Chasing our Long Tails

Rethinking the value of long-tail keywords
Once upon a time, everyone in PPC “chased the tail” on keywords to find keywords that no one else is bidding upon. But search marketers are starting to look at their long-tail keywords and ask if they’re worth the hassle.

The evolution of the long tail
The long tail— the concept of finding value in low-volume search terms got a bump into mass consciousness in 2006 when Chris Anderson published “The Long Tail”.

The concept of the Long Tail is a popularized view of a statistical norm that basically says: Although consumers buy only a small percentage of products in high volumes, there is demand for a wide variety of products in low volume.
This distribution curve illustrates a “long tail” effect (original illustration by Hay Kranen/PD, Wikipedia Commons license). Read more about long tails.

Think of this curve as it applies to popular search queries: There are millions of searches for “Britney Spears” (a “head” term with high search volume), but only a few for “Al Yankovic” and related Al Yankovic terms. As you move down the tail, you find a more fragmented audience—but these are people who are looking for something specific, like “Al Yankovic fan clubs”, and may be more receptive to your messages.

In the past, the cost to reach this audience through traditional marketing methods diminished its value to marketers. But search marketing allows us to target this potentially valuable “tail” audience at a lower cost than ever before, and, as such, PPC advertisers build huge keyword lists of these tail terms. The assumption is that tail words are more targeted, and therefore, more likely to convert. And since there is likely to less competition for the keyword by other advertisers, the cost-per-click may be less.

Mine’s Bigger!
PPC advertisers used to brag of their giant portfolios: “I manage 50,000 keywords!” We were told to expand, expand, expand. Mine your logs, ask your friends, concatenate lists, memorize a thesaurus and torture a linguist!

The idea of the long tail is still going strong, it seems. At CES recently, Bill Gates promoted the “Long Tail Olympics”, which might finally give a chance for stardom to badminton champions the world over.

But over here in PPC land, there’s been a gradual change in attitude about the tail and its eventual value in a keyword portfolio. Some campaign managers are starting to reduce their keyword portfolios, not increase them, and with a valid concern: long-tail keywords with low-impression or zero-impression scores may weigh down an overall quality score in ad groups. As a result, we recommend that you isolate these words into separate ad groups to prevent them from dragging down your campaign performance.

Also, the maturity of the market means that many long-tail terms can now be more expensive than their head term counterparts, since head terms are usually less targeted and may yield lower click/conversion rates compared to their impressions (”Britney Spears CD” as a term may cost far more than “Britney Spears,” since the former is specifically tied to a potential product purchase as opposed to, say, a public breakdown).

Keep Chasing the Tail
However, once you’ve set your low-impression keywords up in their own ad groups—and you’re aware of the ones that could potentially cost you more—there’s no reason to avoid long-tail terms. So go ahead—keep chasing the tail and looking for possibilities. And if you hate going to the IT department and asking for referral reports (or your tracking software is a pain to use), you can set up yourself and a few competitors in a service such as Compete.com or KeywordDiscovery, and quickly discover new keyword ideas.

Ultimately you may decide that some long-tail keywords cost more to manage than they’re worth, and that’s a perfectly valid conclusion, too. If it costs you more time to write an ad, manage bidding and track ROI for a keyword than it generates in a year, then heck yeah, cut it! But for those who love the idea of catching a hyper-targeted customer at a ridiculously low price, sometimes the possibilities are worth the risk.

– Dana Todd, Executive Vice President, SiteLab Interactive, Inc.

Posted by Administrator

[ Categories: Uncategorized ]
Bookmark with del.icio.us Send Blog via 360°

45 Comments Add your own

1. Pay Per Click Journal | March 26th, 2008 at 12:14 pm

Love the term “chase the tail!” Very true…keep chasing it as it’s more about quality than quantity when it comes to keywords.
2. Massive PPC Roundup for T&hellip | March 28th, 2008 at 9:53 am

[…] Chasing our Long Tails - Interesting thoughts about effectively targeting the long-tail keywords so they don’t weight down the quality level of your ad groups and campaigns. “Must read!” […]
3. Alexis Kauffmann - Galeria de Arte Géssica Hellmann | March 29th, 2008 at 2:48 pm

When it comes to probability, many people go beyond common sense. 50 thousand keywords for a single campaign is insanity. Great article.
4. QuickSet Learning | April 2nd, 2008 at 1:54 pm

Great article, Keywords can be very trying at times.
5. Nash | April 6th, 2008 at 11:54 am

A great tips for the newbies like me.

