Website and Search Engine User Behavior Analysis

Website And Search Engine User Behavior Analysis

By: Cristian Mezei
2006-11-01

How do users interact with the list of ranked results of WWW search engines? Do they read the abstracts sequentially from top to bottom, or do they skip links?

Tracking Eyeballs to Determine Search Efficiency

How many of the results do users evaluate before clicking on a link or reformulating the search? The answer to all these questions was found by a team of Cornell Universitys Human-Computer Interaction and Computer Science Groups on an Eye-Tracking research paper.

Below, you will find a heatmap of Googles search results pages most clicked and wanted sections:

What these guys did with the research, is to seek to obtain a more comprehensive understanding of what the searcher is doing and reading before actually selecting an online document. And they managed that using data that consists of 397 queries from 26 searchers (participants). They analyzed all behavior before a user clicks on the first link, or exits the page otherwise.

To better understand the above graph, Ill explain:

If a user clicked on the 5th result in a search engine results page, then he already scanned (but didnt thought it would help him, so he didnt clicked) 2-3 results (SERP positions 2-5) above the 5th result and close to none or rarely one below.

If a user clicked on the 7th result in a search engine results page, then he already scanned (but didnt thought it would help him, so he didnt clicked) 6 results (SERP positions 1-6) above the 7th result and 3 results below the 7th result.

The numbers in the graph above have decimal separators, because only 26 participants and 397 queries were tested, so an average number was calculated. You can round the numbers yourself, aproximating a 3,4 position to either a 3 or a 4.

One cannot consider this study as a highly relevant research paper, mainly because they only tested 26 participants, altough its a start and it shows us some basic usage and patterns of searchers.

Here is the original Cornell paper via SeoResearcher.

Enquiro also published a research paper on Eye-Tracking and their data is:

Organic Listings:

Rank 1 - 100%

Rank 2 - 100%

Rank 3 - 100%

Rank 4 - 85%

Rank 5 - 60%

Rank 6 - 50%

Rank 7 - 50%

Rank 8 - 30%

Rank 9 - 30%

Rank 10 - 20%

Paid Listings (PPC - Sponsored Ads):

Rank 1 - 50%

Rank 2 - 40%

Rank 3 - 30%

Rank 4 - 20%

Rank 5 - 10%

Rank 6 - 10%

Rank 7 - 10%

Rank 8 - 10%

So if your Adwords ad resides in the 5th place, you get 10 times less click exposure than the one residing in the 1st place on the sponsored ads section (Googles right sponsored ads sidebar).

There are a few companies that offer eye-tracking services and analytics:

1. Eye Tracking
2. EyeTrack III
3. Etre
4. Applied Science Laboratories
5. Arrington Research
6. Eye Tools

Jakob Nielsen, Chris Sherman from Searchenginewatch, Seth Godin and Dan Farber from ZDnet also have a few words on the Eye-Tracking technology, which is best used as an extra usability service to websites.

Dont forget about MarketingSherpas Search Marketing Benchmark Guide 2007: 185 charts & 18 eyetracking maps (no aff.) bookwhich gives you:

Eyetracking lab study results: how consumers view your search ads

3,944 real-life marketers search budgets & results

PPC paid search advertising data

SEO - search engine optimization stats

Alternative/complementary solutions to eye-tracking could be heat map tracking solution like CrazyEggs.

Comments

View All Articles by Cristian Mezei

About the Author:
Cristian Mezei is the Founder of Infomarket Network, a web marketing agency based in Bucharest, specializing in search marketing, social media and usability/accesibility. Cristian shares some of his views in Seopedia, his personal blog.