We are currently going through the keyword analysis stages of the search engine optimisation process for one of our clients. For me its the most important part of the SEO process. Choose the right keywords/phrases and the rewards are fantastic, choose the wrong ones and you’ll be scratching around at 3am in the morning trying to build more and more links from social media networks in order to save your reputation with the client.
The nice thing about this client is they have a very niche product and that means we are more likely to find less competitive keywords. However with very niche keywords the problem often arises that you can’t get any data on who (if anyone) is searching for those keywords. If no one is searching for those phrases, is their any point in optimising a page on your site for that keyword?
Wordtracker and Keyword Discovery are fantastic tools for checking the viability of certain keywords (and we use them every day) but sometimes for very niche words the data just doesn’t exist.
At this point Pay Per Click takes on another use. Run a small campaign for those keywords you can’t get any data on, making sure you put enough in your budget to ensure your ad will be seen. This will allow you to get a pretty good idea about the search habits for that keyword. What’s even better about this tactic is you can get data for a specific location, which is something you can’t get with the keyword databases such as wordtracker (although they do now offer data purely for the UK). Working on optimisation projects in South Africa for the first time, this becomes an invaluable tool when our clients target market is local.
- Perky
Read more posts by Mike Perk
Related posts:
Good points Mike. Also, if budget permits, you can speed up the process by allocating a large overall campaign budget to the keyword(s). This means you can get the results faster. So, if the phrase is something obscure like ‘elephant tables’ create it in a dirty test Adwords account (there’s a reason for this I’ll come onto) at a high budget per day (e.g. 50) and set the geographical targeting options. As long as the cost per click is reasonable enough to get, say, top 5 positioning then you’ll know that the impressions figure matches the maximum possible per day. Run it for a few days and although its highly unlikely that Google will use up your budget allowance, it’ll give you a good idea of the average searches per day on that phrase.
A couple of things to be aware of – 1. ensure that the phrase links to page that contains that phrase (Google will penalise positioning otherwise); 2. do it in a ‘dirty test’ Adwords account. This is important because if you did it in a proper Adwords account and later decided to reduce the budget and cost per click down for that phrase (if using it for PPC) then Google will definitely penalise. However, doing it in a dirty account (I know, strange terminology – made it up) means you can then delete it completely and create it afresh in the ‘real’ adwords account.
Andy
“Dirty Tests”. I like the sound of that. Does Google correlate between the same credit card being used on the “dirty” and “real” account?
Can also give http://www.KeywordSpy.com a try for a keyword research tool, with results actually reflecting what advertisers are using at the current time.