Keyword research is an area of SEO that all too often gets overlooked. I like to write about it but I like even more to actually do it. It fuels not only my love of crunching data but also my enjoyment of making judgments.

Keyword research seems like a straightforward task; you pull up a huge list of keywords, narrow it down to the relevant ones, weed out the ones that get no traffic and dig for the ones that don't have much competition. Job done.

Very rarely, does it actually work like this. Well certainly not if there is a client. Doing really good keyword research is about investing time and effort into balancing the requirements of all areas of the business. Sales want more leads, marketing want 'on-brand', management want 'trophy keywords' and so on. But that's a story for another time.

If you are tasked with doing the keyword research for a campaign then you hold the power – so use it wisely, it could be the difference between thousands of targeted visitors and the digital equivalent of tumbleweed.

All other aspects of SEO are directly linked to the work done at the keyword research stage – so why do so many spend such a small amount of time doing good research? The industry is guilty of rarely recognising the work of the keyword researcher but a good 'hit list' forms the bedrock of a successful campaign and is really foundation of a successful website.

So how can you do better keyword research?

Think like a user

It is the basics of human psychology - speak the other person's language and they will listen to you.

Put yourself in the shoes of the customer. I've worked on client's websites that have never had SEO before and to be frank their website is horrendous. I could sit there for a fruitless 20 minutes trying to decipher what the business actually offers – I'm getting paid, do you think a customer will hang about this long? I doubt it. This might seem like more of a usability and conversion issue but actually it is something that really good keyword research can help alleviate – you can't improve conversion if you A) don't know what people are there for and B) Don't speak the user's language (metaphorically).

If you think like a user and develop keyword lists that are centred around potential customers then your clarity and purpose can permeate across into other areas of the campaign. Everyone wins.

Develop a strategy

Without heading into the realms of corporate jargon; developing a keyword strategy is so important. Knowing the strengths of the website you are working on is important here because it means you have an idea of the kinds of keywords you are looking for.

For example, if you are working on a new site with relatively few backlinks and low domain authority then finding generic keywords to target is probably pointless.

A better strategy would be to look for immediate keywords to target that are longer tail and then over say an 18month to 36 month timescale have medium term and longer term keywords to target that are more and more generic.

E.g.

IMMEDIATE = cheap patterned laptop skins

MEDIUM TERM = cheap laptop skins

LONGER TERM = laptop skins

That's an example of a concept that I have found to work really well. Specifically because the website will already be targeting the longer tail phrase (and hopefully will have anchor text backlinks for this term) you can then break down the backlink anchor text and on-page content until you are successfully targeting the more generic phrase with more search traffic!

You could also opt to mix targeting of long tail and generic terms or just target many hundreds of long tail terms rather than one generic term. Perhaps if you are working on an authoritative site you could well just dive straight in with some big generic keywords.

Dig deeper

If you want to do really good keyword research then you need to invest some time in going the extra mile. Just because your data shows that a keyword has relatively little competition, you should really look a little closer to ensure that you're not going to be competing with one or two high-authority domains. It would make targeting

A Google search on the keyword and an investigation (link profile, on-page factors etc)  of the top 10 should give you an idea of what you are up against, it will also help you to understand what's working and what's not in that vertical – crucial if you are serious about jumping in there.

Develop natural keyword variations

This point gives me the chance to include my favourite joke about this industry...(you might have heard it but permit me the indulgence) "How many SEOs does it take to change a lightbulb, light bulb, bulb, lamp, light, eco-bulb, bulbs, fluorescent tubes?!"

Joking aside, in essence what you are trying to do here is to come up with closely related keyword variations that can be used with subtly across the SEO campaign; in linkbuilding (helping to improve the 'naturalness' of your link profile) and your on-page efforts (helping to target multiple keywords with one page). For example, shoe repair shops Chester and Shoe repair shop in Chester.

Normally I'd say it is a bad idea to target more than one keyword per page (usually more effective to optimise for one really well) but if you carefully select really tight-knit keywords you can seamlessly and easily integrate them into your work.

Remember that too much linkbuilding focused on one keyword with zero variations could result in a Google penalty for that phrase on your site so that's where variations come in handy.

Compare data from multiple sources

We all know that to do anything less is foolish and/or lazy. Cross reference your data between different tools and data sources. If something doesn't look right then check it out and if all doesn't seem spot on then check back tomorrow because even software can have an off day.

Utilise technology but don't rely on it

Robots can't replace people (not yet anyway). I am a big fan of technology particularly when it makes my work better (and my life easier) but when it comes to keyword research I prefer to utilise technology rather than completely rely on it. Experience and common sense are very much a part of doing good keyword research. Unfortunately they've yet to develop software that has these attributes in-built. You just can't automate really good keyword research.

Don't get bogged down

This might seem like some odd advice from someone that has just spent 500 odd words telling you to go granular and really get stuck into research but it is also important to not get too bogged down in detail. Obsessing over minute details will only make you mad and probably won't make much of an impact on results.

You can easily spend hours poring over graphs, statistics and charts when doing keyword research but remember that

Above all, use your common sense. If in your heart of hearts you feel that a keyword just isn't right for whatever reason, move on. If you do SEO for a day job and you've got some experience you will be surprised how often your gut instinct is right on these things. In summary, don't let your own prejudice cloud your judgment... strike a balance between what you think and what the data tells you.

By James Agate - http://www.skyrocketseo.co.uk