Long tail dead

Yup, facing extinction. (Image credit: searchengineguide)

The “long tail” in search engine marketing is dead. I delivered this sad news at ad:tech in San Francisco last week (scroll to the bottom of the post to see the slideshow), and I also introduced a new term – the “wide tail,” which has replaced the long tail. This post elaborates on both of these terms and the significance of this shift for search engine marketers.

First, let’s discuss what the long tail is. The long tail of SEM refers to low-volume but high-intent queries. For example, a phrase like “San Francisco bad credit refinance mortgage rates” probably doesn’t get many searches every year, but it has a high degree of consumer intent embedded in the query and is likely to convert very highly (after all, we know the geography, credit rating, and home ownership status of the searcher, so we can create a highly targeted landing page and ad text).

In the olden days (read: five years ago or longer), finding long-tail keywords was an effective way of creating profitable SEM campaigns, simply because there was a decent chance that you could find keywords that your competitors didn’t, and thus get clicks at a CPC below what you should be paying. In economic terms, this was an inefficient market, which means that a keyword phrase worth $2 if all advertisers were bidding on it might only go for $.25 because many advertisers never knew the keyword existed.

This sort of inefficiency was great for SEMs but terrible for Google because the true value of the keyword wasn’t being unlocked. So, over the last several years, Google has relentlessly changed its rules and algorithms to maximize return on all keywords. The result has been greater revenue for Google and fewer and fewer opportunities to profit from long-tail price inefficiency.

Image credit: maximumpc

Here are just a few of the ways that Google has destroyed the long-tail opportunity:

Google Instant Search: Studies suggest that Google Instant Search reduces query length by 17%, meaning that shorter queries (head terms) have gained volume over longer queries (long tail).

Elimination of misspellings: Buy the word “morgage” today and you will end up in the same keyword auction as advertisers who bought the word “mortgage.” This wasn’t the case in the past.

Better, proactive keyword tools: Google’s keyword tool – in addition to many other 3rd-party tools – has reduced the advantage that savvy SEMs once had at finding obscure keywords. And Google no longer just offers a passive keyword tool; it also includes an “Opportunities” tab in AdWords that proactively suggests new keywords to advertisers. As a result, the ability to find keywords no one else has discovered is reduced, and prices on these keywords have increased.

Elimination of geo-modify advantage: In 2005, I downloaded the Census Bureau’s list of 22,000 U.S. cities and appended these names to words like “mortgage rates” and “refinance rates”; instantly I had tens of thousands of long-tail keywords! Today, Google determines geo-intent by either the geo-modifier in the keyword or the IP address of the searcher. As a result, when someone types in “San Francisco mortgage rates,” advertisers who bought the term “mortgage rates,” geo-targeted their campaign to include California, and (by default) have elected to show ads to users by search intent and physical location, will have their bid entered into the auction. This effectively eliminates my trick of buying 22,000 geo-specific keywords.

Product Listing Ads:  Google has reduced the advantage of retailers buying thousands of product-specific keywords by introducing product listing ads (or PLAs). PLAs enable ecommerce companies to upload a feed of all of their products and then have their ads show up whenever Google determines that a query is related to a specific product. So rather than having to create one million ads for one million SKUs, one product feed can now cover millions of different queries.


Add to all of these innovations the overall increase in advertiser participation in AdWords (any advertiser worth his/her salt has an AdWords campaign today; that may not have been true five years ago), and the ability to truly make much money off the long tail is very low.

So what’s an AdWords advertiser to do? Well, at PPC Associates, we recommend a two-pronged approach. First, rather than focusing on the long tail of keywords, focus on what we call “keyword sculpting.” The concept here is to spend more time creating highly targeted ad text, bids, and landing pages for your head terms, rather than trying to uncover secret long-tail keywords. We generally find that 95% of clients’ revenues comes from the top 5% of their keywords; as such, making sure you are competitive on these top words is just a better use of your time than mining the long tail. To learn more about this concept, I recommend you download our white paper about the “Alpha Beta Account Structure.”

Second, to be effective at search engine marketing today, you have to basically do more than search engine marketing. I call this “wide tail” marketing (remember, you heard it here first!). The wide tail refers to all of the different channels now available to SEMs. Here are a few of the channels we currently use for our clients – ones you should also be considering for your business:

– Google AdWords
– Google Display Network
– Yahoo/MSN AdCenter
– Google Mobile Ads
– YouTube
– LinkedIn
– Facebook
– DoubleClick Ad Exchange

As online marketers, we have no choice but to go where consumers go, and increasingly, consumers are going to many more places online than they used to. As a result, simply throwing all of your money at text ads on Google’s search results (SERPs) is only going to get you part of the available traffic for your business. SEM today requires text and display ads (and video), mobile and desktop (and tablet), and social networks and search engines. With all of these different platforms to monitor, the long tail becomes even more difficult to manage, simply because it’s hard to imagine many people having enough time to spend on long tail and the wide tail simultaneously.

You can view the PowerPoint slides from my Ad:Tech presentation here:

David Rodnitzky, CEO


  1. David Rodnitzky April 9th, 2012

    This post is recommended for Steve Rowley in particular, since he is an aspiring SEM.

