10 replies to this topic

[jerry]

Jill, I know you say to take Wordtracker data with a grain of salt (or something like that), but I usually consider it to at least be in the ball park.

Sometimes I see results that look like a total joke. For example: Wordtracker says that a KEI of 10+ is GOOD, 100+ is VERY GOOD and 400+ is EXCELLENT.

Today I saw the following data on Wordtracker:

south beach diet, KEI = 4,375 and the expeted hits per day = 7,339
atking diet, KEI = 3,889 and expected hits per day = 10,624
weight loss, KEI = 946 and expected hits per day = 18,159

To me this says that these popular terms would be easy to optimize for and you would be getting more hits than you could handle.

I don't have a site about diets and don't plan to have one, but as the country song says, "What were they thinking?"

To me this says that it would be no trouble to get a top 10 ranking with these popular phrases. I don't believe it.

Likewise, when I find keyword phrases that have a KEI of 3.0 or even 0.3 should I really forget about them?

Jerry

[Jill]

I don't say take WordTracker with a grain of salt, I say don't look at KEI as it doesn't really tell you much of anything.

If you read through the other threads in this Keyword Research section of the forum (oops, I see you didn't put this post there...I'll move it), you'll see lots of useful advice on how to judge the competitiveness of your phrases.

Jill

[jerry]

Jill, I didn't mean to mis-quote you. I was going from memory. Now I understand that it's just the KEI that you say to take with a grain of salt or forget about all together. I can see why based on the above data.

Thanks for your fast response.

Jerry

[DanThies]

KEI is useless. Try this formula:

Popularity * Relevance = Targeted Popularity
Popularity = number of searches
Relevance = % of searchers who would find your site relevant

You can get a "popularity" count from Wordtracker. You and/or your client can come up with relevance. So if the search term gets 1000 searches a day, and 80% of people using that search term would be interested in your site, the "targeted popularity" is 1000 * 80% = 800.

Now if you rank your search terms based on the targeted popularity, the next question is how much competition there is, because if you don't have any chance to rank in the top ten for a search term, that might not be a good target right now.

Some ways to measure competition, that are better than what Wordtracker/KEI use:
- search on Google for "allintitle: search term" to see how many pages have that search term in the title. Using keywords in the TITLE tag is basic optimization, but it's a better indicator than counting all the pages in an "exact phrase" search, which is what Wordtracker uses.
- Using the Google toolbar, visit the top ten sites. Take the average PageRank score displayed in the Google toolbar for those sites, then compare it to that displayed for your page.
- Use Marketleap.com or some other service to check the "link popularity" for the top 10 sites on Google, Hotbot (using the default Hotbot/Inktomi search), and Lycos. Compare this to your own site's popularity.

[don1]

I see that many people here use Wordtracker. What for then, if KEI is "useless"? I would also guess that most people are paying for the subscription. If that is the case Wortracker is useful for something. What? Tell me so when I sign up I use my time wisely.

[qwerty]

Don, it's useful in that it helps you go past your own brainstorming on keywords. You tell it what you think you ought to be optimizing for, and it will show you similar terms people are actually using to search, and how much they're using them. Some of them you may not have thought of, and others you may not have realized were so popular among users.

I agree that KEI is pretty much useless, though.

By the way, you don't have to buy a one-year subscription. You can get it for as little as a day at a time.

[BrianR]

The really useful thing about WordTracker is the way it can help you de-jargonise your keywords - eg. you're thinking of optimising for the industry term 'gardening implement' and WT suggests that most people are using 'spade' or 'shovel' - it can be a real lifesaver at times.

BrianR

[DanThies]

Wordtracker is the most reliable source, if you want to determine how often a particular search term is used.

You can get higher quality data from Adwords, if you run a campaign, but you have to know in advance which search terms you want to investigate, and it takes time.

You can also get free data from Overture, but it's very unreliable.

[don1]

OK, very funny folks. Are you all getting referal fees from Wordtracker? I bought into it and have been using it for several hours. What do I find? I find that there are some new terms, but I also find they must be hitting my website store (that is the part I am working on) because the terms they are suggesting are the ones I use as keywords in the same order I have placed them (and some of them are mizpelled on purpose).

[Jill]

Don, you are correct! That's how they get ideas. They quickly go out and visit the sites that have those terms on them, and then bring back the keywords from the Meta keyword tag.

Jill

[awall19]

I think eventually (way less than a year out) google adwords and overture will offer a competitive thesaurus with their products. the idea would most probably be a low cost solution which does a great job of adding value to the advertisers accounts. until they do though, wordtracker will continue being used by many SEO people.

I doubt they will use the meta keywords tags like wordtracker, but I think they will find some other way to determine what like terms are (most likely from on page copy). now the meta keywords data is very accurate, but many (about all) search engines are ignoring this data. going forward (a few years out) this data will become less reliable for two reasons
1.) grammar and culture change (and the world is somewhat turbulent right now)
2.) many people are no longer using the Meta Keywords tags

for now i think wordtracker is certainly worth a $6 one day trial though