ellipsis flag icon-blogicon-check icon-comments icon-email icon-error icon-facebook icon-follow-comment icon-googleicon-hamburger icon-imedia-blog icon-imediaicon-instagramicon-left-arrow icon-linked-in icon-linked icon-linkedin icon-multi-page-view icon-person icon-print icon-right-arrow icon-save icon-searchicon-share-arrow icon-single-page-view icon-tag icon-twitter icon-unfollow icon-upload icon-valid icon-video-play icon-views icon-website icon-youtubelogo-imedia-white logo-imedia logo-mediaWhite review-star thumbs_down thumbs_up

Beware: the search advertising sky is falling

Sandeep Krishnamurthy
Beware: the search advertising sky is falling Sandeep Krishnamurthy

Have you ever opened a ValPak direct mailer? Have you ever clicked on a text ad? I pose these questions to my students regularly. Typically, the answer to the first question is "yes" and the answer to the second question is "never." The results are even worse if I inquire about banner ads or rich media.

How can this be? Search advertising has risen to become a behemoth; yet why is it that so few customers seem to engage with the advertising? My students are, after all, more tech-savvy than the average Joe Schmoe (I teach 11 miles from Redmond, WA, and eight miles from the Google Kirkland office). How is it that despite extended use of search engines on a daily basis, so few savvy individuals have clicked on an ad? Why is it that everybody is in love with Google's stock price and yet so few people love the ads that it displays?

Could it be that the ads really don't work the way they are supposed to? Could it be that we are in the midst of a search advertising bubble driven by overconfidence in the abilities of overrated services and an abundance of small, undiscriminating advertisers eager for any bump in traffic?

There are really two things driving this: a superclicking class that is artificially boosting ad prices, and an overworked and under-appreciated blogging community whose free labor is providing the content where ads might be placed.

If you are an advertiser, you should take this seriously.

A recent study shows that there is a category of people they call Natural Born Clickers. The study, which premiered at the iMedia Brand Summit in February, shows that a segment of the online population accounts for six percent of the online populace, and yet it is directly responsible for 50 percent of all ad clicks. This is a stunning number. What do we know about these superclickers? They tend to be young (mostly between the ages of 25 and 44) and poor (income less than $40,000). They have their own distinct behavioral pattern -- they tend to spend dramatically more time online than regular humans, and they are disproportionately more likely to visit auctions, gambling and career services sites.

If this is indeed an unbiased study of the online population (and it might not be -- Starcom Media, which conducted the study may have its own agenda), the implications are paradigm-shifting.

For the longest time, Google, Yahoo, Ad Center and other online ad services have focused on the click rather than the attitude change. This is problematic in two ways. First, this tends to work well for small businesses, which do not really have strong brands and do not have an interest in branding. As a result, it is likely to work for Ma and Pa Inc. rather than Big Enterprise Inc. Second, we know that it is common for consumers to act on an advertising message well after seeing an ad. By only counting clicks, the online advertising universe underestimates brand impact -- something that large advertisers care about.

Online advertising services have never provided information about who is clicking the ad. Therefore, many small businesses today have the illusion that they are measuring all sorts of details about their ads, but they do not realize that they don't know the profile of those who clicked on an ad. For instance, the profile of superclickers in the Starcom study points directly to an intern in a back office who has been ordered to go click on ads. How is a small business to know that?

This brings up the old bugaboo -- click fraud. Even search engines such as Google have long argued that by mitigating the impact of click fraud by advanced detection and discounting suspicious clicks, it has never gone away. The reality is that the system in place right now does not provide a reliable way of tracking click fraud.

Recently, a blogger at ShapingYouth.org wrote to Target complaining about a racy image with the target logo in an awkward spot. Rather than interacting with the blogger, Target sent the blogger a note that said, "Unfortunately we are unable to respond to your inquiry because Target does not participate with nontraditional media outlets.''

Even the egalitarian-sounding Craigslist recently went after a blog titled Craigslistblog.org. The letter to the blogger said, "We need you to stop using the infringing domain CRAIGSLISTBLOG.ORG immediately and arrange for transfer of it to us asap -- using/selling/transferring infringing domains is illegal, and penalties up to $100,000 per domain can be applied."

