adwords

http://blog.washingtonpost.com/securityfix/2007/04/virus_writers_taint_google_ad.html

The above article describes a method used by some bad elements of society to install malware on vulnerable pcs as users click on adwords ads. Replies to the thread tend to point out the vulnerabilities of Internet Explorer Browser as the main culprit in the issue. In effect, the ads would point to a virus which would attempt to install on the users pc & then quickly redirect to the advertised site. Targeted keywords for the ads included well known institutions such as the BBB.

Today Google announced a Pay Per Action beta test destined for their adsense contextual advertising product. In essence, PPA does not incur a cost to the advertiser unless the visitor takes an "action". This could include a purchase or completing a form.

Needless to say, this has lots of Search Engine Marketers, Publishers, & Affiliates buzzing about what it means to the industries.

Nothing... yet.

However, it could become significant in a number of ways.

Some see an opportunity for smaller advertisers to begin advertising online as they would not have the risk of click fraud. Advertisers could determine what their profit per visitor is & then bid a percentage of that profit for a PPA advertisement. It's measurable.

Assuming PPA grows beyond the content network into the search network, the advantage over PPC still eludes me, though. Search engine marketers do this all day, every day with CPC (cost per click). The really scientific ones could tell you conversion rates & revenue per visitor by keyword & time of day. Most of the time this can be done without surrendering conversion rates to a third party like Google.

PPA requires conversion tracking in order to count an "action" (and charge the advertiser). Congratulations. Now another company has almost as much information about your online business as you do... and the resources better analyze the data.

While the PPA Advertising option is in beta & limited to the contextual network, I wouldn't bet on it staying there. If there is any indication that Google can increase profitability by expanding this, it will be scaled. Increase profitability... that's a clue.

Companies spend money to make money. The enhancements that Google has offered with Conversion Tracking, Web Analytics, etc. are given away for FREE (!) despite their development & maintenance costs. The true cost to advertisers is data. Nothing is free.

Some say this is the end for the Affiliate networks & aggregators or even the end of CPC - probably not... but truly it's too early to know if or how this will be adopted.

In the interim, I'm staying on the sidelines & building Content.

Google Adwords just released a Beta program for CPC ads on Site Targeted campaigns.

We just launched a test ad on a site with 500k+ impressions per day to try it out under fire. Look forward to a full review when we have some actual data.

Adwords has been available for site targeted campaigns for some time (...since 2003?). However, the site targeted model was based on CPM impressions as opposed to cost per click.

Site targeting can be effective if:

A) You have a product everyone needs

or

B) The site is a good fit for your niche or demographic.

Let's see how it goes ;)

----------------------

Ok here's the update. Wasted money.

It took a while for the ad to go live & we targeted a product that every adult driver needs & is mandated by law (clue.)

When the ads did start running - the budget was used up quickly. After about 100 visitors no conversions. In fact, none even ventured past the landing page.

Now I remember why we don't use the content network.

Your mileage may vary.

Eric

Google recently introduced the Adwords Quality Score to improve the experience for Google's users... and perhaps their shareholders as well.

For those who are unfamiliar with the term "Adwords Quality Score", essentially it is an algorithm which is applied to ads and keywords on Adwords which measures the relevance of the landing site to the keywords chosen & the ad itself... Yes, it's the site & not just the landing page. A-B Testing different domain names with the same ad & basic landing page content has confirmed this well enough to convince me.

Shortly after the Quality Score was introduced, Google added a tool to allow advertisers to see their quality score measured as Poor, Good, or Great. It's nice to have a little transparency. But with only 3 ratings, the key word here is little.

Unsurprisingly, our (good) sites which have always had good ad placement & a proven history of higher than average click-through rates showed a quality score of "great" for most keywords. Some longtail keywords fell into the "good" category. Some of the thin affiliate sites will need some work.

To give an example of financial services sites, our keyword minimum bids are running from .01 to .10 cents. Anyone in this niche knows that Google will gladly set minimum bids as high as $5 to $10 in this category if G doesn't like the site for whatever reason.

Ok, so we survived the Adwords Quality Score rating. But many didn't. Their options include dropping they keyword, raising the quality of the page to match some mysterious criteria in the quality score algorithm, or raise their bids.

Now, Adwords positions (and cost per click) are determined by a few main factors... click through rate, maximum bid, & now quality score. For more competitive niches, the end result of implementing a Quality Score is that many advertisers who wish to continue advertising for a keyword with a poor quality score will have to raise their bids. As they raise their bids, it will affect the cost to maintain a position for even those with a high quality score because the cost per click is determined by both bid & click through rate relative to other ads for the keyword.

Consider what that means in a niche that has thousands of advertisers across a gazillion keyword phrases.

Higher bids in reaction to the Quality Score, will drive up the average cost per click for even those advertisers with a high quality score.

I should have loaded up on GOOG at the $86 IPO.