Lohtia,R . "Evaluating the efficiency of Internet banner advertisements" Journal of business research [0148-2963] 60.4 (2007). 365-370.

    In this article, the Data Envelopment Analysis (DEA) tool is introduced and is used to measure the efficiency of banner advertisements. On a technical level, DEA converts “multiple inputs and outputs into a single measure of performance, which is generally referred to as relative efficiency.” An input variable is advertisement related, such as incentives, emotional appeal, and specific graphic elements. The three output variables observed are click-through-rate (CTR), attitude towards the ad, and recall. The DEA produces “an efficiency frontier that represents the optimal levels of output for given levels of input.” The author stresses that DEA can be a “useful tool for pre-testing advertisements” as well as comparing individual advertisements. The authors found an inefficient advertisement in one banner that had “3 colors and a message length of 6 words. It had no interactivity, animation, incentive, or emotions.” Using the DEA tool, the advertisement could achieve efficiency with “ higher CTR scores, higher attitude scores, and higher recall scores.” Therefore, the advertiser could take this output information and manipulate the banner design as well as reconfigure the placement of the ad to improve these specific aspects. Inputs also have the ability to direct advertisers to evoke more of an emotional appeal or improve the graphics in hopes of elevating efficiency.
    Primarily, this study is able to show that banner advertisements are important. They are important enough to catch the attention of an analysis tool, and have someone adapt it to internet advertisements in hopes of increasing their efficiency. While specific characteristics do not seem to be involved (e.g. Change the color of the banner from brown to blue), the tool’s ability to focus on a certain element of an advertisement is unmatched.  In addition, the banner advertisements examined in this study lay the groundwork for identifying characteristics of effective and ineffective advertisements. One of the appeals of this tool is that it provides actual numerical information and data. Instead of creating banner advertisements from scratch with seemingly sophisticated graphics, advertisers are able to justify their decisions with a proven efficiency tool.