  2. Douglas Tarr April 9th, 2012

    You might get more interest if you called it the fat tail.

  3. Rikk April 9th, 2012

    I like the reasoning you present, however I would like to ask one question. Many times, isn’t the benefit in targeting the long tail, not that you can get it cheaper, but that the conversion rate is usually significantly higher? In fact, I have on occasion made significant bids on high value long tail to essentially achieve a bid lock on the keyword. No one was (realistically) going to out bid me for it. Granted, I do agree instant search is doing more to cut down on long tail, but at the same time, I am seeing an increase in some cases in the number of terms in a search phrase. The point being, sometimes people select a longer search phrase from the suggested search Google presents then what they normal would type in.

    Also, I would suggest that misspellings was never a valid long tail anyway. You always bid on broad matching and broad matches almost always caught mispellings anyway. The fact that Google now corrects misspellings doesn’t realy impact this much. Does it?

    Like you said, we see 95% of our ROI on 5% of our keywords. I think you could list Pareto’s 80-20 Principle to why the long tail is dead. Spending 80% of your effort for 20% return is not where you should be spending your time.

    Anyway, we have been chasing the wide tail. It’s just that now I know what to call it!

  4. David Rodnitzky April 9th, 2012


    I think we are really saying the same thing here. My argument (and I think yours) is two-fold:

    1. Google has reduced long-tail opportunities;
    2. Because you have limited time to spend on SEM, you should focus on the top 5% of revenue-generating queries, rather than chasing the tail.

    There are always going to be high-value, low-volume queries out there that work well, but they are fewer in number by the day.

  5. Mike Stewart April 9th, 2012

    Great article… I agree with Douglas, I like "fat tail" over wide. Even better would be "apple bottom tail"…. much more interest.

  6. Jobette Escobanas April 10th, 2012

    Well by testing, long tail is still working for SEO.

  7. David Rodnitzky April 10th, 2012

    Jobette, it may well work for SEO, I'm definitely not an expert in SEO.

  8. Seo Specialist Chennai Rajesh April 10th, 2012

    I always encourage my team mates to use Long Tail Keywords. Is this waste of time now?

  9. David Rodnitzky April 10th, 2012

    Rajesh, If you are doing SEO, it may still be a good option. I was speaking specifically with respect to PPC paid marketing.

  10. Reg-NBS-SEO April 10th, 2012

    Perhaps Google has made it harder to get an edge on a long tail but this is expected as the industry grows.
    It still boils down to conversion. On page factors.

    No matter what you call it, long tail is about the longer phrases, the ones with the most words.
    “Long” applies to frequency as well as length. “Fat” and “Wide” don’t seem to fit as well.

    I don’t think Google has reduced long-tail opportunities, they have done the opposite by expanding their suggestions.
    In my world, the long tail are the income generators.


  11. David Rodnitzky April 10th, 2012


    For SEO this may be the case. My article was about how paid search has been impacted.

  12. Terry Whalen April 17th, 2012

    Timely: http://adwords.blogspot.com/2012/04/new-matching-behavior-for-phrase-and.html

  13. David Rodnitzky April 18th, 2012

    You are so cynical Terry. It’s all about what’s best for advertisers and users, not about eliminating arbitrage opportunities for savvy SEMs! ;)

  14. PPCAssociates June 21st, 2012

    From the archives: The Long Tail is Dead; Meet the Wide Tail – http://t.co/pJy6oHGS

  15. Ryan November 23rd, 2012

    Hi David,

    Top-notch stuff. A quick question about the following:

    ” Today, Google determines geo-intent by either the geo-modifier in the keyword or the IP address of the searcher. As a result, when someone types in “San Francisco mortgage rates,” advertisers who bought the term “mortgage rates,” geo-targeted their campaign to include California, and (by default) have elected to show ads to users by search intent and physical location, will have their bid entered into the auction. ”

    Does this mean that we should no longer break our local campaigns down into the following?

    a) 1 campaign not using an IP-radius tareting, but bidding on keywords like “san diego plumber” etc

    b) 1 campaign using ip-radius targeting and bidding on keywords like “plumbers”

    Is the above to say that a) above is no longer needed since the ip-radius targeting by default will bid on “san diego plumber” when the keyword is “plumbers”? If this is the case, are these two campaigns canabalizing each other?

    Your thoughts are greatly appreciated!

Leave a Comment

David Rodnitzky
David Rodnitzky is founder and CEO of 3Q Digital (formerly PPC Associates), a position he has held since the Company's inception in 2008. Prior to 3Q Digital, he held senior marketing roles at several Internet companies, including Rentals.com (2000-2001), FindLaw (2001-2004), Adteractive (2004-2006), and Mercantila (2007-2008). David currently serves on advisory boards for several companies, including Marin Software, MediaBoost, Mediacause, and a stealth travel start-up. David is a regular speaker at major digital marketing conferences and has contributed to numerous influential publications, including Venture Capital Journal, CNN Radio, Newsweek, Advertising Age, and NPR's Marketplace. David has a B.A. with honors from the University of Chicago and a J.D. with honors from the University of Iowa. In his spare time, David enjoys salmon fishing, hiking, spending time with his family, and watching the Iowa Hawkeyes, not necessarily in that order.