Blogging is one of the big content engines driving the current price of text ads. And yet bloggers are among the most under-appreciated, utterly exploited and overworked content workers who rarely see the fruits of their labor. Bloggers simply have to deal with a lot. They put up with snippy PR folks who constantly see bloggers as illegitimate actors whose voices need to be quashed, and IP lawyers who would be glad to shut them down on the pretense of IP infringement. On top of that, services such as Google's AdSense do not pay bloggers their ad revenue money until it exceeds $100. Unless you are on the top 100 list of blogs, you are making very little money. The result might very well be that the current blogging movement is simply unsustainable.

The sky is falling.

Search-based advertising is simply not sustainable. Why? 

First, traffic to search engines has been pretty flat over the last year and will continue to be. See this chart from Compete.com that confirms this pretty clearly.

The trend is even more diagnostic if one looks at the number of page views per visit.

In its most recent 10-K statement, Google claims that its increased revenue is "primarily from an increase in the total number of paid clicks." If the number of monthly unique visitors and number of pages/visits have been pretty flat for a year, the total number of paid clicks must have come from greater clicks from existing visitors. 

Second, growth in blogs has slowed down and will continue this trend. See the growth in the number of unique monthly visitors to technorati.com, wordpress.com and digg.com below.

Even though wordpress.com has really grown cumulatively, the rate has slowed considerably since November 2007. This might be the beginning of a great slowdown that will drive down the number of people visiting blogs. 

This pessimistic summation is because I see a slowdown in the number of visits to search engines and growth in content (especially blogs). 

What does all this mean if you are an advertiser?

1. Higher rates. The math is simple. If the supply of content and visitors slow down, as competition among advertisers for keywords goes up, rates are likely to trend higher. This means more money for search engines and lower effectiveness for advertisers.

2. Poor targeting. Pay attention to the superclickers. They are the folks who are easy to target. But, they cannot be everybody's customer! So it is going to prove difficult to target people that you want to go after (e.g., women 18-45, working professionals). 

Pay heed! The sky is falling! 

Sandeep Krishnamurthy is associate director, graduate programs and associate professor of marketing and e-commerce at University of Washington, Bothell.


to leave comments.

Commenter: Aaron Goldman

2008, May 08

I just read this column and agree with the comments that Sandeep's "assessment" of the search landscape barely scratches the surface here and is riddled with poor data and even worse extrapolations of what that data means.

The key point to remember is that, even though search marketers pay for the click, they can track what happens on their website after the click. 90% of the search marketers out there wouldn't keep running search campaigns if people we're taking some sort of action post-click that benefited their business. (The other 10% just want the awareness that comes from showing up on the search results page.)

100% of the heaviest search marketing spenders track some sort of event on their website (online sale, store locator, etc.) and manage their search budget based on those metrics. If they aren't getting more in return than what they spend, they move their money elsewhere. The fact that Google and other search companies revenue keeps growing is a testament to the fact that something more than just clicks is happening here.

Shame on iMedia for giving a forum to an outsider's misguided ramblings and speculation.

Commenter: paisley x

2008, May 07

please quit trying to sound like an expert.

Commenter: Nick Kastner

2008, April 30

You make a lot of great points but it all comes back to the marketing mix. No advertiser should ever put their eggs in one basket when it comes to advertising strategy. There are a number of factors that play into the success and failure of any marketing campaign, including PPC advertising.

I work for a digital marketing firm, Red Clay Interacitve, and what we have seen is that PPC, although not always great for direct conversions from click throughs, has been extremely powerful for branding. One of our clients in particular has seen an increase in sales from the PPC campaigns we have ran, mostly from increased search for their domain. We believe a lot of modern searchers see these ads, don't find exactly what they are looking for in the organic search but recall the domain that they may have seen in the paid results. This particular clients sees 100%+ increases in searches for their domain everytime we run PPC ads for any given product category relative to their offer. I personally believe that Google will eventually move to a CPM system in conjunction with some form of cost per sale.

