Digital Ad Archive

About this project

After being dissapointed by existing repositories of online advertisements,^[1] I made my own. Advertising can be a rich critical tool for historical analysis—their form and content are shaped by, and in turn shape the institutional, economic and social contexts in which they are placed. Some lines of analysis can include what subjects were represented (and went unrepresented) in these ads, how they were represented, and how these ads drew from prevailing semiotics and discourses.

This collection includes over 800 static and animated online ads spanning the emergence of the early commercial internet. Sourced from the Internet Archive's Wayback Machine, metadata for each ad is stored in a json file that allows for filtering by resolution, advertiser, source, subject and file format. A Python script automatically generates much of this metadata, while alternative text and subject metadata are inputted manually by me.^[2]

The first banner is widely reported to be a 1994 AT&T ad that ran on HotWired, the online counterpart to the print magazine Wired.^[3] Although, like many of these sorts of titles, the title of "first banner ad" is a disputed claim that probably hasn't been helped by citogenesis. Many of the archived ads come from key players in the dot-com financial bubble: start-ups and other dot-com companies demanding users for their web sites.

This is not and cannot be a comprehensive repository of web banner ads. For one thing, many ads during the time period were displayed using Macromedia Flash,^[4] which I have chosen not to archive. This archive does not display any ads for pornographic websites, but you can still access these ads through the json file linked above.^[5] More importantly, whether an ad is accessible in the first place is dependent upon whether it has been crawled by the Internet Archive. Content on the web can be altered or disappear without ever being crawled. Since ads can be served based on user data, the configuration of the Wayback Machine's crawler could be a source of systemic bias into what ads have been archived.

This archive is a pet project of mine. If you find any issues with or questions about the site, please reach out to me via email or leave me a message on my guestbook.

Further reading

If you need help accessing any of these books, feel free to contact me, I may be able to help!

• Ankerson, Megan Sapnar 2018 Dot-Com Design: The Rise of a Usable, Social, Commercial Web NYU Press 9781479882182
• Jessen, Iben Bredahl 2010 The Aesthetics of Web Advertising: Methodological Implications for the Study of Genre Development Brügger, Niels Web History New York Peter Lang 9781433104688

Footnotes

1. Existing historical ad repositories largely favor print and television ads, while the few existing online ad archives fall short of what I wanted. Existing archives include James Gardner's now defunct blog adverlicio.us and Tyler Grant's archive of Flash ads, neither of which are still active. The Web Design Museum's web banner exhibition is interesting, but is closer to illustrative than comprehensive. Huang and Zhao (2024) has the most systematic approach, but their automated search method excludes advertisements that do not match specific IAB size guidelines.
2. Manually writing alternative text is a slow process, but I'm not satisfied with using an automated method. I may try to crowdsource this work in the future.
3. You can read more about the ad on Wired's retrospective article, written 16 years later.
4. Later Adobe Flash.
5. These hidden ads are marked in the json file with the key "adult": true.