports, send a mail to the Digital Marketing team mentioning your Gmail id. You will be given access to Google Analytics within 48 hours. Once the access has been provided, log on to https://www.google.com/analytics with your Gmail id and Gmail password.

Terminology used in Analytics tools

Page view

A page view is recorded every time a page is viewed. Or, more technically, every time the track Pageview method is executed in the script. We generally say it as ‘Hit’ but this is a wrong term so please does not use this. It should be ‘Page view’. When a visitor hits the back button, a page view is generated. When a visitor hits refresh, a page view is created. Every time a page is opened in the browser, regardless of whether it has been cached, it generates a page view.

Visitors

Defined by a unique ID set in a visitor’s cookies. Whenever the tracking code is executed, it looks for cookies on the browser set by the current domain. If they can’t be found, new cookies with a new ID are set. Google Analytics emphasizes visits over visitors because of the inherent inaccuracies of trying to track individual users. For example, a visitor who deletes their cookies, uses multiple browsers or shares their computer will show up inaccurately

Bounce

A visit with one page view. It doesn’t matter how long the visitor was on the page or how they left. Technically, it’s a visit with only one interaction. Any Event Tracking calls would negate this. It’s just a one-page visit.

Time on Page

Time on page is measured by subtracting the time a visitor hit a page from the time they hit the next page. (e.g. If they hit Page 1 at 12:00 and hit Page 2 at 12:03, time on Page 1 is three minutes.) This means that the time on page for the last page in a visit is always zero because Google Analytics has no way of tracking when a visitor closes a page.

Time on Site

This is the sum of all times on page for a visit.

New Visitor

A visitor who did not have Google Analytics cookies when they hit the first page in this visit. If the visitor delete their cookies and come back to the site, they would be counted as a new visitor.

Returning Visitor

A visitor with existing Google Analytics cookies. Those cookies typically will also record the number of visits the visitor has made to the site.

Direct Traffic

Ideally, this is the traffic that came to a site via bookmarks or by directly typing in the URL. It is the traffic for which the code couldn’t determine a source. Depending on the site and the browser, some links may not show a referrer and instead would be categorized as direct.

Pages/Visit

Page views divided by visitors.

Extract GA reports for any website/application:

The Dashboard is the first report you see as you enter Google Analytics and it gives you an overall picture of the major metrics for your website.

The dashboard includes the following metrics for a given time period

Site Usage Information:

  1. Visits
  2. Page views
  3. Pages / visit
  4. Bounce rate
  5. Average Time on Site
  6. New Visits
  7. On mouse-over you can find the data for a particular day.
  8. You can change the date by clicking the arrow in the right side of the date range and extract the data accordingly.

Visitors Overview

Total number of visitors who visited your website during the selected period (included multiple visits by a single person). You can click on view reports to get further details about this metric

Content Overview

This gives you the details for each page on your site. On clicking overview, you will find detailed metrics which include:

  1. Page Views
  2. Unique views
  3. Bounce rates
  4. Top content (this gives the data for every page on your site ranked by the number of page views)

What is Predictive Analytics?

Predictive analytics uses existing data and extracts information from that data in order to determine patterns and predict future outcomes and trends. Predictive Analytics does not tell you what will happen in the future.

Benefits of Predictive analytics

Many of us have wondered why earthquakes cannot be predicted the same way as cyclones or tsunamis are predicted. A huge number of parameters that have to be looked at for arriving at a conclusion, make predicting earthquakes a near impossible task. While past data is useful, predicting an earthquake is immensely difficult as it means analyzing not only past data but also predictors which show the probability of an earthquake.

Parameters are typically looked as it is not possible to crunch every parameter and arrive at an insight. To improve the accuracy of predicting earthquakes, many parameters should be brought together and analyzed.

Earthquakes can be predicted 20-30 days before they occur. This is possible by analysis using big data and Satellite technology – Proved Successful by Terra Seismic
Apart from Big data and satellite technology scientists must consider the parameters like

  • Ground water level.
  • Bizarre behavior of animals and fishes.
  • Variations in seismic velocity.
  • Sudden cloud formations.

The company Terra Seismic successfully predicted few earthquakes.

  • Example. Apr 5, 2013 the firm issued a forecast for Japan.

Predictive analytics advantages in Insurance

Predictive analytic techniques allow insurers to better understand their data and how to use it to predict future events. Proper implementation of predictive analytic techniques can improve an insurer’s consistency and efficiency in products development, help to define target markets and market selection, increasing the number of policy price points “rating”, underwriting and claims processing, fraud detection and segmentation for proper attention and action

  • Helps marketing department more precisely identify potential policy sales through analysis of customer purchasing patterns
  • Reduces the employee hours’ underwriters may have spent researching and analyzing an applicant who ultimately is not a desired insured
  • Provides predictive modelling scores for applicants that can be used as a rating mechanism for determining a variety of policy price/product points
  • Helps identify potentially fraudulent claims
  • Scores claims based on the likely size of the settlement, enabling an insurer to more efficiently allocate resources to higher priority claims

Conclusion

Analytics are very useful in areas such as Insurance, Banking and media. Analytics will help business managers and executives to improve the business and attract more and retain existing customers. Analytics tools like Google Analytics will be used to measure the performance of a website, in turn helps to improve the business. As the market leaders put forth the effort to transform data and apply analytics with increasing sophistication, the demand for predictive analytics will sweep the property and casualty insurance industry to new heights.

References

http://www.google.com/support/googleanalytics/?hl=en_US
https://en.wikipedia.org/wiki/Analytics
http://blog.clicktale.com/2010/11/17/a-brief-history-of-web-analytics/

João Barros

joao.barros@bconcepts.pt