Commenter: Caleb Whitmore

2008, April 30


You've made some excellent points, however I have to disagree with the premise that Search is flawed as a form of media. What IS flawed is how many marketers, particularly some large agency buyers and simple small business owners, approach search media: that being an approach focused solely on the "click".

Getting beyond the click and into attitude is exactly right. What you seem to have overlooked the reality that measuring that "attitude" can be accomplished very easily. Tools like the new AdCenter Analytics and of course Google Analytics, not to mention other web analytics tools help smart marketers measure what happens after the click - things like bounce rate, pages per visit, time on site, and all the way to the bank, a.k.a. conversion or point of sale.

Reason would lead us to conclude that if a certain keyword and ad consistently brings visitors who click but then "bounce" (view only the landing page and then leave), then that keyword is attracting the wrong kind of person or the landing page isn't meeting the visitors expectations. You can make a decision from that point: change your ad, keyword choice, or landing page, and keep watching.

When you look beyond the click you'll find tremendous value in sponsored search as many other readers have attested.

I hope to see another article that investigates this with greater thoroughness.


Commenter: Troy W

2008, April 30

This post is provocative, but unfortunately the execution leaves a little to be desired.
I am astonished that any professor of marketing, anywhere, would ask students what they do and then use this as a basis to extrapolate user behavior. The weaknesses of focus groups have been evident for more than fifty years. They may be good for evincing feelings about something, but as a predictor of what people *do* they are useless. You need to watch users to learn that, as marketers like Paco Underhill know (and Google).
Where is the $5 billion in revenue coming from? Why do people still believe online ads work if they don't? Even a naive user can easily see whether or not his spend is making money--and if you don't believe Google, then look at your A/R department.
I work in the industry, which has its share of hucksters (and they're not at Google). This stuff works. It is also still in its infancy. 99% of the bosses in this country are the pointy-haired kind, and they still think you need to spend $2 million on some TV ad that won't do bupkis. The day online ad spend hits 10% of ad budget for every business in this country is *years* off, as any search marketer will tell you. We're where TV was in 1958. There's a huge amount of room to grow, and even more room for ads to get better as various other posters point out.

I can't believe people are still making this argument! I'm more and more convinced that it's a fundamental, profound ignorance of just how search ads work and how insurmountable Google's lead is. If this were space travel, nobody would pay a minute of attention to a Nepalese "assessment" of the US space program--they are a wonderful country and wonderful people, but they don't have the technological chops. And yet people swallow "search analysis reports" by folks like Compete.com, who if anything are less qualified to comment...

Commenter: Neil Squillante

2008, April 30

So true about bloggers being under-appreciated (and under-used I might add) by public relations professionals.

But I disagree with your suggestion that savvy people don't typically click on search ads. I work in the online advertising industry, but I click on search ads.

When? When I'm shopping. In fact, Google enables you to just see ads associated with your search, which I sometimes use since the ads are often more relevant than the search results. I would think industry insiders like you and I click more often because we tend to notice ads (I'm always grading the copy in ads).

I believe the untold story in your article is that search ads work best when used as a tool to catch people near the end of the sales cycle. Other advertising vehicles work best earlier in the sales cycle. Of course, exceptions exist. You can use Google to offer a white paper (early sales cycle offer), and you can use other vehicles to offer a discount (late sales cycle offer).

Commenter: Retah Elgoog

2008, April 30

I agree with most of the comments above. I do think that the title "The sky is falling" is a little extreme. I do agree that advertisers may begin to shift some of their Paid Search budgets to other online channels during the next few years, but not necessarily for the same reasons stated in this article.

