Choosing keywords for SEO

+ See all sections of the How to SEO Guide

- Hide the How to SEO Guide menu

New clients often approach me with the simple request. "I want to rank for the keyword ....". But we have to wind the clock back first to avoid costly mistakes. I'll show you why keyword research is vital and how to do it effectively.

Why you need to consider the keywords you want to rank for

Before you start working tryin to rank for a keyword or phrase you need to know:

• Does anyone actually search that keyword or phrase
• Will that keyword or phrase be of any benefit even if it did bring in visitors.
• Arre there alternative keywords or phrases which would be better or which should also be targeted.

Now you're probably thinking those are three very sensible checks, why wouldn't anyone want to do them? The simple answer is that these activities involve time and cost and in the excitement to 'just get on and rank' many people lose touch of their senses. Don't be one of them.

True - I could just take cash from clients and search engine optimize for the keywords they have chosen without doing any checks and then shrug my shoulders when they complain that the ranking they paid $1,000s to achieve "doesn't work". But I prefer happy clients because they are walking, talking advertisements for link2light. In the long term that makes more business sense than grabbing a few bucks now.

You're reading this which means you want to rank and you probably already have some keywords and phrases in your mind. You might even be thinking "If I could only rank for those my website would be a success".

Maybe. But how do you know? Luckily there are ways to test if what you are thinking matches reality in two steps so you don't waste time, resources and cash on pointless SEO.

Step One: Do internet users search the keywords you have in mind?

Once you have put together your own list of keywords that you think people search you need to open an account with Google Ads and use the Keyword Planner Tool - here's the tutorial. Its free, you don't have to start using Google Ads to access it.

Make sure you have used the 'advanced options' to set the language and geographical location that you are aiming for if this applies to your business. If you are running an online store, for example, and only ship within the US you don't want to go search engine optimizing for a keyword phrase that is only really popular in the UK or Australia.

Now the Google Keyword Planner Tool is going to tell you how often a particular keyword phrase has been searched over the last 12 months and it will give you monthly figires so you can see if that keyword is trending in anyway. But you have to be careful with this data. Here are a few pointers:

1. If you are not paying to advertise with Google Ads the numbers are approximate.
2. Treat the results with a pinch of salt as they do fluctuate so if one term has 1,000 searches a month and another has 1,100 you can assume that they have the same popularity. Use the keyword tool tomorrow and you might well see the opposite result.
3. If the keyword Tool says there are zero searches don't throw away that keyword just yet, I'll explain why in a minute.

While you are checking through your keywords also take a good look at the suggestions the tool brings up lower down the screen and add any of these which look good to your list.

Your aim with the Keyword Planner is to flush out any words and phrases that you won't be able to rank for because the competition is too fierce. That's a call you have to make but if you are doing your own SEO the chances are you won't be able to rank for keywords with more than 100,000 monthly searches.

If I'm starting a brand new site I'll be looking for keywords that barely register on the radar. 100 monthly searches or such like. These are the keywords that everyone else is ignoring because they don't look juicy enough so if I can hoover up ten or twenty of these I'm looking at 1,000s of potential visitors from the word go.

Once the website starts getting links I can start focusing on keywords with higher competition but thats for another day.

These rarely searched keywords are called Grey tail keywords or as search engine optimizers often say, "The low hanging fruit".

The Keyword Planner will give you some rough idea on how steep the competition is (Low, Medium or High) and how much you would expect to pay per click if you were using Google Ads. Personally though I find both of these wildly innacturate unless we're talking about keywords which are heavily searched. I other words ignore the ones with very large dollar values.

Step Two: Optional - other keyword tools

Google's keyword planner is free but there are also buckets of pay to use services which claim to be more accurate or give greater insights or offer better alternative suggestions. AHrefs, Majestic, SEO PowerSuite. They're three solid services but in truth I don't use them that much.

They, like the Google Keyword Planner, are also unreliable in spotting low volume keywords and often claim zero searches. I've seen this over and over but when we launch a page that ranks for that keyword phrase it brings in traffic.

Yes, if you are optimizing a substantial website and aiming for seven figure annual turnover from the start then they're worth the investment - every extra keyword opportunity you can spot is another piece of content your team can create. But if you don't have a team you won't be able to action what you see anyway.

Step Three: Checking keywords with zero searches

Remember being number one for a word or phrase that no one ever searches is like placing a billboard in the desert. It may look nice but no one ever sees it so its something you want to avoid.

But you shouldn't take what the keyword tools say at face value.

So head over to Google and start typing in the phrase. If Google's auto complete kicks in and you see the phrase coming up that means it has been searched before, and regularly. The autocomplete only kicks in if Google has a fair amount of data which means that phrase is searched often by Internet users.

Now run the actual search, even if there was no auto complete and scroll through the results. We're looking for four things:

• Who ranks for this keyword phrase? Are they household names that you'll never stand a cat in hells chance of outranking?
• Look for Google's "People also searched" to get further keyword ideas
• Look for Google's "Suggested searches", usually at the bottom of the page.
• Look for Barnacle SEO opportunities - are there sites on the first page that you could get a link from, like directories or on a Map if Google shows one.

Adjust your list of keywords as needed, adding in ideas and stripping out the search terms that have no auto complete.

Step Four: Go beyond the keyword tools

Other tools that will help you check and find keywords include:

• Google Trends - is a keyword or phrase becoming more popular and what's going out of fashion
• Google Correlation - what words do people search at the same time as other words?

If you're targeting low volume keywords or phrases Google Trends and Correlation won't work that well but their always worth checking and might fire some keyword ideas you hadn't thought of so far.

