47 replies to this topic

[DanThies]

Just to go back to keyword research for a minute, I'll toss in a few more handy measurements in looking at competition.

1. "allintitle" search on Google - if the search term isn't in the title, they aren't targeting it. The number of matches here is a much better indication of "how many" sites are competing.

2. "Average Toolbar PageRank" of the top 10 sites. If your page shows PR6, and the average page in the top 10 is 5, all it will take is a little optimization and maybe a little link-building to compete on Google.

3. Total Link Popularity of the top 10 sites on Inktomi. Quantity is more important than quality when you get away from Google. If you can compete on Google, Inktomi may be tougher, or vice versa.

4. Pay per click bids - what does it cost for the top 3 spots on Overture? For the top 10? If the bids are high, there is competition whether it's apparent now or not. There will eventually be heavy competition for free listings whenever there is heavy competition for paid listings.

Those are "quick" measurements. A more in depth analysis would consider the text of links pointing to the competing sites, the quality of optimization techniques they have employed, etc.

[cline]

There are a lot of cases where search terms important to the client don't even show up on Wordtracker or any other research tool. Especially in cases of a client who offers an obscure product or service, it's really important to pay attention to what the client actually provides. The actual traffic on the terms important to them may come under the research radar, but yet provide highly profitable traffic to the client.

I took on a client earlier this year that was exactly in this situation. After I did the keyword research, I had this "oh ****, what am I going to do?" feeling. So I researched their obscure products, bought Adwords on every obscure term I could find, and I waited. Indeed there was traffic out there. Eventually I got enough of a pattern with that traffic that I knew what to optimize the site for. It was a lot of work for a modest amount of traffic (although it was a 10 fold increase for the client, so they were thrilled), and since the client's products have high margins and are very hard to find, the campaign has been highly profitable for them because the SEO work made those obscure items vastly more findable.

[BrianR]

Well said, Cline - happens a lot with the highly niched B2B sites I often work on - WordTracker is just about useless.

I like your AdWords strategy - I'll normally use Ink PFI to test the more important obscure keywords, but your AdWords strategy is another good one - thanks!

BrianR

[Scottie]

IMO- In my opinion. (IMHO is in my honest opinion or in my humble opinion.)

Inktomi PFI is pay for inclusion. You pay an annual fee per URL and Ink will spider and update the page on a regular basis. It's a good way to see changes reflected and how they are affecting your rankings. PositionTech is one of the companies that offers this service.

[btreloar]

Not to bash WordTracker, because I do find it an extremely valuable service, but I often have problems with niche keywords. When you're working with phrases that are searched 10-50 times in Wordtracker's database, they can very easily be thrown off by single individuals searching.

I recently came across a term that WordTracker showed with a frequency of 11. Yet the spelling had been mangled so badly that I find it inconceivable that more than one person would spell it that way. So those 11 searches were one individual.

WordTracker results need to be taken with a certain degree of skepticism, and as was pointed out earlier, relative positions are often more valuable than absolute frequency counts.

[cline]

well, you can limit your daily Adwords spending to prevent runaway expenses on research and targeting errors. But, there's really no substitute for frequent manual monitoring.

[cline]

By "manual monitoring" I mean log into your Adwords account and look at the data.

[DanThies]

I recently came across a term that WordTracker showed with a frequency of 11. Yet the spelling had been mangled so badly that I find it inconceivable that more than one person would spell it that way. So those 11 searches were one individual.

Yeah, that's either a really dumb person using the search engine to navigate, or someone running rank checking software.

We always find those when we're doing research for our clients. Usually it's obvious, like when the count for "fred's widgets" is higher than the count for "widgets."

In "real life" it would be really unusual to find a two-word search term that's less popular than a 3 or 4 word variation. It happens all the time on Wordtracker.

I have little trust for the counts below 20, since that could easily be someone running a weekly rank check that includes Dogpile and Metacrawler.

[omahonydonnelly]

Great thread!

We have to remember SEO is an "art" not a science -- and Wordtracker is just one of many tools to use to create a page optimised for a specific product/topic and market.

It has helped me to find some great niche phrases that have been really successful for my clients. Phrases with enough searches to make it worth optimising for, but with very little competition.

Being outside the US, I also have to factor in that most of my target audience are not using the metacrawlers that Worktracker uses, and that regionality is a much bigger issue.

Often when I am short on time I do "compressed searches" only. Any opinions on whether this is short changing me?

Thanks, Ann

[DanThies]

Ann:

"Compressed" search just changes all the search terms to lower case, and removes unnecessary punctuation. Which is pretty much the basic expectation we would have for what a search engine does.

The "comprehensive" search tries to do word stemming, but what they really do is "partial word matching." This means that "poo" and "poopoo"
will match, but also that "poo" will match "poorhouse" and "spoon."

[Fritz]

Don't want to start an unnecessary new thread so I'll jump in here...

I am in Wordtracker today for the first time and am a little confuzzled. (And I feel like the pressure is on because I only paid for a day of use. May upgrade later.)

My challenge is to get my mortgage website some exposure in the search engines. Highly competitive, obviously.

What am I missing when I see that "home mortgage" has had only 380 searches in the past 60 days? Obviously that just ain't right.

Also...any particular advice on using Wordtracker to optimize a site in a highly competitive market? (I know that is too broad but...thought I'd throw it out there anyway...)