I live and work in the heart of silicon valley, around the corner from Google and down the road from Yahoo head quarters. I agree, if you ask people in this tech savvy valley if they ever click on text ads or banner ads, they will say "never". But if you asked if they ever type in a brand name in Google and click on the brand's search result link to simply navigate to the brand's website, they will say "absolutely". Often, the links that they are clicking on to quickly navigate to the website happen to be sponsored links(paid search). In their quest to quickly get to a website, these tech-savvy users might not care or even realize that they are clicking on a sponsored link...they just want to get to the website. This user behavior is very common - instead of typing a website address into the URL bar, they simply use Google and search links to navigate to a website. Why do Paid Search campaigns get credited for so many conversions based on "last click wins" tracking technology? Much of their success can be attributed to this common "navigational" behavior by online users. Search sponsored ads might not be truly driving conversions through their targeting and messaging, but they are definitely a popular vehicle being used to navigate to ready-to-convert users to a website. I can definitely guarantee that outside of these tech savvy cities and valleys, the average joe is clicking on sponsored links intentionally or not intentionally.

What's going to truly impact Paid Search:
I also agree that online advertising is continuing to change and new technologies are allowing advertisers to gain better and deeper insights on their online marketing spend. Some 3rd party ad serving providers are already looking beyond the last click metric that often gives 100% conversion credit to Search. This will allow advertisers to analyze how ALL of their different media buys and channels are interacting with the consumer while they browse, research, and move around the web prior to converting. Soon, the last click will not get 100% credit for the conversion. This is what will have the biggest impact on Search. Consumers will continue to use Search to navigate to websites and will continue to be the last click event prior to a conversion....but will no longer receive 100% credit. It might only get credit for 50% of the conversion. Clicks and Views from display campaign banners or emails or even organic search that are occurring prior to the last click will finally get recognized and awarded partial credit.

Commenter: Adam Russell

2008, April 30

P.S. Excuse my inconsistent numbering in the above post...


Commenter: Adam Russell

2008, April 30

I can't help myself - to say this article seems poorly researched would be a compliment and to make matters worse there seems to be a serious misunderstanding in several key points:

1. What does the Superclicker segment sound like? To me they sound pretty much like the future - saying that most clicks come from young individuals with below average income that purchase online a lot is like putting down the viewers of Nickelodeon because they are, on average, under twenty and earn nothing. The audience you describe mirrors what you could expect, to a certain extent, for online purchasers as a whole.

2. The Craiglistblog example seems an odd one to chose - I'm not sure what this blog covered but I think Craiglist's response is understandable because the domain suggests the blog is official. A far better example would be Apple's response to Think Secret.

4. It has been widely noted that Google's latest growth comes from overseas markets, not the US. I think this means that the Compete figures for Google.com are probably irrelevant, no?

3. Search isn't purely based on the content network. The content isn't purely based on blogs. Search includes sponsored search links (unsurprisngly) and blog traffic is actually a very minor part.

4. The fact that, as you point out, small blogs aren't getting paid reflects the fact that small blogs aren't generating clicks. Even if they did all drop off the radar (they won't) the clicks lost as a result would be a small proportion

5. Bloggers don't all do it for money - in fact, most don't. Those that never hit the $100 minimum payment level (this includes me with my blog) don't blog to get paid - they blog to express themselves and to give an opinion.

Only in digital would people worry about a decrease in the rate of increase so much - the number of blogs, searches or clicks cannot increase forever.

Commenter: linda smith

2008, April 30

I agree w/ many of Matt W's comments. There's a low barrier to entry into PPC, and thus many who are non-marketers are 'experimenting' and creating bad ads that don't perform, and annoy those who do click, as they don't deliver on their promise.

I wouldn't say the sky is falling, but I do think there's a bubble that will burst, and the inexperienced will realize that they're throwing their money away and they can either hire specialists...which the web space is requiring more and more of, or they will get out of the PPC game.

As for blogs, I agree, that too will level out whereby the blogs that deliver value will remain, and those that thought they could have 3 seconds of fame, and didn't achieve, will move on to other things.

No one knows what the future holds. That's the excitement of the web. It's a roller coaster ride. I just hope the arrogance of some who think they know all, subsides, because after working in this space for 10 years, it gets old.

Commenter: Darren Klein

2008, April 30

If the author's small sample of people who indicated "never" actually was never, then Google wouldn't be making billions from people who are clicking. Superclickers aside, the click rate might be low, but the volume of overall clicks makes up for it.
I think "the sky is falling" is a little extreme.

