Software Type: Web Based
Cost: $189 a month

Keyword Intelligence was launched by Hitwise earlier this year as a way to access the small business market with their search engine keyword intelligence that couldn’t afford their enterprise offering. Here is how they describe their service:

“The data featured in the Keyword Intelligence product is based on search engine usage of over 25 million people, across all the major search engines, including Google, Yahoo! and MSN.

Keyword Intelligence data is a subset of the more extensive search term data available within the Search Intelligence features, found in the Hitwise Competitive Intelligence Service. Search Intelligence helps marketers understand which keywords are driving traffic to their competitor websites together with charting and analysis tools for maximizing search marketing campaigns. “

At $189 a month for the top 1000 keywords, I would hardly call this service inexpensive. However, sometimes this kind of intelligence can pay for itself hundreds of times over if it really is as valuable as they say it is.

I checked it out and here is what I found.

My Experience

Once I had signed up and set up my account I was ready to jump right in and get to work. I found out that this service will supposedly work from only one machine which is a shame as I have multiple machines and work on different ones depending upon my location.

Once I logged in, I input my usual keyword search phrase “fly fishing” into the text box and hit submit. The system generated a large list of over 1500 keywords which looked great on the surface.

Unfortunately, the only way that search volume was presented was as a percentage. The volume for each keyword was presented as relative to the rest of the keywords on the list. They take the total keyword volume and then for each keyword present the percentage of traffic that keyword contributes to the total volume. However, they seem to have a threshold whereby keywords that fall below that threshold (don’t get many clicks at all) are not given a percentage and do not form part of the whole volume measure.

The other interesting measure they gave for these keywords was an indication of “Success”. This basically means the percentage of people who were happy enough with the results of the search that they clicked through. For example, the term “fly fishing” gave only 73% as they obviously were not happy with the results. Perhaps that term generated results that were too general and did not meet the specific needs they had.

As there was no indication of the actual volume of the sample size, I decided to check out some of the keywords specifically and exported all 1500 words. I looked down the list to assess the quality.

What I found disappointed me for a service that is as expensive as this one. As I went through I discovered that perhaps the real reason why they give only a relative volume measure is because the data sample size is simply not big enough. This means that the pool of keywords that they gather the data from is too small and the results produced lack statistical significance.

For example, the seemingly 27th most popular key phrase was:

oregon fly fishing guide tomwatersriverguide 

This seems to me to be a one-off phrase. I don’t think there is a high chance that more than one or maybe two people would be entering that term in the search engines more than once. Some may argue that it is an automated query to create these kinds of entries for exposure. However, I would disagree. These kinds of queries are normally URLs, not arbitrary search terms like this. I typed this phrase into both Wordtracker and Keyword Discovery and it was not within the top 500 keywords with “fly fishing” in it.

I decided to check out the industry search to see if I could drill out some more value from this service. This is an add-on service that you get as part of your subscription. For $189 a month you get to select five industries to analyze what the most popular keywords and phrases are for that industry.

Unfortunately, I was disappointed here as well, mostly because the data appeared to be completely skewed. According to Keyword Intelligence, the search term “cabelas” (which is a online outdoors store) was 13 times greater than the search term “fishing”. Clearly this didn’t make sense. I checked this term on Keyword Discovery and their data made a lot more sense. Their results showed that “fishing” was searched 10 times more than “cabelas”. Quite a contrast. Similar results were found at Wordtracker.

Why is this data skewed? I think, for two reasons. One thing they taught in my psychological research and statistics classes at college was that the sample size is important to get accurate results. It appears that the sample size is too small for this data. I think the sample size would become statistically significant only on very popular keywords.

Secondly, I think this data is skewed due to the way they sample it. Think about this scenario. You have to conduct a survey that asks the question, “Are you hungry for some donuts?” Now do you think the data is going to be different if you did that survey at the front door of Dunkin’ Donuts compared with doing exactly the same survey in the same town, but asking the people who are going into the a book store? Of course the results will be different.

So how does this relate to Keyword Intelligence? In this case they don’t collect the data directly from the search engines, they collect most of the data on the websites that the search finishes up at. Consequently, you are going to get search terms that relate specifically to those particular websites. For example, look at the industry search “fishing”:

1 cabelas  
2 bass pro shop 
3 bass pro shops
4 cabela's
5 bass pro
6 cabellas
7 cabelas.com
8 www.cabelas.com
9 fishing
10 basspro

The top keywords for the industry are not generic search terms. They’re simply the terms that reflect the words that are found on the sites that the data is sampled from. Keyword Intelligence either needs to get the data from a larger cross section of sites, or else remove the terms that relate to those individual websites.

They do not have any other tools for Keyword Research.

Summary

So what did I end up with after using Keyword Intelligence?

• 1500 semi-relevant keywords with relative demand and success measurements. Many not so relevant industry terms.

Strong Aspects

The strong points of this product:

• A different data source.
• Lots of keywords
• Nice interface.

Weak Aspects

The weak points of this product:

• Expensive.
• Skewed data set.
• Limited data.
• No PPC or other useful tools.
• No actual volume shown.
• Good only for people working in very popular and competitive keyword environments.

Who is it ideal for?

People working in very popular keyword environments.

Ratings

For Niche Miners:
For Search Engine Optimizers:
For Pay Per Click Advertisers:
For AdSense Publishers:
Usability:
Value:

Site: Check Out Keyword Intelligence Now!