Cheers,
Office furniture computer chair
6. Suchmaschinenmarketing: D&hellip | April 15th, 2008 at 7:27 am

[…] dem Yahoo Search Marketing Blog ist in einem Artikel von Dana Todd (Executive Vice President, SiteLab Interactive) zu lesen, dass […]
7. JohhnyRock | April 30th, 2008 at 1:19 pm

First of all, consider the source: Gee, do you think that maybe, just maybe, the Yahoo folks would rather see us biding up and paying more for popular keywords instead of getting quality targeted traffic at a fraction of the cost? Hmmmm…

On my site (a real estate site in CT), in terms of View:Visit ratio, LT keyword visits out perform popular general keywords 50 to 1. Sure, we get 400 visits a month for “CT real estate” (@ $1+/click), but the view:visit is 6:1, half our average of 12:1.

On the other hand, its not unusual for a LT keyword to post, 50:1, 100:1, even 200:1.

Nice try, though
8. James M. | April 30th, 2008 at 1:20 pm

A lot depends on what you are selling. For us (a commercial real estate firm) a specific “long tail” keyword could mean a five-or-six-figure deal. I’d much rather pay $5 for “I need office space in downtown San Diego” than 10 cents for “downtown office”.
9. drboyce | April 30th, 2008 at 1:21 pm

While long-tail keywords might be on the decline for PPC, the real world of organic search and seo knows that there is a much higher conversion on most long-tail keywords. If there is a higher conversion, according to analytics, then perhaps I’m glad everyone esle is abandoning using these terms!
10. Tampa Furniture and Mattress Clearance store | April 30th, 2008 at 1:35 pm

We’ve “chased the tail” for some time now with much success. However, I appreciate the concept of separating the long-tail terms into their own ad group. I’d often wondered if they were hurting my quality score. Aparently, they are.
11. martone | April 30th, 2008 at 1:38 pm

…or you might use a service like clickable to save time managing the campaigns.
12. Kathy T | April 30th, 2008 at 1:46 pm

I found this helpful - even if it means I’ve been doing things wrong all these months. As a newbie (about 8 months into this), I’ve just loaded up keyword after keyword thinking that “more is better.” So, this article definitely helped me understand this whole strategy a little more.

I guess it’s over to the tutorials for me! I evidently have a LOT to learn…
13. Dan | April 30th, 2008 at 1:50 pm

“The Long Tail” is an extremely interesting concept. If you havn’t read the book, however, don’t bother. The paragraph above that explains the concept is perfectly sufficient. The entire book simply rewords this paragraph over and over and over again. Also, i liked post number 7.
14. Ross | April 30th, 2008 at 2:00 pm

1. The headline and storyline (”But search marketers are starting to look at their long-tail keywords and ask if they’re worth the hassle.”) does not match the actual article contents. Mr. Todd questions whether using longtail kw’s are worth it, then suggests that they are worth it, but simply to put them into their own ad group to not pull down the quality scores.

2. I agree 100% with JohnnyRock’s comment. You should always look at the putative intentions of the speaker (basically Yahoo in this case). So if you are watching a commercial by the corn growers association touting Ethanol’s environmental benefits, you really can basically guarantee that it’s a lie.
15. Erik | April 30th, 2008 at 2:01 pm

We enjoy great success with long-tail terms for both organic and paid search. As we provide B-to-B technical services on a global basis, longtail search terms for multiple industries and scientific disciplines work very well for us. By adding reasonable amounts of long-tail terms in multiple languages we further increase our results. As a result I drive literally millions of dollars into our global business at high margin. The time and effort required for longtail marketing are well worth it, at least for our global markets and technical services.
16. Smart Credit Card Choices | April 30th, 2008 at 2:02 pm

I’m with JohnnyRock and Ross on this one.
17. Brian Renner | April 30th, 2008 at 2:30 pm

So true. It’s cumbersome to manage tons of keywords without some sort of comprehensive bid management system. It’s best to focus efforts on the higher trafficked terms and test out different landing pages in order to increase conversion rates.
18. Bonsai Market USA | April 30th, 2008 at 2:49 pm

Maybe Yahoo is trying to save server space!
19. Jason | April 30th, 2008 at 2:55 pm

Thanks for the only good information on this page JohhnyRock.
20. Steve | April 30th, 2008 at 2:59 pm

