Robinson,H . "Internet advertising effectiveness: the effect of design for click-through rates for banner ads" International journal of advertising [0265-0487] 26.4 (2007). 527-541.

    This empirical study investigates characteristics of banner ads and the role they play on online advertising effectiveness in the world of online gambling. The relationships between the click-through-rate (CTR) and seven design elements (banner size, message length, promotional incentive, animation, action phrase, company brand/logo and casino games) are examined, and although the author states up front that that the intention of the study is to see how these relationships differ between the online gambling industry and other e-commerce, this information can still be rendered useful. The study found design elements of effective banner advertisements include: “a larger size, long message, absence of promotional incentives and the presence of information about casino games.” The company/brand logo was found to be an “ineffective characteristic” in attracting a direct response.
    While it would not be accurate to generalize this data to all banner advertisements, the authors do say, “these results also suggest that the effectiveness of banner ad campaigns for online gaming (as measured by CTR) is governed by the same principles of artistic execution as those found in other, non-gambling-related sectors of the internet.” The inability of click-through rates to assess other effects such as emotion (which an ad evokes) or memory (of a particular brand on an ad) seems to be magnified in this study, for it is heavily based around the already faulty click through rates. Specifically, the finding that the brand logo is an ineffective characteristic in attracting a direct response seemed troubling at first because brands and logos play a major role in eliciting credibility. However, credibility can still be present, even if it is not evidenced by a click through (possibly a limitation of this study)

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.