Trends Report – Internet Arbitrage: A Short History

The top five companies in our area account for north of $500 million in annual revenues. Almost as remarkable, these companies make the vast majority of their revenues from advertisers paying them only for the actions they generate. Hundreds of millions of dollars in action based revenues requires incredible distribution. While many of the top companies have a strong affiliate arm to help with this traffic, almost all at some point realized the need to actively purchase media in order to achieve the scale they wanted; and the scale many have now attained. Given that these companies receive payment only for the actions they generate, means that they had to purchase media on a CPM or CPC basis even though they received payment on a CPA basis.

Those familiar with the space refer to this activity as Internet arbitrage, borrowing a term from a long established activity in the financial markets. Arbitrage may seem new to the web, but in reality it predates many working in the space today. It only seems new because it has finally gained acceptance among other advertisers, publishers, and the capital markets. To understand the origins of internet arbitrage, though, we must look back in time to the origins of one of the most popular and now infamous ad units, the pop-under. For assistance, Trends Report sought out Eric Roach, currently Director of Marketing for Oemtec and one of the driving forces behind the pop-unders’ acceptance as well as an early pioneer in Internet arbitrage.

When the pop-under first came to market, only the pop-up had some level of adoption. The early pop-ups were 300×250 in size and as the name implies launched (ER: popped) upon the loading of a web page. While it had rather high click through rates, the pop-up generated a higher level of negative user sentiment, and many top tier sites did not implement them. The pop-under did two things differently than the pop-up. First, it did not, for the most part, distract users from the site they visited, loading behind the active window and keeping the content window the active window. Second, rather than the diminutive pop-up, the pop-under mimicked the shape of roadside billboards with its rectangular 720×300 sizing. Besides familiarity, this size gave advertisers greater real estate with which to work, and its perception as being beneath the window gave top tier publishers such as Yahoo and FoxSports the comfort to embrace it.

The adoption of the pop-under by publishers came about due to a confluence of events in the Internet advertising world. By the second quarter of 2001, the Internet bubble had started to burst. By that time, though, publishers saw the potential of advertising revenues and had in many respects become reliant upon it. This need for internet advertising revenues, along with a shortage of those spending, set the framework by which a new ad format could be rapidly adopted. The pop-under had the additional benefit of not requiring any modifications to the content of the page. Unlike other display ads such as leaderboards, skyscrapers, and medium rectangles web sites did not have to redesign the look and feel to accommodate this unit. That said, because the ad existed didn’t insure its adoption.

The pop-under started life with the classic chicken and the egg problem. Websites did not want to include it unless they received guaranteed payment, and advertisers did not want to pay for the placement until they saw traffic. Enter the pay for performance and the arbitragers. Among the beauties of pay for performance advertisers is their willingness to try new units. While they care about their brand, it takes a second seat to revenues. Most people today think of Netflix and Orbitz when they think of the pop-under, but it was the early direct response advertisers X10, Lowermybills, and Reliaquote that paved the way for both the pop-under and arbitraging. Armed with these first generation advertisers along with good timing and tenacity, companies such as Advertisement-Banners, for whom Eric Roach convinced websites such as Yahoo to use this new ad unit. Not long after, others such as Traffic Market Place followed suit by aggregating media for its advertiser base.

The lifecycle of the pop mirrors that of many other ad units, such as banners, email, and desktop. The initial lack of dilution led to incredible returns. Aside from simple adserving, no form of optimization was necessary to produce the 2x to 5x yields. However, the transparency of the business and low barriers to entry allowed for quick competition for other risk takers. While prices didn’t change, effectiveness did, forcing companies to look for margin in either new inventory, new advertisers, or through the use of technology to better optimize the ads. Fastclick, for instance, in seeing the success of the pop-under used technology to help automate two pieces of the pop-under puzzle. They provided intuitive and powerful self-serve interfaces for publishers and advertisers. They timed their entry at a point where most small to medium websites had seen the pop-under and wanted to add it to their site but didn’t know where to go or where to get the sales resources to do so. Others like Advertising.com focused more heavily on the optimization so that they could acquire traffic through purchase rather than organically through publisher signups.

Today, the pop-under still runs strong, generating perhaps upwards of a billion dollars annually. Although few might have expected its association with infamy, as it is the ad-unit of choice for desktop software. Unlike the pop, Internet arbitrage shows no sign of slowing down. It does not rely on a particular ad-unit or distribution channel. It simply looks for opportunity. The same lack of barriers to entry and higher than average transparency still exist, which makes our space especially competitive. Each company looks for edges in advertisers, inventory, and technology, looking to build walls whenever possible. Several companies for instance have had phenomenal success finding niche sources of traffic, be it keywords or specific media buys for their products. Ultimately, success comes down to one’s ability to identify potential pricing inequalities and execute on them. That has fueled the growth of some of the largest players in our space and will likely continue to do so in the future.