LT is not only the only thing I can afford for our tiny business, but is also the only thing that makes sense. We sell a software product that lets the end user create CBT (Computer Based Training) courseware. CBT by itself is a useless term as there are millions of CBT courses available but only a small handful of products that let you create one. A million and one clicks on CBT is probably money wasted on paying for a million clicks for a product the end user is not looking for.
21. Dave | April 30th, 2008 at 3:33 pm

First of all, Dana Todd does not work for Yahoo. She is the President of SEMPO and EVP of SiteLab (as it says at the bottom of the article). As for the long tail, it all depends on your business. If you operate locally, then obviously you need longtail keywords with geographic modifiers or broad keywords constrained by geotargeting. I have to agree with the spirit of the article, though. Think of the 80/20 rule, if you get 80% of your results from 20% of your keywords, then why spend considerable time and effort to manage an exhaustive list of keywords that gets few impressions and even fewer clicks? Of course, if you can’t afford some of the higher volume keywords, then you need to stick to a more profitable niche.
22. Ronnie | April 30th, 2008 at 3:39 pm

I have found that brand recognition is a huge factor in determining whether to expand a list of long-tail phrases. In my experience, big brands perform very well with short-tail phrases, probably because users already know what to expect, in terms of quality, from that particular brand. Adding hundreds or thousands of long-tail phrases only dilutes the campaigns performance.

Again, this is my experience with larger brands. These types of campaign strategies should always be customized and tested.
23. Russell | April 30th, 2008 at 4:09 pm

Great piece,
Management of a campaign is everything and you certainly want to boost performance in any adplaced.

Thanks
24. Scott's Debt Advice | April 30th, 2008 at 4:47 pm

One thing I find useful is to work backwards. Create your ads before creating your keyword list and write each ad to focus on solving one specific problem that your site addresses/solves. Only then decide what keywords the user might enter that are 100% relevant to that ad.

You may end up creating more ads this way but that’s how you focus your traffic and lower your costs.
25. randy | April 30th, 2008 at 4:49 pm

It might be that the head words are slightly more profitable since the minimum bid have gone up in price.
26. The Unit | April 30th, 2008 at 5:10 pm

For the amounts of profit that I make on my obscure keywords that no one else is targeting, I will continue to chase my tail.
27. Al | April 30th, 2008 at 7:25 pm

I’ve also found it useful to partially allocate our advertising budget based on which search engine’s customers are more likely to receive our email responses to quote requests and service trial signups.

For example, we found that a high percentage of users of a certain search engine (starts with Y) also had email accounts with that same firm. Looking at the server logs, it turns out our quotes and trial signup emails were getting blocked and deferred by that firm’s email system for quite some time - even though we don’t send out unsolicited emails. So much for all that money spent.
28. Free Traffic | April 30th, 2008 at 10:47 pm

Chasing long tails is very profitable
for narrow niche market targeting.

Head terms usually for big companies
that have bigger budget to burn.
29. Web Business Secrets | April 30th, 2008 at 11:05 pm

Why not just get all your P.P.C. free?
30. Rich Boy | May 1st, 2008 at 12:22 am

Interesting ideas. I personally think long tail keywords can be worth the chase if you target the proper ones. Like it or not, PPC is not the most affordable or best converting way to get people to your website.

Its a good start, but you should diversify if you want to see long term profits.

Thats coming from a real rich boy.

See ya

Rich Boy
http://www.TheRichBoy.com
31. Daniel | May 1st, 2008 at 4:00 am

Interesting,But I see it differently from my own real experience.
I have 60k keywords in my campaigns since I have many products and landing pages, like most ecommerce has.
I use automation technology ( kenshoo.com ) for managing it.
Although brand terms are the most powerful (not 80-20 but…). I get VERY good ROI with low CPC on my long tail keywords, which are connected directly to the product pages.
Using the technology really helps in the management and optimization, and the bottom line it even saves me time.
32. Mike | May 1st, 2008 at 4:09 am

I agree with Daniel. I work for a price comparison site, and we need to have thousands of keywords. We used to run a campaign with a smaller amount of words but the shortail competition really drove up prices. (Maybe why Yahoo is motivated to write this article).

I have seen it with my own eyes, the more long tail keywords the lower the costs. I have words that get about 10 impressions a year but convert at least 2 or 3 times. These are exactly the words that Yahoo doesn’t want because I pay the minimum amount because there is no competition. Yahoo wants everybody to bid for the short tail words to drive up their profits.

