Are Ads Less Effective on Cluttered Sites?

A billboard on a highway in the middle of the desert is more likely to be noticed than the same ad in the middle of Times Square. It’s only logical that it’s harder for an ad to get its message across in a highly cluttered environment.

Does that principle hold up on the Web? Is there a relationship between site clutter and ad (or branding) effectiveness? That’s the question that research, presented to the Advertising Research Foundation (co-authored by yours truly), sought to answer.

Clutter is an important issue for media planners. If site design is a factor in ad effectiveness, then the issue of site quality, not just demographics, should be considered when deciding where to run advertising.

It’s important for publishers, too. Sites like Lycos (which I wrote about last week) have been trying to improve their sites as vehicles for their advertisers. To that end, should they consider decreasing clutter on their site?

The Study

The first step in our study was deciding how to define site clutter. In traditional media research, clutter is usually defined by the number of ads in a given space or time period. Research has shown that too many ads in a row on TV, for example, can dampen their effect.

Unlike television, where the ad environment for a 30 second spot is uniform, Web sites vary tremendously. Some sites that accept ads are clean and sparse, while others are as crowded as, well, Time Square. And their degree of visual clutter doesn’t seem to directly relate to the number of ads on a page; some sites, with only a few ads, seem a lot more crowded than other sites with more.

For that reason, the partners in the study (AdRelevance, Dynamic Logic, iVillage and OglivyOne) decided to define site clutter by the number of elements (words, graphics, interest areas) on a Web page.

For the test, we took a Web page from iVillage (whose uncluttered environment has proven better than normative averages over dozens of branding tests) and altered it, creating a high, medium and low clutter version. Then we exposed respondents to ads on the three different versions of the site, and followed up by asking a number of branding questions relating to the product advertised. We compared each one of these research cells to a control group. We also asked respondents to rate how cluttered they thought the sites were that they saw.

Results

The hypothesis was ad effectiveness would suffer on the high clutter site we created. The results proved it wrong. There weren’t significant differences in branding metrics (awareness, message association, favorability and purchase intent) among the high, medium and low clutter pages we created.

The study uncovered something else. Ads worked significantly better on pages that respondents perceived, or rated, as uncluttered. In other words, among respondents who rated the site the saw as cluttered (regardless of the way we defined it), ads didn’t work as well.

Clutter is in the eye of the beholder. It seems that while our definition of clutter didn’t hold up, it is important to make sure users felt sites weren t cluttered. It s up to the next research project to decide what exactly makes people perceive a site as cluttered.

Actually, it’s up to each and every site to determine their own users feelings about clutter, since every site is different. Visitors to gaming sites, for example, probably define clutter differently than those focused on finance.

All of which proves the point I made last week. Happier users make for more effective advertising, and thus, happier advertisers.

Happy holidays!