Banner Ad Placement Study

Theory: Two ads on a page will generate a higher click-through rate

Method

Two ad locations were tested for this study. The ads were tested only at the top of the page and then tested with the same ad located at the top and bottom of the page. Each location was tested for 1 week. The goal was to run the studies on high traffic pages.

Each of the boxes shown below represents the web page (note: the pages were longer than the visible area of the monitor, therefore the ads were not seen at the same time). The smaller boxes represent the ad locations on the page. The name of the company whose ads were shown is given above the boxes. The URL of the page on Webreference is shown between the two rows of boxes. Each column then represents the page and company. Each row represents the ad location being shown on the web page during that week.

Results

Listed below are the results from the study. Hypothesis testing was used to test the statistical significance of the results. View hypothesis testing calculations.

The Shopsite study was not included in the results due to inaccuracies caused by logistics problems.

Note: The link to the HoloNet advertisements shows how the ads were run at the top and bottom of the page for the second week.

HTML and Netscape Page (view advertisements on page)

Conclusions

The study was inconclusive because the results are not statistically significant. (Note that the Holonet ad was not a new ad, and has been displayed on the site for 6 months.)

Back to Home

Comments are welcome