We now use an automation solution to manage our search cmapaigns and it works very well. We get the benefits of the long tail with out wasting alot of time and work.
33. Barbara | May 1st, 2008 at 4:40 am

For a small business, long tail keywords are the most profitable, at least in my opinion. Not as many clicks, but a much higher conversion rate does it for me. What good does it do me to have a lot of clicks on a keyword with few sales as opposed to a small amount of clicks with a much higher rate of sales?
34. Reginald Poe Carelock | May 1st, 2008 at 7:11 am

Hello

This was a very interesting read. Definitely makes one think. #28 and #33 really has caught my attention. I was testing my keywords as I have not checked them since 4/9/2008. Which I will do after I submit this email. Being that I have a small business/small web page. What good does it do me to have a lot of clicks on a keyword with few sales as opposed to a small amount of clicks with a much higher rate of sales? Keep in mind that I am just starting out/starting my online business. I am four months in.
35. brojoe | May 1st, 2008 at 12:21 pm

Comments 28 and 33 ditto
It all depends on what you are providing.
I throw everything against to wall to see what sticks, and long term sticks more for us.
36. Discount Cruise Guy | May 1st, 2008 at 1:15 pm

Interesting article. I’ve been a long-tailer in the travel industry for years. It’s saved my company an unbelievable amount of money a boosted conversion dramatically. I think the more specific a person is in their query, the more ready they are to buy.

Good article.
37. Safe Cleaning Products | May 1st, 2008 at 10:13 pm

I agree with the idea of separating LT keywords into different ad groups. I like to keep them in a larger group, so I don’t have to write too many different ads for small groups of keywords.

Then, as I see some impressions on specific LT keywords, I might pull them into a separate group, since now they “deserve” their own ad.
38. R Dilip Kumar | May 1st, 2008 at 10:44 pm

The idea of the “long tail” is probably what we small marketers thrive upon. This is what generates sales for us and keep our cash registers ringing. Had we had to bid for the main keywords we would have gone bankrupt by now. With softwares like the Micro Niche Finder finding such long tail keywords have become even more easier.So keep chasing the “long tail” is a right quote.
39. Jewels of Denial | May 2nd, 2008 at 7:23 am

Interesting article, though I would have to say our results completely favor the use of the long tail.

Sure, head terms bring a lot of non-qualified people to our site, but it’s those specific tail searches that bring in the conversions.

Head = Hits
Tail = Sales

At least that how it is for our company.
40. UK Loan Express | May 2nd, 2008 at 12:38 pm

The longer the monkeyword tails the better.
Why…?

1. Autopilot PPC is possible and less mental stress as bloody PPC competition is not as hash;

2. ROI is FAR much higher;

3. Surefire PPC profit is almost certain.

4. Good cash flow is possible as total CPC is low.

5. Not fear of PPC spying tools like…:)

Good Luck everybody.

Ray
41. Carl Loop | May 4th, 2008 at 1:37 pm

Good information, and it seems intuitive that long tail KWs would become the most competitive words, since by definition, they convert better. Interesting idea about separating long tail and general terms in a PPC campaign.
Carl
42. Thersia | May 5th, 2008 at 2:14 am

The article is very interesting, and by reading the comments, I think LT keywords are worth the try. Less clicks, but more sales makes sense to me.
43. Will Nguyen | May 8th, 2008 at 12:54 pm

One major factor you didn’t mention was that the development costs for mining keywords and setting up their campaigns can be cost prohibitive depending on the size, cost, and benefit of the keywords themselves. For instance if your ppc person spends 7 hours (at $40/hour) to generate 50 long tail words that generate 2000 impressions over a year period with a click rate of 5 percent at a cost of $.50/click and (keeping score: $50 + $280 to the ppc guy) a conversion rate of 8 percent that generates 8 sales at $30/each, you actually lost $90.
44. UKLoanExpress | May 12th, 2008 at 10:56 pm

To Post 43:

I don’t understand why you will need 7 hrs just to **FIND** 50 long tail kws.

“For instance if your ppc person spends 7 hours (at $40/hour) to generate 50 long tail words that…”

Regards,

Ray
—–
45. Rachel Rokeach | May 13th, 2008 at 6:29 pm

I’ve tried both ways for a long time - since the days that overture used to be GoTo - and still find that the long tail works for businesses that I am in, where the head tail is so high & competitive that it eats all the profits.

Rachel Rokeach
http://www.PokerChest.com