Commenter: Julie McHenery

2008, April 30

I completely agree with Matt's well put comments. PPC campaigns are pretty simply to create but not so simple to manage properly, therefore there are many inexperienced people out there setting up poor, untargeted ads at the moment.

As the PPC medium matures, and online marketers get better at understanding web analytics, the 'bad ads' will naturally reduce. PPC is one of the most powerful marketing tools we have, we just need to get better at using it to its full advantage.

As far as I'm aware, the 6% stat does not take into account B2B accounts - For example, I would suggest that there is a substantially greater proportion of people looking for enterprise software who will click on an ad than the 6% suggests.

Commenter: Bill Burke

2008, April 30

Sorry for the duplicate post- Sandeep, just curious.

Were you aware Compete.com still hasn't figured out how to measure web traffic (nor do they appear motivated) to/from domains that have sub-domains?

Nor can they track traffic to/from domains that host exclusively on a "www.VXYZ.com" CNAME?

Commenter: Bill Burke

2008, April 30


If I may be so bold as to give unsolicited advice, Compete.com's statistics are questionable, at best which is why very few major organizations pay any real attention to them.

Secondly, isn't it also possible that Google (why bother discussing the other 2 SE's - they share less than 30% of the search marketplace) has improved their search results to where Googlers are finding what they want within the 1st set of results, and need search no further? Wouldn't that account for the flattening of search traffic numbers and the accordant page views?

As to growth in blogs slowing, could you share what you base that opinion on?

Re: click fraud, there is little to be done there; web crawlers and other miscellaneous factors are too many to be contended with.

As to PPC, I think the market is adjusting, not collapsing. We're seeing growth in ads for commodities, and a slowdown in PPC for services; although that's a general, unprofessional observation

Bill Burke

Commenter: Matt Wasserman

2008, April 30

The sky is not falling. There will be an adjustment, maybe sooner, maybe later, but PPC remains one of the most effective and efficient means of driving quality traffic.

The article assumes that all ads are created equal, equally relevant and equally targeted. But the fact is that most ads are simply bad and the situation is, at this point, getting worse instead of better. The DIY nature of paid search at the bottom end of the market and the desperation to harvest clicks at the top of the market results in a huge percentage of ads that are not relevant, not targeted, and do nothing to qualify vistitors. This is, to my mind, a much more plausible explanation for the relative flatness in pageviews growth. Bad ads rarely generate more than one pageview per click.

Keyword algorithms like Google's broad matching and Yahoo's advanced matching exacerbate the problem. They are designed to generate revenue for the publisher, not the advertiser, but they do generate a lot of clicks.

As the mom and pops become better at managing their ad spend, and the larger advertisers move away from their obsession with reach and towards more engagement oriented metrics as a measure of success for PPC, things will change.

As for the flatness in blog growth - blogs were a fad and there was never much money to be made from them for the majority of bloggers. Not to say that they are going to go away entirely, or that they are not an incredibly valuable source of content, but after the initial craze most people started wandering off to do other things. The market will evolve around commited bloggers who actually have something to say, and the hobbyists will fall off.

When the mom and pops that can't figure it out(unfortunately) spend themselves out of existence and the big spenders introduce some sanity into their strategies the market will start to grow again, this time from a more rational and sustainable base.

Commenter: Steve Pohlit

2008, April 30

Hello and thank you for the post and data. Regarding the data on blogs I am wondering if it captures the blog hosted on server supporting a domain name. For example http://www.stevereports.com is a Wordpress blog platform hosted on space I lease. I set it up without going through the Wordpress site

The substantive issue is on paid search. I just spent a year with an eCommerce business model where paid search continues to be an important part of the marketing strategy. Every campaign was carefully constructed and revenue vs. cost was continually monitored. To give you an idea of volume the company invests about $15,000 a month in paid search.

The results continue to exceed forecast and as long as there is profit after cost of the ad and cost of the product sold the ads will continue. As this is written the performance of paid search is very lucrative for the business.

Marketing strategy is a meaningful component of The Profit System I developed to help companies improve revenue and profits. I continue to recommend and assist clients with the implementation of paid click advertising.

Steve Pohlit, Business Development Consultant