Huckabuck adds scheduling and alerts support

Fri, 16 Feb 2007 16:34:00 GMT

The Huckabuck reporting service has been updated now allowing users to schedule reports to be automatically run at daily, weekly, bi-weekly or monthly intervals. In addition, users may opt to receive email notification when certain events are fired. For example, you may wish to have your Pageshare report run on a daily basis, being notified by email when your pageshare or rank changes.

Create your free Huckabuck account today.

What's your pageshare?

Fri, 26 Jan 2007 22:11:40 GMT

If your business relies at all on Internet sales then you have more than likely at one time or another performed a web search to see what position your site holds in the major search engines. You may have even performed some SEO (search engine optimization) tricks to try and make sure your site comes up in the top 10, or even better, the coveted top 3 results for search terms related to your product or service. However, even if you've made it to the top of the mountain, the reality is that several million new pages are indexed daily and the competition is not far behind.

In this article I intend to explain why a Pageshare report is needed and also the algorithm behind the report. Necessity is the mother of invention, so let's start with the need.

As the founder of a small software company whose sales are 100% online I found myself constantly performing web searches to see where I and my competitors ranked for various search terms. I did this on a regular basis using the major search engines Google, Yahoo and MSN. Based on these results I would create and execute a marketing plan to maintain and/or improve my search ranking. Unfortunately manually analyzing these search results on a routine basis would take up a lot of my time, time away from focusing on critical software and business development tasks. Like many small business owners I couldn't justify the expense and effort of hiring/training a person to perform these tasks but at the same time I could not afford to ignore these duties myself. As a result I decided to focus a majority of my time on the core areas of the business, checking search ranking when I had spare time.

As expected spare time was rarely to be found and occasionally I would be less than pleasantly surprised to see a competitor jump in front of me in the search results. When this happened I would scramble to try and regain top-position before losing potential marketshare. Once regaining position I would then refocus my efforts on the core areas of the business ... that is until it happened again and the cycle repeated itself.

After repeating this process a few times and telling myself I needed to be more proactive v.s. reactive to these happenings, I began to long for a report that could quickly tell me my web sites position across all the major search engines. Additionally, I wanted something that would tell me my position in relation to my competitors and that could quickly convey whether my position was improving. After further consideration I decided that what I needed was a report that indicated marketshare for a given search term across the major search engines, hereafter referred to as pageshare.

After a long search I could not find such a report. I found some basic reports that would provide me with my ranking for a single search engine, but nothing that did everything I wanted. Therefore I decided to create my own, hence the birth of Huckabuck and the Pageshare report.

We (a group of talented programmers from across the U.S., Costa Rica and Ukraine) started with the idea that a website's pageshare for a given search term is directly related to it's position within search results for said search term. Websites with results closer to the top are obviously more likely to be seen/clicked by users so will have a higher pageshare than those listed farther down in the results. Additionally, since a website may be listed more than once in search results this also had to be considered when calculating pageshare. For example, if your website occupies both positions 1 and 3 for a search term this is obviously better than if you just occupied position 1.

To start we decided to focus on the major search engines of Google, Yahoo and MSN. Using the criteria previously mentioned we set out to create the first version of the Pageshare algorithm which is described below.

Pageshare algorithm version 0.1

1.Perform search against each search engine and store results.
2.Group results by domain.
3.Assign points to each search result based on position. See Point Assessment.
4.Calculate the total number of points for each domain and divide by the total number of available points to determine pageshare percentage.

Point Assessment

Results 1-3 = 10 points
Results 4-6 = 6 points
Results 7-10 = 4 points
Results 11-15 = 2 points
Results 16-20 = 1 point
Results 21+ = 0.5 point

After running several reports using this algorithm I found it to be very useful in quickly identifying my pageshare and how I compared to my competitors pageshare. We soon realized however that the algorithm was flawed. The problem was that not all search engines are created equal. Google obviously has a much larger marketshare in web search than Yahoo or MSN, therefore one cannot consider a #1 ranking on MSN to be equal to a #1 ranking on Google. Furthermore at this point we had added del.icio.us search results to the mix and they command an even smaller share of web search than Google, Yahoo or MSN, so to assign del.icio.us search results the same weight as a Google search result would make little sense. To solve this issue we needed to determine what marketshare each search engine had and factor this into the total pageshare calculation. Fortunately someone had already done the work in this regard.

comScore is a company that houses vasts amounts of information on consumer behavior. One of the metrics they report on is called qSearch, which tracks the total number of searches performed by Americans within a given month and what search engines were used. According to December 2006 results and over 6.7 billion searches, Google has a marketshare of 47.3% followed by Yahoo and MSN with 28.5% and 10.5% respectively. This means that roughly 4.5 times as many users performed searches using Google v.s. MSN. No information was provided for del.icio.us however as a rough estimate I assumed a 1.0% marketshare of web searches. Using these values I could then more accurately calculate pageshare using the following algorithm.

Pageshare algorithm version 0.2

1.Perform search against each search engine and store results.
2.Group results by domain.
3.Assign points to each search result based on position.
4.Multiply the points for each search result by a factor equal to the marketshare for that search engine.
5.Calculate the total number of points for each domain and divide by the total number of available points to determine pageshare percentage.

Using these results I found that I often received a completely different, although I believe more accurate view of my pageshare and competition when compared to results using the first version of the pageshare algorithm.

Today the Pageshare Report is made available to all via the Huckabuck service. Please feel free to try it out and let us know your feedback. We would be especially interested in knowing how we might improve this report or service in general.

Huckabuck - search tag

Huckabuck adds scheduling and alerts support

What's your pageshare?