Persuasive Advertising - J. Scott Armstrong [176]
Ogilvy claimed that four times as many people read captions as read the body copy, and that all pictures should have captions.
I found little evidence related to the effectiveness of captions. Their use relies primarily on common sense and received wisdom. Also, in Wheildon’s (1995, p. 122) study, 67 percent of his readers preferred pictures that had descriptive captions. They did not like having to search ads to determine what the pictures represented.
9.3. Text
By text, I refer to words other than those in headlines, captions, and brand identifiers. The following principles are especially applicable when strong arguments exist.
9.3.1. Make the first paragraph relevant, short, and easy to read
The first paragraph, and especially the first sentence, should be relevant to the key message. Once people start reading an ad, they are likely to continue, so the beginning of an informative ad should be short and easy. This is consistent with evidence from the Zeigarnik effect: once people start a task, many of them feel a need to finish it.
An analysis of eye-tracking studies for full-page Dutch magazine ads found that approximately three-quarters of viewers did not read the whole ad (Elpers, Wedel, and Pieters 2003). This is not necessarily a problem because a good ad should quickly indicate whether it is in the customer’s interest to keep reading.
An ad that shows the most important information first is more likely to retain the interest of the target market. In addition, even if consumers do not continue with the entire ad, they will have read the most important information. Consistent with this, website-design experts such as Nielsen (2000) advise against the use of a “splash page”—an attention-getting page with no information—because it delays customers from reading the most important information.
Ogilvy claimed that an opening paragraph with fewer than 11 words would increase readership of the full text. Our WAPB analysis found that of the 480 full-page print ads by leading U.S. firms, only 11 percent used an opening paragraph of 11 words or fewer. But 71 percent of them did put the main message in the first paragraph.
Evidence on the effects of a relevant and short first paragraph
Our analysis of quasi-experimental data on print ads supports using a relevant and short opening paragraph:
Print ads with relevant first paragraphs had much better recall. Our WAPB analysis found 46 pairs of ads in which one ad placed the main message near the beginning, while the other ad did not. For example, an ad that began with the claim, “the new Isuzu Rodeo comes with a powerful new 24-valve, V-6 engine” had a recall score of more than four times that for another Isuzu ad that did not state its main message in the beginning. Recall for ads with relevant openings was 1.74 times better than for the other ads.
Print ads with a short first paragraph had better recall. Our WAPB analysis found 36 pairs of print ads that had 11 words or fewer in their first paragraph, while the other ads had more. Recall for shorter opening paragraphs was 1.13 times greater than for the other ads.
We also analyzed non-experimental data in WAPB. In comparison with the industry norms for each ad, the persuasion score for the 151 ads with a relevant message in the first paragraph was 7 percent higher than the comparable score for the 25 ads without a relevant opening.
Given that people are exposed to hundreds of ads each day, should the text be short?
9.3.2. Provide long copy when reading time is not constrained
The more you tell, the more you sell.
Hopkins, 1923
Long text—say at least 50 words—is especially persuasive for high-involvement utilitarian products when the target market needs information. Thus, it is particularly relevant for new or improved products. For example, following Apple’s 1984 TV ad, its 20-page ad in national magazines was read by