Analytical Solution

Key Performance Indicators (KPI) for Campaign Targetting

2013-01-28T14:18:00.000+05:30

We have launched the new new campaigns and want to measure the effectiveness by using results. When I tried to do that we observed following metrics (Key Performance Indicators) very useful. I hope those will help you too.

1. % of visits by medium : Does the distribution of our marketing budget match the received visitors?

2. % of goal conversion rate by medium: Which medium is contributing more for getting goal conversions?

3. % of visits by source type: To know any online channel from which visitors arrive at your website.

4. Campaign Quality Index (CQI) : How well targeted your campaigns are at driving qualified leads.

CQI = (% goal conversions from campaign) / (% visits from the same campaign)

Interpretation of the CQI:
a. CQI=1 tells – a visitor from said campaign is likely to convert as a visitor from from any other campaign.
b. CQI= less than 1 (<1) tells – less likely to convert than a visitor from any other campaigns.
c. CQI= greater than 1 (>1) tells – visitor is more likely to convert than a visitor from any other campaigns.

Only Google Can Launch Multi Channel Funnel Report Without Funnels?

2013-01-28T14:12:00.000+05:30

Why Multi Channel Funnel?

Twitter Spectrum : Twitter Topic-Words Association Analysis

2012-08-03T00:18:00.000+05:30

It is a visualization of the topics for two words in the latest results from Twitter. First word is colored blue, the other red, and the associated topics are colored and positioned based on how highly they are associated with the two words. Click on any topic to see the related Tweets from the twitter.

Like this allows you to enter your own terms of interest. Press the 'Enter' key to generate the spectrum after entering your words.

We can use this application for competitor analysis.

When I had chosen India and China I found following discussion topics on twitter and its association spectrum shown below:

Blue and Red color represents the associated topics of those two words whereas Purple represents the common words.

You can click on any topic in the spectrum to see related tweets. My interesting topic Olympics is highly associated with China than India. When I clicked on it to see relevant tweets I found following:

This Spectrum may changes with time due to continuous updates on twitter.

It is created by Jeff Clark in May 2008. He had created News Spectrum first which was visualization of the words used for two topics in the latest results from Google News. He modified News Spectrum application to take data from Twitter instead for Twitter Spectrum.

Bounce Rate: Facts Keep In Mind

2012-07-23T23:26:00.000+05:30

There are almost more than 80% website owners (as per my assumption) are really bothering about the Bounce Rate of the website, and why not? They should. Bounce rate represents the percentage of visitors who enter the website and leave the site rather than continue viewing other pages within the same site.

Here are the few things we need to be considered before jumping into the conclusion.

A) What is the definition of bounce rate used in my web analytics tool

Bounce Rate Definition:

It is usually measured in two ways:

1) The percentage of website visits who see just one page on your site.

2) The percentage of website visits who stay on the site for a small amount of time (usually five seconds or less).

All of the websites are mostly using free tool such as Google Analytics (GA). GA is using bounce rate as the percentage of website visits who see just one page on your site.

B) What bounce rate need to measure for your website

You need to ask few questions to yourself

Q 1. Do you want the visitor who must see more than one page?

This makes sense for sites where visitor need to follow well-defined conversion steps requiring multiple page views. There we recommend using just one page on your site bounce rate definition.

Q2. Do you mind if the visitor is seeing one page on your site but spending more than 5 seconds.

It is value for sites where visitors are likely to find what they are looking for on the entry page itself so they will spend more than 5 seconds on your website. Despite of that if you are using only one page visit as a definition of the bounce rate doesn’t make sense here.

Looking forward for your thoughts/questions. Enjoy new learning.

Track Social Media Marketing - BE SMART

2011-02-25T12:07:00.000+05:30

Suppose you have pages on Facebook, Twitter and LinkedIn as given below:

· www.facebook.com/mysoftware

· http://www.linkedin.com/groups?gid=12345&trk=at_side_k

· http://www.twitter.com/mysoftware

· http://www.facebook.com/pages/oursoftware/123456

· http://www.linkedin.com/groups?about=&gid=12345

· http://www.youtube.com/user/mysoftwareanalytics#grid/user/5A123D2B

· http://www.flickr.com/groups/12345@B23/

You are doing social media marketing through above sources for following website:

http://mysoftware.com/

If you already have implemented any web analytics tool for given web site correctly then you are able to see that you are getting traffic from the following referrals such as Facebook, Twitter and LinkedIn and report will look like this:

Referrals	Visits	Page Views	Bounce Rate
www.facebook.com	874	1224	30%
www.twitter.com	745	934	55%
www.linkedin.com	832	1324	40%
www.flickr.com	214	234	89%

Here we are able to know only the source as referrals but are we sure that all the traffic has come from your community on Facebook (for example)? No.

When we drill down to the report for www.facebook.com it will depict following:

http://www.facebook.com/l.php?u=http%3A%2F%2Fwww.mysoftware.com%2Fnewsoftwares

It means that the visitor came from Facebook has entered on the page www.mysoftware.com/newsoftwares .

