Keyword Research for Niche Terms

SEO Question: I am trying to do keyword research for a client, and he is focused on a niche phrase that does not show up on any of the major keyword research tools. I was wondering if there was an accurate way to estimate search volume for these long tail keywords.

Answer: In this case, almost any sampling method you can think of is going to be wildly inaccurate.

If something does not get much search volume the easiest way to estimate search volume is to pull out your credit card and run a Google AdWords campaign targeting the keyword. Make sure your targeting is for search only and that your daily spend limit and bid prices are high enough that your keyword is showing for most (if not all) searches. If your ads are targeted to broad match (instead of phrase or exact) Google will also show your ads for many related keywords based on your keyword list. It also helps if you have an AdWords account that is aged and trusted so that they give you significant exposure right away.

I think a more practical solution than to look for exact keyword volume for niche terms is to use the keyword research tools to show their low search frequency, then use the same tools to show what words are important, and for parallel keywords to come up with rough estimates for search volumes for the relevant basket of related keywords. Then use a tool like my keyword list generator to come up with a relevant list of keywords to bid on.

After you get some market feedback from that PPC account, use your server logs and ad campaign stats to track short and long keywords your ads were relevant for, adjust your page copy, internal anchor text, and inbound link anchor text to help focus on the most important phrases, while also covering related phrases in a way that avoids keyword cannibalization.

A few other ways you can come up with relevant keyword phrases are:

  • track what you are ranking for in organic search results

  • look at the navigational structures, page titles, and page copy from competing sites
  • see what keywords Google recommends based on your URL
  • see what AdSense ads Google would target to your pages. Look at their ad copy, copy from their landing pages, and copy from other relevant parts of their website.
  • On rare occassions competitors may make their stats public for one reason or another.
  • You can also pose as an interested potential site buyer to acquire statistics.

There are also services such as HitTail that aim to help you extract other useful keywords based on the keywords you are already ranking for, and private tools such as HitWise and KeyCompete may show you a few keywords that public tools do not.

Published: March 13, 2007 by Aaron Wall in Q & A


April 24, 2007 - 4:26pm

I've been using WordZe[.com] and they seem to get me better results. What is your take on this service?

March 13, 2007 - 5:59pm

Assuming the niche term is relevant to the business product/service, then volume is probably not that important. It may bvery well be a small niche market that is highly pre-qualified. What is more important is backing up SEO efforts with other online and offline efforts to advertise the company/website in terms of 'that term' because if it has a low search volume, you can bet that with decent SEO he will likely rank high for 'that term'. Align the website with 'that term' via other media, articles, blogging, print etc etc and you're on a winner, let the search volume take care of itself.

March 13, 2007 - 7:43pm

hi, Aaron, how are you?

Great post (in fact, great posts, you've been busy!).

This tactic isn't just useful for niche terms - you can use a tactic similar to this to gather new keywords / base a whole broader campaign around:

1. pick a few core keywords
2. put them into an adwords campaign on broad match
3. use web analytics to gather all of the actual search phrases visitors used
4. rank those by number of searches, cut out the 'tyre-kicker' phrases & you have a hit-list of terms to focus on via SEO or to target more carefully using PPC

I've rarely seen mention of this tactic, but I've always thought of it as 'PPC keyword fishing'.


March 13, 2007 - 8:21pm

om strat -

I don't necessarily agree with that tactic. In fact, it seems like, more or less, a real bad way to throw away a client's PPC spend. I'd rather use some free keyword tools & find niche keyword phrases than pay for untargeted users to click on an ad & see what search term they used.

I suppose sure, if you could sign off on getting some budget in the preliminary stages to experiment with that strategy, then you might find it useful. But I wouldn't go about doing it.

March 13, 2007 - 9:09pm

Daniel's tactic sounds good to me. If the keywords do not have much competition, you could probably expect them to work even with a small bid, right?

March 13, 2007 - 11:00pm

I like to use this approach.

In many cases, you will spend very little with Adwords to quickly glean extemely valuable marketing information. In a matter of days or weeks, you can find out what your customers are looking for, what kind of copy made them click through to your site, and how to tweak your web site for better sales conversions.

This sure beats the "wait and see" guesswork approach!

March 14, 2007 - 1:20am

I like your insight, Aaron, but before I spend money to sort of guess, other keyword phrase research could be done on the specific results the niche product delivers, or specific problems people have in the niche field. I know it's basic, but there was still no mention of searching for keyword phrases related to results or goals or problems.

I really find it hard to believe that there would be no searches on a topic in a niche field that is profitable online. If that really is the case, however, offline is where to find that market. Check out the SRDS for lists of buyers, or repeat buyers in the niche.

Thanks for your work lately, Aaron!

March 14, 2007 - 1:44am

I ran into this same problem myself. I basically printed out Overture search data for the client and showed them where we could really dig into their niche. Once the client understood what type of search numbers we were looking at vs his competition, the client relented to let me choose the niche pending their ultimate approval. I ended up having to teach the client some ideas behind search then it was much easier to convince them to do it my way. Their business is doing very well now.

March 14, 2007 - 5:19pm

Aaron, I am new to the SEO world, but have become a strong addict. Thanks for your great blog.

This may not be completely on point, but deals with niche terms.

I am a web designer, but have become an SEO addict due to my positive experience with niche terms. For me, going for a term like "web design" for organic or PPC was a non-starter. But by choosing several qualifiers and going for "affordable denver web design" I was able to achieve a Google page 1 position 1 position.

All of the keywords we chose were qualified in a similar way, and were chosen to reflect our existing corporate values and the values of our client base. I've seen our contact form submissions quadruple. I've also started a PPC campaign for the same qualified keywords, and pay much less for what are much more targeted customers.

So anyway, from this newbie...niche terms have worked great for me.

March 14, 2007 - 7:12pm

Thanks for the mention, Aaron. HitTail recommends cutting down your time on keyword research by letting some nice algorithms do your research, then you spending your time doing writing based on these keywords (for natural search), or shoveling them directly back into AdWords with exact match (for PPC). The important thing is saving time on the research phase. We've implemented a crazy feature called "keywords forever" which basically uses every keyword you've ever considered in the past as a filter for keywords that you're thinking about targeting in the future. It's sometimes a trick to understand, but what this does is ensure that everything new that you consider, has never been on your consideration list before, and thereby saves you tons of time. Hope that helps.

March 15, 2007 - 7:49am

Thanks Aaron for pointing it out.

Add new comment

(If you're a human, don't change the following field)
Your first name.
(If you're a human, don't change the following field)
Your first name.
(If you're a human, don't change the following field)
Your first name.