VijayWeb.com

When we all go to Search Engines like Google and do a search - we obviously expect the relevant results.

What should Google deliver as the results in order to satisfy the searcher’s needs. How would Search Engines know exactly what the searcher is looking for with a mere keyword that you have provided in the search text box? Have you ever disappointed with the results that Search Engines deliver for you?

Imagine I went to Google and just typed a keyword phrase “web analytics” in the Google’s search text box and just hit the “Search” button - Google does know that I am looking for something related to web analytics which is evident from the search keyword I have just entered in the search text box and it can also know since I am doing this search geographically from India, it would know I may be looking something related to “Web Analytics in India”. But, how does it know exactly what I am looking for?

Am I looking for what is web analytics or what it means? Am I looking for how web analytics works? Or Am I looking for where in the world this web analytics stays Or Web Anlaytics Tools Or Web Analytics Vendors Or am I just looking for Web Analytics Blogs or Articles on Web. When we target a generic keyword phrase like “web analytics” there could be hundreds of intentions could be related even from Search Engne Marketing (SEM), Search Engine Optimization, Web Reporting & Web Analysis, and many more online marketing functions.

How would Google going to satisfy my search intention by providing relevant results which it always says it would when I can relate my intention to many of the above mentioned web analytics topics. What do you think that Search Engines would give me as the top 10 results that would satisfy my search need? What happends if Google give me results about Web Analytics Companies but I am looking for Web Analytics Reporting Techniques, what happends if Google gives me results on Web Log Analysis Software but I am looking for Web Analytics Jobs in United States - so it never matches.

The answer is Search Engines provide relatively different results that means it would provde a mixture of all the topics that can be highly related and commonly searched. The sites that can answer most of the search intentions or provide links to the resources can get the advantage to be ranked in the top 10 results of the search query.

The Search Results page for the keyword “Web Analytics” would normally look like 1 or 2 results on what is “web analytics” and 1 or 2 results from web anlaytics tools or vendors or web analytics companies and 1 or 2 results that can provide links to most of the web analytics topics such as a wikipedia or a directory listing.

But, that may not be the right always for satisfying the searcher needs - the probablity of not satisfying the searcher’s intention would always be 50%, that is the reason Google gives related searches at the bottom of the results page as you see in the above screen shot. When we do more specific search we many not get the recommendations from Google on Related Search but how Google ranks the top 10 results would remains the same - “Relatively Different” and unique content that can answer many of search intentions or provide links to the similar resources would always have an advantage to be ranked in top 10 results.

Lot of times, I have seen SEO’s spending time in analyzing the competitors results and their keyword relevancy and proximity to guess what works for optimizing their web pages (Tools like Webposition Gold, WebCEO provide similar analysis and reports) but I personally say a big “NO” and it is recommended to provide unique but relative content on the web pages. If we try to provide the same content that is already ranked in the top 10 there is likely to be filtered from top 10 results for that search query unless we convince Google that you have better content than the competition on the same topic.

Click Here to read Formula for Search Engine Optimization

Posted on February 17th, 2008 by admin
Filed under: Search Engine Optimization (SEO), Web Analytics | No Comments »

Committed Visits -  Every visit to the website has an exit, but every visit to a web page may not have an exit.  A visit to a web page is calculated a count of “1″ everytime a visitor visited the website and stayed on the site for a defined amount of time and never visited again before the session-cookie is expired in the visitor’s browser.

Looking at Visits alone can not give the right perception of the metrics data, especially when you are looking at Campaign Reports and making analysis of it, it can give a wrong perception of the data.

People visit the website for many reasons which also include people visiting websites mistakenly,  people visiting website that is irrelevant to them.  Though it is important to measure the bounce rate on the landing page, it is also important to measure the qualified traffic that is entering into the website and measuring various success metrics in using them in place of actual visits.

For example, a holiday campaign for Widgets.com -  Conversion Rate is 1.5% and Landing Page Abandonment Rate is 70%.  Conversion rate is generally calculated between Visits and total orders from the website. To get a better idea on the website performance - It is highly recommended to consider the committed visits and calculate Conversion rate for them in other words, Calculate the success metrics using committed visits. That way we will be able to measure the performance of the web site since we are removing untargetted from the analysis for web site performance.

Calculating Committed visits is as simple as subtracting Singe Page Visits from Total Visits.

 Time2Click - Time2Click is the metric that tells the average amount of time that was taken in order to click a link on a web page.  This metric helps web masters and marketers on how much time at an average, visitors are taking for making a decision to click a link or button on the web page.  This metric is simple to calculate and great use to optimize the call-to-action buttons, positioning, anchor text in the links and much more without using any other expensive Split-Testing or multivariate testing tools.