The reason behind it is you have given a link of www.mysoftware.com/newsoftwares page somewhere on FB and visitor clicked on it and comes to your website. Also you are not sure that these are your efforts or not. Web Analytics tool will not tell you that this visitor is from your FB community. We need to use our own way to find it out. Tool will track the urls that what we set for visitors.

If you would like to know that how many visitors are coming exactly from those sources you activated on different social sites, you add url parameters to your website url before giving any link on any of the social site.

For example if I add ‘smm=mysoftware’ to my link url such as www.mysoftware.com/newsoftwares?smm=mysoftware then whenever visitor clicks that url or its tiny url link it is been tracked in web analytics tool and you will get to know that these are your efforts only. When you will drill down the referral report it will differentiate your efforts and it will show in the report as below:

http://www.facebook.com/l.php?u=http%3A%2F%2Fwww.mysoftware.com%2Fnewsoftwares&smm=mysoftware

Try it out and let me know your thoughts. I tried it successfully for my web site.

Web Analytics Tools

2011-01-23T22:58:00.002+05:30

As I mentioned in my earlier post there are two ways to track the website visitor data one is ‘Page Tagging’ and another is ‘Web Logs’. If we miss the tagging of javascript code on any of the page of the website tool can’t track visitor data for those pages. As a result of missing tags we didn’t get a clear picture of visitor data. It seems we are analyzing the missing data.

10-20% of visitors are typically missed by web analytics page tags. The embedding of javascript page tags onto every page is a costly, time consuming and error-prone process. Even when page tags are done correctly they miss users due to browser issues, reporting issues, and server or spyware blockers. Also the page tags can even slow down the web page for the user there we need to consider the downloading time of the web page before and after adding tags.

Many companies use multiple web analytics vendors in their company and often on the same site. This leads to some pages having as many as half a dozen page tags that affects the page download and visitor data collection. You will find that one tool shows 2000 visitors so other may show 2300.

There are lots of Web Analytics tools available in the market. Here are few of them which are mostly used and popular too:

1) Unica NetInsight:

For anyone who has ever used page tags in traditional Web analytics applications, chances are the experience was frustrating and time consuming. Fortunately, there’s a better way. Unica’s approach to Web analytics data collection eliminates much of this pain, substantially reducing marketing’s dependency on IT to get started or make changes.

Many other Web analytics software packages require extensive custom JavaScript code to tag pages for customer segments, content groupings, funnel steps, campaign details, etc. In contrast, Unica “soft tagging” approach minimizes the need for code modifications to deliver advanced, personalized metrics and reports. As a result, marketers can easily set these configurations in NetInsight’s visual user interface.

2) Webtrends:

There are few Products and Services they have mentioned separately as:

Analytics: Here it monitors website traffic and applications

Optimize: This model is used for testing, segmenting and targeting.

Ads: This model is used for improving ad performances

Segments: This model profile your prospects and customers

Social Measurement: This model includes four stages such as Listen, Engage, Measure and Decide.

3) Omniture:

Omniture® SiteCatalyst® is now Adobe® SiteCatalyst®, powered by Omniture®. Adobe SiteCatalyst provides marketers with actionable, real-time intelligence about online strategies and marketing initiatives. SiteCatalyst helps marketers quickly identify the most profitable paths through their Web site, determine where visitors are navigating away from their site, and identify critical success metrics for online marketing campaigns. SiteCatalyst is part of the Adobe® Online Marketing Suite of applications for online business optimizations.

4) Google Analytics:

Google Analytics is a free web analytics tool provided by the Google.

1) It has capacity to track your Paid Campaigns.

2) It has the power of alerts you can set for your account according.

3) GA can track mobile visits and E-commerce, industry benchmarks too.

4) It is very popular for its advanced segmentation and flexible customization.

Google also providing a paid web analytics tool called Urchin which is very similar to GA.

Website Tracking - Click Stream Data Collection

2011-01-16T17:15:00.001+05:30

It is important to understand tracking methods and measure the expectations or requirements of the online marketing channels to analyze website visitor behavior and channel’s performance.

Click stream Data Collection Types

* Web Logs – Log files (Tool is known as: Log Analyzer Tool)
* Web Beacons – 1 x 1 transparent images that are placed in web pages
* Javascript Tags – A standard few lines of javascript code (Tool is known as: Page Tagging Tool)
* Packet Sniffing – a layer of software that is installed on the web runs “on top” of the web server data layer.

Page tagging and weblogs are the most popular methods to capture website visitor data. These both methods have their limitations.

Page Tagging: JavaScript code is pasted on each page of the site. The data is collected via the visitor’s browser. Every time the tagged page opened by visitor’s browser the JavaScript is processed and visitor information collected by the cookie.

Advantages of Page Tagging:
1) Tracks events like JavaScript, Flash and Videos etc.
2) Collecting and processing data in real time
3) Provides information on web design parameters such as browser, versions, screen resolution, connection speed etc.