Step Five: Do the keywords have value to you?

This step is missed so often but its critical. Just because a particular word or phrase is related to your site and searched often doesn't mean it will do you any good to rank for it.

Too many webmasters spend months working towards higher rankings and when they get there they realise the traffic generated is of no use to them whatsoever. These keywords and phrases are known as trophy terms. You can feel good about ranking for them, they can bring you lots of traffic but you get no benefit from either.

How can this be? Three reasons:

• The buying cycle - the keyword is a vague term people search. They are not ready to commit or buy yet until they narrow down the options. Like 'laptops for sale'. When the person decides which laptop they'll search for that make and model to compare prices and sellers. 'Laptops for sale' might be high volume but no one is buying at that stage.
• The re-search phenomenon - people search this phrase before realizing they would get better results if they used a different search term. The search results are such a mixed bag, most of which are obviously not relevant to them that they know they need to rethink how they are searching. You searh 'red mullet' and then realize you actually need to search 'red mullet recipe' because Google has totally failed to read your mind and understand your planning a dinner party.

So how can you test the value of a keyword to you?

There's a tricky cost/benefit decision to be made here. If the keyword or phrase you have in mind will, after you have considered the competition, be fairly easy to SEO for then get a page to rank and see if it does you any good.

However if the competition looks tougher, your website is already up and running and you can see that you will need to invest heavily (even if it is just your own time you are investing) then it is definitely worth running a test on the value of that keyword or phrase.

You do it by ranking at the top of Google immediately. How do you do that - through Google Ads and its not going to cost that much. For low volume keywords you'll be hard pressed to spend more than $50 per keyword. Many that I test don't even reach a total spend of $10 and given the time and money you'll waste working to rank for the wrong keyword - that's a bit of a bargain.

OK - here's the step by step.

1. Create a Google Analytics account - (Tuturial)
2. Set up goals in analytics ( Tutorial ) or if your site is ecommerce enable ecommerce ( Tutorial )
3. Set up sub goals in analytics - the set up is exactly the same as goals but now you'll include pages along the journey. Your main goal might be to get visitors to sign up to a service. Fine. But how many reached the sign up page and didn't act. That's good to know as we'll see later.
4. Join your Google Ads and Anlytics accounts together ( Tutorial )
5. Create an Ad campaign that targets the keyword (as an 'exact match') ( Tutorial ). You might need to create more than one campaign but start small if you are a beginner with Adwords.
6. Wait until any particular keyword has bought you about 200 visitors and then look to see how many of them achieved your goal.

That's it? Not quite...

Consider what you know - 200 people searched 'keyword X' and came to your site but few achieved your goal. Why?

• Your advert was poorly worded - Use Google Adwords Quality Score to see if you got your wording wrong (Tutorial). This might also help you later to refine your page title and meta description for better results in organic search.
• Your site doesn't convert - it may not be a problem with the keyword, it might be a problem with the landing page the visitors arrived on. For example, is there a clear call to action? Does your website design inspire confidence or remind people how the web used to look in the 1990s?
• Your expectations were too high - in ecommerce the industry rule of thumb is that for every 100 visitors only 1 or 2 will buy - a conversion rate of 1%-2%. If you got 3 sales and you're thinking bin this keyword - don't. Conversion rates are generally much, much lower than people expect. If your website is displayig adverts and getting paid when people click on them go even lower - 1,000 visitors might earn you about 10$ so 200 visitors would have earnt you in the region of $2.
• It almost worked - remember the sub goals I talked about earlier. Did people react to your content well and say, go to the sign up page, but then not take the final plunge and register. If so focus on the issue here rather than binning the keyword.

Now you know what you're looking at and you understand what to expect there's also a goldmine of data that you've just created and its waiting there for you to tap into regardless of how successful the test was.

Step six: What did ranking for this keyword teach you

Analytics will tell you for any particular campaign who liked your content and who didn't. Was it mobile users who didn't convert? Were there users from a particular country/location who warmed to your content more than others?

Use this knowledge to tweak your content or even subdivide it into multiple pages which better target individual groups of people.

Google Ads will tell you what search terms fired your ad. Yes, you should have set your keyword to 'Exact Match' so your advert only displayed when people searched that exact term but Google Ads is a little more easy going than most people realise.

If you're phrase was 'mens tennis shoes' and someone searched 'tennis shoes for men' or 'gents tennis shoes' then Google will say "that's close enough to be exact". Google ads will show you the search terms which actually fired your advert so lots more keyword ideas can be found there.

Again remember just because a search term fired your advert but didn't seem to do you any good remember to follow the chain right through. Was it because, for that particular search term, your advert appeared really low on the page or even on page two. Google ads will tell you the average position to give you some idea.

If the ad generated traffic did it perform badly for that search term because your landing page doesn't seem that relevant to the search term and so didn't inspire the visitor to act? Or did they act and achieve one of your sub goals? That means the keyword is not a complete failure but something on your site is undermining visitor confidence.

If you used multiple adverts or dynamic ads which advert performed best. This will tell you how to refine your page title and meta description for best performnce in the organic search results.

Choosing the right keywords is the dull part that is often skipped in the excitement of setting up a website. But failing to plan in planning to fail.

As you've seen there is so much you can do to make sure you search engine optimize for the right keywords and phrases which could be bringing the right visitors to your website within days.

Choosing the right keywords is absolutely fundamental to SEO. Everything else you are read or hear or learn can be pointless if you're optimizing for the wrong keywords.