Keyword research is an all too often under-appreciated aspect of SEO.

I've written a few keyword research posts here on Moz and that's because I believe it to be the blueprint of any successful SEO campaign.

Here are some of the more common mistakes that I see people make with their keyword research.

#1 – You're being Unrealistic

"It is better to have a bigger slice of a few smaller pies rather than not getting even a slither of a much bigger pie."

Keyword research appears to be a very straightforward task. You fire up your keyword research tool of choice and find the keywords that relate to your industry with the highest search volumes. Sadly, that's not the way to do it if you want to see real results.

To many businesses, high-competition keywords are simply out of reach – at least in the short and medium term. Part of good keyword research is about being realistic and selecting appropriate keywords for targeting that take into account the site's age, current authority and any future optimisation that will take place.

Targeting one word keywords is quite often unrealistic but it may also prove unprofitable – someone searching for 'Toshiba l670 laptop' is likely to be much further along in the purchasing process that someone who searches for 'laptops' – think about which searcher is likely to have their credit card out already.

There's nothing wrong with targeting generic keywords, I'm simply saying that if your campaign has limited budget and you need results in the short to medium term then targeting less trafficked, less competitive keywords is a much better way to utilise resources.

Lower traffic but lower competition keywords might not seem as exciting to target but if your website can dominate these areas fairly quickly then you are going to see far more traffic from the search engines than failing to effectively target a much more competitive term.

#2 – You're looking at broad match instead of exact match

A seemingly simple mistake but one which many people continue to make...

Search volume is of course a very important metric when it comes to keyword research but all too often people make the mistake of looking at broad search volumes rather than the exact match figure when using tools like Google's Keyword Tool.

There can be a huge difference between broad match and exact match search traffic for example:

There are 135,000 broad match searches each month in the UK for 'dog kennels' but only 14,800 exact match searches for the same keyword. Still, this wouldn't prove particularly problematic as this is obviously still a keyword worth targeting – it would knock traffic and ROI projections way off kilter if you do these kinds of things though.

The real problem comes when you choose to target a keyword like 'ladies leather handbags' which has a broad match search volume of 2,400 but an exact match search volume of only 260 – failing to base your research on exact match data might mean you think you are targeting a reasonably well-trafficked keyword when in actual fact, once you've factored in data inaccuracies, you could be looking at a very low search volume keyword indeed.

It is widely accepted that Google's Keyword Tool isn't entirely accurate when it comes to search volumes but using exact match gives you the best data available when assessing how viable a keyword is to target.

#3 – You're targeting plural instead of singular

It is very common to see a website targeting the plural version of a keyword but in most cases, it is the singular version of a keyword that people are searching for.

I see this most often on eCommerce websites where the site owner optimises category pages and because they sell more than one product, they naturally focus on the pluralised keywords for example "tablet PCs" which actually gets 91% less searches than "tablet PC".

I will readily admit that Google is much better at determining that a singular and plural version of a keyword are one and the same, but in many cases there are still differences in the search results. Failing to target the singular keyword can be the difference between your search listing being highlighted in the SERPs (=higher clickthrough) and it can also mean your website appears lower (even slightly) than marginally better targeted pages – that could be the difference between making a sale and not.

#4 – You're ignoring conversion

This one could easily turn into a rant for me because so often I come up against clients who want to rank for [insert trophy keyword] when in actual fact they'd do better (financially) targeting a different keyword or set of keywords. I try to explain that a keyword that brings in traffic is wasted bandwidth if that traffic doesn't convert. You don't hire my company to get traffic for traffic's sake...you presumably hire us to help you ultimately make more sales.

The online world is competitive and it's only going to get more competitive, therefore making the most of every penny being invested is vital.

This makes conversion and language analysis a vital part of keyword research. The human mind is the only software capable of performing a good quality 'conversion audit' of a keyword list because whilst there are programmes out there that can filter and sort keywords to make your life easier, there's no real substitute for industry experience and SEO knowledge.

There are some very basic indicators for example prefixes such as 'buy' might be a clear indicator that the traffic from this keyword is going to convert.

A keyword conversion audit is more complex than that however since each situation and market is individual. I find existing data to be a very useful way to determine which keywords are likely to convert well. If you have goal tracking setup with Google Analytics, you can easily determine the highest converting keywords your site currently gets traffic from, try to identify patterns in your highest converting keywords and then translate and apply this knowledge to other areas of keyword research.

#5 – You're selecting keywords that are out of context

This is yet more rationale to further humanise the keyword research process because most keyword tools struggle to compute words and their meaning in the way a human would.

For example, a searcher looking for 'storage' could be looking for a self-storage centre, boxes and other storage furniture for the home or even professional storage solutions for a warehouse or office.

Opportunities for confused targeting are abundant which is why it is essential the keywords you decide to target are highly-relevant and laser-focused towards what your business offers.

A good way to do this is to search manually for the keywords in Google and see the kinds of results that come up, you will likely be able to get a feel for whether the keyword is applicable to the product or service you intended to target.

#6 – You're failing to conduct keyword reviews

It is accepted that SEO is an on-going process but rarely are target keywords reviewed and audited. If a marketplace is shifting over time then you would also expect customer search behaviour to develop and evolve over time too – this makes regular keyword reviews essential.

In most markets, I find an annual review is perfectly adequate. Any time period shorter than this and there is a risk that targeting becomes a bit chaotic with efforts focused on new keywords before results on old keywords have been achieved or evaluated.

That being said, in some competitive and very fast moving markets a more regular keyword review may be required.

The aim of a keyword review is to:

• Weed out poor performing keywords
• Identify opportunities and areas for growth
• Shape your SEO strategy for the future

To do a strategic and actionable keyword review you can use this adapted version of the Boston Matrix that I like to use.

Large brands use the Boston Matrix to assess the health of their product portfolio and to identify where to concentrate their resources.

You can do the same thing for your keyword portfolio.

Sort your keywords into four categories in order to better shape your search strategy for the future.

• Question marks – these are keywords in areas where growth is likely but at present you're not getting the performance you'd expect. These are very often untapped keyword opportunities and you should plan how you are going to improve performance on these kinds of keywords.
• Stars – high-performance keywords and loads of room for growth – find ways to capitalise on growth. My advice is to focus your resources of gaining results in these areas for maximum ROI in a short period of time.
• Dogs – the poor performing keywords with little or no chance of growth – bin these in favour of other keywords, reallocate any resources to other areas.
• Cash cows – the high performing keywords that show little opportunity for growth – look for ways to enhance and maintain performance whilst identifying patterns and translating this learning to other areas or verticals.

What mistakes do you see happening in the keyword research process? Please share them in the comments section below...

By James Agate, founder of Skyrocket SEO and a regular SEO contributor to leading blogs and publications across the web.