Disadvantages of Page Tagging:
1) Set up errors lead to loss of data
2) Firewall can restrict JavaScript tags
3) Can’t track search engine spiders (robots) and bandwidth
4) Unable to directly track non html pages
5) Vendor specific

Web Logs: Here no need to use JavaScript code for tracking purpose. The data is collected by the web server independently of a visitor’s browser. It captures all the requests made to your web server including pages, images and PDFs.

Advantages of Web Log Method:
1) Historical data can be reprocessed easily
2) No firewall issues
3) Can track bandwidth and completed downloads
4) Track search engine spiders (robots)
5) Vendor independent

Disadvantages of Web Log Method:
1) Proxy and caching affects data collection – if the page is cached, no record is logged on your web server.
2) Requires data storage and archiving to be performed by your own team
3) No event tracking
4) Robots multiply visit counts

If you are using ClickTracks for instance you may observe significant differences in visit numbers. You may get 20%-30% more visit numbers. Standard visit duration is 30 mins. If any visitor gets inactive for more than 30 mins, session gets expire. The change in session duration affects the total number of visits. Therefore it is very keen to know that the basic definitions defined by the tool which you are using for measuring web site data.

Overview of General Analytics

2011-01-09T18:24:00.000+05:30

If you search for the key term “Analytics” in any search engine, there are lots of search results. The simplest definition of Analytics from http://en.wikipedia.org/wiki/Analytics is "the science of analysis". A simple and practical definition, however, would be how an entity (i.e., business) arrives at an optimal or realistic decision based on existing data. Common applications of Analytics include the study of business data using statistical analysis in order to discover and understand historical patterns with an eye to predict and improve business performance in the future.

We need to focus on the questions for which we are looking an answer. Analytics is also a kind of research. The research mainly depends on the research designs. Analytics is based on the two types of research designs.

a) Observational Design

b) Experimental Design

For getting more accurate answer we need to choose the proper design.

Observational Design: All of you have heard about astrology on the web or personal horoscope. They ask you your name, mail id, birth date (sometimes birth time), sex, country etc. Actually what is this information? This is the primary data required to provide your horoscope. This is also a kind of analysis on the basis of supplied data they offer your horoscope.

Analysis of already collected data comes under the observational design.

Experimental Design: In this method you can establish the objectives and processes necessary to deliver results in accordance with the expected output. Web site optimization techniques such as A/B testing and Multivariate testing (MVT) comes under this category. We use factorial experimental design to optimize our website. The process of experimental design is matches with Deming’s PDCA cycle, where P-Plan, D-Do, C-Check and A-Act.

Doing experiments and achieving the expected outcome is a continual process.

We analyze data for the following reasons:

· Testing our assumptions and making decisions

· Predicting the future trends

By doing this we try to find out what happened, what is going on and what should we do to achieve a success.

About Web Analytics

2011-01-02T21:57:00.000+05:30

On March 20, 2007, a search on Google for “web analytics” + definition displayed 642,000 results in 0.11 seconds. On December 31, 2008 a similar search on Google for “web analytics” + definition amazed all, when the search revealed 138,000,000 results in 0.09 seconds. When we looked for the same search on 31, Dec 2010 it depicted 253,000,000 results in 0.07 seconds. In about 2 years, the search results for “web analytics” grew by 83%. This is truly incredible and most interesting revelation!

The birth of Web was in August 1991, but it didn’t become commercial until 1995. Many online businesses boomed, and the use of Internet and web services increased in comparison to many other sales channels like phone and retail etc.

On the net usually online data of the users can be tracked using server Log Files or by adding small JavaScript code on each page of the site, and this was the beginning of a new revolution called Web Analytics.
(Analytics: The branch of logic dealing with analysis or the science of analysis)

What is Web Analytics?
Quite simply the main objective of web analytics is to understand and improve the experiences of online visitors, while increasing revenues of online businesses.
The official definition of web analytics proposed by the Web Analytics Association (WAA) (http://www.webanalyticsassociation.org) is as follows:
Web Analytics is the measurement, collection, analysis and reporting of Internet data for the purposes of understanding and optimizing Web usage.

Here are few reasons of booming of Web Analytics:

Boom in online business:
It is always good to exist online along with offline business. Online marketing costs much less than other. You are easily in touch with your customer at any instance, worldwide while pursuing your own brand. Also 2008 survey shows 55% of US have “always-on” broadband connection leading to a significant change in the way internet is consumed.

Effortless collection of online data:
If you can’t measure it, you can’t improve it. – Lord Kelvin what a practical quote it is! Once you set some basic tags for your site, all visitors’ information get tracked on web analytics tool. Most of the free tools are available in the market.

Easily available data:
As compare to all other channels, online business data will get easily available with lowest cost and time. Tracked information can be available within 36 hours. So you can observe your current data in very few hours.

Understand & take action:
Data is able to reveal instant trends. We can get comprehensive information at a glance by doing analysis of data. Hence we can improve the experiences of online visitors, while achieving the goal of our online businesses.