Cold Hard Cache: The Quantified Truth

One of the most perplexing issues about ad counting centers around caching. But wait! before you delete this article and go about your business, know that my objective today is not to sing the praises of any cache-counting methodology. Rather, I have a message: You need to know how fast caching is growing on the internet, and just exactly what that means to your business.

As an advertiser or publisher, you hate caching. It leads to underrepresented ad counts and a reduction in page traffic. It prevents your ability to update current content, and results in over-inflated click rates and underrepresented inventory, and prevents tracking of all unique visitors. In essence, caching creates this hidden black hole in the internet that you can’t control and have no clue how to get around.

No one can be exact, but NetRatings figures that 30 percent of content comes out of browser cache. According to an Ernst & Young audited MatchLogic cache study, an average 76 percent of ads are delivered out of browser and proxy cache, reaching as high as 674 percent for some publishers. And according to PCWeek, studies show that infrequently changing web content accounts for 70 percent of network traffic — much of which gets cached.

Regardless of who’s numbers you believe, the message is consistent: Caching is a huge problem for advertisers, and it’s not likely to go away anytime soon.

Look at it this way, if you were (or are) an internet service provider or large corporation, you likely have one thing on your mind: saving money. Secondly, you may want to speed up the users’ internet experience. A cost-effective way to accomplish both tasks is to install caches.

Really, put yourself in the shoes of an ISP. According to caching appliance vendors, ISPs and corporations can save anywhere from 10 to 80 percent in bandwidth reduction just by employing caches. Consider an ISP whose bandwidth costs are $1,000 a month. Buying a cache appliance for $6,000 may save the ISP 50 percent in bandwidth costs and the ISP would pay off the appliance well within a year. (Sooner, even, in places where bandwidth may cost $10,000 a month — such as overseas.) Factor in increased customer satisfaction from faster connections.

So do you think the internet will see more caching in the near future? Damn straight.

Brendan Hannigan, director of network strategies at Forrester Research, expects 90 percent of Fortune 1000 companies to have deployed caches by end of 1999. Of those surveyed in early 1998, 50 percent were already installing caches. Collaborative Research estimates that 80 percent of ISPs and 56 percent of enterprises plan to use caching in the next year. Collaborative also says that the internet caching market will be $4 billion by 2002.

Infolibria, maker of DynaCache, recommends 2.5 cache units per POP. Oh, that means put 2.5 cache systems at each dial in center. And considering that there are thousands of those across the country (and world) among ISPs and corporations, that’s well, lotsa caches!

With a little research on my part, I counted 16 different cache products ranging in price from free to $60,000. So price or selection won’t inhibit the use of caching. And these 16 cache-appliance vendors claim caching levels of 35 to 80 percent. So what that means, folks, is that your ad and content:

  • WILL be cached
  • WILL be hidden
  • WILL be hard to count
  • WILL be over-clicked
  • WILL lead to underrepresented inventory

Whew! Get the picture?

Caching isn’t a short-term phenomenon. It isn’t something at which we can keep hurling busting techniques. And it isn’t something we can ignore — we can’t keep living the high life of false click rates and undercounted inventory levels. The harsh reality is that there is this black hole called “cache.” It is real, and it is something we need to work with not against.

Fixes like cache busting are short-term and ultimately get us nowhere. There needs to be a long-term solution to a growing problem.

Related reading