This is part 3 of a Wordtracker review (part 1 – introduction, part 2 – related keyword search)
In this section we will look at how Wordtracker helps to identify vertical keywords, but first a couple of words about Wordtracker service. Three days into my seven day trial I identified a problem with UK vertical search, and published my complaint on GetSatisfaction. Within less than an hour I got several responses from Graham and Steven, saying the problem has been fixed, and indeed, it was. Absolutely fantastic service. They also extended my free trial by another week. Again, great!
I will continue with my paragliding example. To discover all the vertical key phrases that include your keyword, use Wordtracker’s “Include” facility :
Be sure to open “more search options”. This allows you to specify matchtype, misspellings and plurals. It is sometimes a good idea to browse through misspelt keywords, as they might be surprisingly popular. Google reduced the need to consider optimisation for misspelled words, but it may be worth experimenting. Just don’t overdo misspellings, you may risk your website’s reputation.
Wordtracker does not always identify the misspelled words - a broad match with misspelling on “paragliding” did not return results for “para gliding”, yet “para gliding” is one of the top 30 paragliding related searches. A search for “para gliding” did pick up “paragliding”.
It is at this stage where I start having major concerns with Wordtracker’s dataset for the UK. In the “paragliding example”, Wordtracker found 692 search terms, of which only 20 had a search volume of more than 10 – this with a data set spanning four years worth of data.
Here are some random comparisons between Google’s results and Wordtracker’s results :
| Keyword | Wordtracker
Volume |
Google
Volume |
| northern paragliding | 22 | 880 |
| paragliding harness | 2 | 1000 |
| tandem paragliding | 2 | 880 |
So Wordtracker would suggest that I prioritise “northern paragliding” way ahead of the other two terms, Google suggests they are in the same ballpark. I don’t know whether Google’s results are skewed, but I am pretty convinced they are less skewed than Wordtracker’s. Besides, Wordtracker is aggregating 4 years of the data for the UK. How do I know whether the highly popular search terms are popular now, or were they popular a couple of years ago?
Wordtracker tracks less than 0.5% of searches, possibly using a skewed sample population (how well does Dogpile and Metacrawler users represent the rest of society?) . I would certainly not hedge my bets on Wordtracker for my long tail strategy, because I might just be barking up the wrong numbers ….
In conclusion – is Wordtracker worth considering?
The functionality is great, the user interface is just fine, the data presentation is useful, but I have a great concern with the UK data. I would probably still use the tool, but not in isolation, and most definitely not for long tail keywords where Wordtracker reports less than 3 or 4 searches.
I wish Wordtracker all the success it deserves, but please please increase the size of your dataset (and if you can give us some cyclical trends that would be even better …)
Hi Hannes,
You’ve obviously spent alot of time with Wordtracker and investigated different options in your keyword research – it’s not often that a customer takes the time to write such a detailed review.
Related keywords are found by analyzing popular web pages that relate to the word that you entered. The words that are found most frequently are shown first – the Related Search doesn’t use the metacrawler data.
I’d like to clear up what seems to be a small misunderstanding about data sources. The US dataset does indeed have the metacrawlers as its source – this gives just under 1% of all US search. The UK dataset is a combination of ISP and search engine data, but due to contractual reasons we’re not permitted to say more than that. It is a fairly small sample set, intended to give a relative guide to search in the UK rather than an absolute picture, but customers have come back to us saying that they’ve found keywords that their competitors aren’t using. While we’d love to say that we have an even split of searches across all niches, this simply wouldn’t be true, and isn’t really possible for any keyword research tool to deliver as search data isn’t released from the major search engines to external agencies.
We’d agree that the UK dataset isn’t perfect, but it can be a good indicator of which keywords may be the most popular, particularly if you’re tracking keyword niche trends. While the dataset contains searches going back around four years, it’s still possible to detect changes in search behaviour if you’re tracking. Another thing to consider is that while search volumes on some keywords may appear low, Wordtracker can offer unique competition data based on MajesticSEO’s independant crawling of the web – sometimes it’s not necessarily the case that the keyword with the highest search volume is the keyword with the highest conversion rate. With any keyword research tool, it’s important to remember that the only real way to measure a keyword’s usefulness is live testing.
I’m happy that you like the interface – our development team has put alot of work into this. You’ll find that over time, Wordtracker will be developing further tools to help with SEO campaigns, as well as further work on the main Keyword Tool.
Good luck with your research, and please do let us know if you have any other comments or questions.
All the best,
Mal
Excellent post. Hope to read more excellent posts in the future.