Time2Click metric is similar to the metric “Average Time Spent”, however, “Average Time Spent” will only tell us on how much time at an overall, an average visitor spent on any given web page in the website.  Average Time Spent gives the perception on how effective the web site content/design and/or any problematic areas.  But, Time2Click metric gives the perception on how effectively the other web pages referring the given web page and the effectiveness of the content that is recommending the users to click on the call-to-action buttons or links.

Average Time Spent metric is available in almost all vendors and even the standard log analyser tools.  However, Time2Click metric can be calculated by taking the average time spent on the referring pages or previous pages in simple words this is the average time spent on the previous page.

Time2Reach is the metric that tells about the total time that was taken to reach the given web page.  This metric helps the web masters and marketers to find out amout of time that is taking to reach a specific web page or a conversion point.

Calculating this metric is quite simple and easy - Summing up time-spent on all previous pages for a session to the given page.

Similarly we could report metrics like Time2Convert, Visits2Convert etc., (pls. note that always try to remove outliers as much as you could like I briefed in Committed Visits).

Posted on December 15th, 2007 by admin
Filed under: Web Analytics | 1 Comment »

Today, I was talking with someone who is a webmaster managing various kinds of web sites. He has one specific question for me which I thought I should talk about it in my blog post.

“How do I Measure the success of a web page?”

I was a bit paused when I hear this question. And, that’s really an excellent question that should be looked into.

Well, when doing SEO I always see, though there are many common characteristics among the web pages, they always have a unique personality like individuals. Because of many uncontrollable parameters, a web page has unique personality which is what a web page with the same kind of content and links can be treated differently in search engines, even by the users sporadically. Perhaps, I will speak more on Web Page Personality in my future posts, now I really want to deal the question I was asked on “Measuring the success of a web page”. Please note, I am not talking about success of a web site but success of an individual web page.

Let’s take a real case as an example, have you ever advertised on Google, Yahoo or any other publisher’s sites or you might be getting some traffic to your website thru whatever source. Do they enter your website thru one of the web pages from your site which we call as entry page or gateway page. When the web site visitors land on one of your campaign landing pages or website gateway pages what would you expect them to do?

1. Buy a Product - Do you agree with people buy products/services as soon as they land on your website?

2. Fill-in a Questionnaire/lead generation form - Do you agree, people drop their email addresses and other personal info. as soon as they land on your website landing page.

3. Explore the website - Do you agree, visitors land on your campaign landing page and enjoy the user experience of your website or is it that you expect them to do?

Pls. ask the above 3-questions to yourself, the answers slightly vary based on what you are trying to achieve with your online campaigns or your website.

But, as soon as visitors land on a landing page or a website gateway page - within the first second they try to ensure themselves they landed on the right page. And, try to smell the scent that was lured them to come to this web page i.e. find the relevant text, links or images that help them to ensure themselves. If they can’t find the relevancy to their need of visiting your landing page…the best they will do is “Abandon Your Website Landing Page” and move their journey back to search engines.

So, the success of Campaign Landing Pages and Other Website Entry Pages is purely depended on how many people clicked on and get-into the next page of your website. These pages are typically I call as “Directional Pages”

A majority of the landing pages and gateway pages are “Directional Pages”. The primary duty of these directional pages is to direct visitors to the right section or source of information on your website. These pages include Home Pages, Site Maps, Product Listing Pages etc., The goal of all these pages is not to “Sell the products or services” but direct visitors to the right section where they find information or even buy something. Hence, the success of the Directional Pages is measured by no. of people getting into the websites. I use a metric called “Committed Visits” which is Total Visits to the page subtracted by exits at the page.

That is all about measuring success for Directional Web Pages. Now I will talk about another kind of web pages - Descriptors.

Descriptors receive visitors from Directional Pages. The goal of the descriptors is to talk to the visitor on your products, services or support in terms of giving an overview about your product, talking about the product features, provide the pricing, how your product works, how is it different from your best competition etc., Each descriptive page has one call to action (infact, several in many cases). On product overview page, you may call the visitor to read about benefits of the product, on the benefits page - you may call the visitor to learn about how is it different from your competition and testimonials. You would need to find out on which point you would need to call the visitor to make the ultimate action i.e. “Click Here to Buy Widgets Online!”

The success of these Descriptors is measured on No. of times the website visitors have taken the action on “Defined Call to Action”. For example - No. of times viewed the product flash demo, no. of times people clicked on Benefits link in Product Overview page. You can mix and match while defining these call to action even combine a few of them for each descriptive web page.

