What makes web ads work?

July 14, 2006 9:11 AM

What data is there, academic or otherwise, on the effectiveness of various types of web advertisements?

I've clicked maybe 20 banner ads in my life, 10 text ads, and zero pop-ups. I am a terrible data point and I don't understand what generates interest in web advertisements at all.* Is there a breakdown of this somewhere? I'm sure that a lot of marketing and media corporations guard this information like gold-pressed latinum, but surely there's an academic somewhere who has looked at the issue.

* For instance, PUNCH BRAD PITT TO GET A FREE IPOD! Who clicks this? Why?

posted by Optimus Chyme to Work & Money (5 answers total)

I don't have the data, but subjectively:
The typical web ad gets a click through rate well below 1 percent. I've seen 0.1 percent, 0.2 percent, things in that range. That, and the personal low usage rate you mention, tends to suggest poor results. However, look at it this way: Americans are exposed to an average of 3000 advertising impressions per day in all media. How many do we typically act on? Certainly far less than 1 percent, most likely less than 0.1%. So a 0.1% clickthrough rate (1 per 1000) may actually be very good. In terms of advertising effectiveness, once a prospect clicks through, that's almost as good as having them physically walk into the store. The advertiser now has the prospect on their own web site, and now it's up to them to sell the product. That, along with the ability to measure and track results and to closely match advertising with interests and demographics, helps explain the continuing interest and growth in online advertising.
posted by beagle at 9:25 AM on July 14, 2006

Response by poster: I'm surprised this didn't get more responses. Thanks for the info, beagle.
posted by Optimus Chyme at 1:13 PM on July 14, 2006

Not sure if you have access to any databases, but I just took a quick look at ABI/INFORM and there is an article from this year that addresses this issue:

The Effect of Banner Advertising on Internet Purchasing /
Puneet Manchanda, Jean-Pierre Dube, Khim Yong Goh, Pradeep K Chintagunta.
JMR, Journal of Marketing Research. Chicago: Feb 2006. Vol. 43, Iss. 1; p. 98

The abstract states:
In this research, the authors focus on a previously unexplored question: Does banner advertising affect purchasing patterns on the Internet? In particular, using a behavioral database that consists of customer purchases at a Web site along with individual advertising exposure, the authors measure the impact of banner advertising on current customers' probabilities of repurchase while accounting for duration dependence...
posted by Razzle Bathbone at 1:27 PM on July 14, 2006

I am a terrible data point and I don't understand what generates interest in web advertisements at all.

It's easier to understand when you realize that people who click on ads aren't usually big fans of the site they're visiting.

1. Search for a phrase, click through to a site.
2. This site doesn't answer my question, maybe this ad will.

(or)

2. This site is boring, maybe I'll punch Brad Pitt.
posted by mmoncur at 11:33 PM on July 14, 2006

I book a lot of banner ads (no not the "click on george get an ipod" kind.

I never expect a return rate over 2%.

I find the point is more about saying "Oh, I see Joshua Bell has a new album out, maybe i'll check that out at some point in the future since I like him" than "Oh theres a banner ad about the new Joshua Bell record, I'm gonna click it and be taken to a retailer i may or may not know!"
posted by softlord at 7:28 AM on July 16, 2006