Lead Generators/Converters are the web pages whose responsibility is to convert the prospect into the defined goal. These are the  web pages that generally upsell your products or services, these are the web pages that pitches about the benefits of your product, offers, testimonials etc,

The success of these web pages is measured by associating the visits to these pages to the no. of goals achieved. For example, out of 100 orders that you made today check how many of them visited these converter pages prior to their purchase at your website.

I was recalling myself when I was learning computers, the teacher was telling computers is all about taking Input from the users and Processes (CPU) to give a defined Output. Similarly, we do have a list of pages that come into this category “Processors” (let me know if I could coin a better word

Processors are the web pages whose sole responsibility is to receive the information as Inputs and process it and give the defined output to the users. These are the pages typically we have in any website like login forms, registration forms, shopping carts, payment forms etc.,

The success of these processors is measured by no. of people started the process and completed the process successfully. The best example is web site shopping cart funnel. The success of shopping cart funnel is measured by looking at the people started adding products to cart, to checkout to payment information to our best friend thank you page or order confirmation page.

Similarly, we can measure the success for a registration form, login form etc.,

Posted on November 22nd, 2007 by admin
Filed under: Usability, Web Analytics | 1 Comment »

There are ways to check where your website ranks compared to your competition or in a panel of websites thru alexa or compete.com… But, do you know if there is a way to check what’s your website reach over total internet traffic.

Right, there is a simple way to calculate website reach on our own. For this purpose, we would use two kinds of data. The first one is your website yearly unique visitors count which you can get it from your log file or analytics system you are using and the second data, we would need is the Internet Usage trends which you can get it form http://www.internetworldstats.com/stats.htm

All we do is, relate the Yearly Unique Visitors to the Global Internet Usage to get the global reach of your website. We can also do this simple calculation for knowing region wise reach or even at the country level. For that, we may need to do a little research locally to find out the Internet Audience count in the country you are targetting.

Just to illustrate better, if widgets.com is targeted for NA and we could get the overall NA internet audience count from the above site, which is 232,655,287 for the current year and imagine your website gets about 2.5 million unique visitors in the same year. That tells us the reach of widgets.com in North America is 1.07% We could show the same graphically as below:

1.06% People out of 232 million internet users from NA Region visited widgets.com

Posted on July 27th, 2007 by admin
Filed under: Web Analytics | No Comments »

You may be wondering what happens if the web pages are dead. The excitement is on for Web 2.0 user experiences. Web 2.0 experiences are not limited to blogs, wikis, gadgets and other user generated content sites but also the regular portals and other shopping sites, and we are seeing a tremendous growth in marketers turning to web2.0 interactions.

Now, the greater deal for the marketers is to measure the web2.0 interactions. As most of these web 2.0 user experiences don’t necessarily load a new web page in the browser. Or the interaction, it used to happen before by going to a new page, now it can happen without leaving the current page you are on.

The current way of doing cookie-based page tracking may not be adequate to measure the complete activity on the web site. For example, if one can complete the entire shopping experience without leaving the store home page, there is no scope for measuring page level activities. We will only end up having one page view in this case. And, since many other metrics like stickiness, abandonment’s, entries, etc., are calculated out of “Page Views” metric, they all can not be calculated for the web 2.0 web sites. Here, the metric “Page Views” makes no use.

Page Views may be replaced with the new metric “Interactions”

With the web2.0 experiences, the metric page views sooner or later will vanish from the industry and I guess, the new metric “Interactions” will takes it place.

Webmasters and marketers are forced to measure the each interaction on the web page. In the same store example above, if we can tag each element on the store home page, we will be able to measure the complete activity like no. of interactions, average time spent on each element of the page, abandonment rate per element so on and so forth. We can convert all our regular metrics to web2.0 metrics which were previously calculated out of Page Views.

What’s the deal on “Engagement Metrics”?

Engagement metrics are typically mean “comments”,”subscriptions”,”wiki-updates”,”uploads”,”no. of times an image or video is shared” etc.,

These metrics have been in use for quite a number of years.
For example, if we go to any vbulletin forum, we can find no. of posts, no. of replies, no. of registered users reported on the forum site. And, another similar example is Yahoo Groups.

Engagement metrics are getting higher importance these days because the metric “Page View” and related metrics are disappearing in the web2.0 world.

The right metric for Web2.0 would be to measure “Interactions” in place of “Page Views”. Eventually, all other regular metrics can be calculated out of this new metric.

Posted on July 13th, 2007 by admin
Filed under: Web Analytics | No Comments »