How to Choose Keywords That Serve User Intent
This article details processes to help improve the pairing of keywords to your content by uncovering the intent behind a search. What is the intent? To be educated, to find a price, to compare brands, to compare products and/or services and so on.
This should be your #1 priority. Makes sense, given Google stresses the most “relevant” content ranks highest? Yet I see little mention of it on page 1 results for the very term I’m “trying” to serve. I focus on blog posts, but don’t be afraid to diversify your strategy.
The articles cover some excellent points, no doubt, but not one focuses on content and serving intent. We’ve become infatuated with links and white-hat SEO (which turns out, is mostly black-hat).
Whilst SEOs agree most link building tactics are becoming devalued, competition and similar metrics are the basis for most keyword strategies? Content still isn’t a core part of the strategy (for most). Hmm.
Failure to realise that most link building tactics are short lived, will ensure your content has a similar lifecycle; short. Google concurs, by every few months rolling out updates such as Penguin. Sure, low quality content does appear in search results, but it’s being culled; it’s a massacre.
Your main competition is not the site, it’s not the site’s links – it’s the content. It has to start with the content. No doubt, links are important, but they act as a catalyst for great content and they simply prolong the death of rubbish.
A test post I wrote gets 20,000+ uniques a month and out ranks the likes of AOL and Entrepreneur.com (look at me, I’m rich…not). I didn’t build links. After it gathered speed, I started internally linking to get some targeted traffic flowing through the site. Some extra page views, but no real change in the SERPS.
It ranks #1 for multiple terms with over 600,000 exact match results, #10 for a phrase with over 2,000,000 and #22 for a phrase with over 14,000,000.
Why? The content serves the search intent better than others. The keywords I used for SEO purposes indicated to search engines the intent I was servicing; the content serviced them. Many other results are half-assed 400-600 word articles. You know – articles that bored readers on article directories when content marketing was just a puppy. They offer search engines an indicator (using keywords) of the intent they are servicing, but they are not servicing it.
My effort? Over 2,000 words, well laid out, plenty of useful references, personal experiences and more. It doesn’t look amazing at first glance. Take into account not just the intent but also the types of people that are likely to be reading; it’s dressed well for the occasion.
You probably start with highly competitive broad match terms, drilling down to long-tail queries. I assume you have your list of keywords. If you’re a new , checkout Erin Bell’s article; it is PPC related.
Cluster keywords into sub-sets whereby they are likely to appear in the same article naturally. I generally target one, maybe two keywords, butalways just one intent. I don’t try and multi-task – I’m not that clever – I find myself trying to serve two or more different intents and instead, I serve none. This results in a reduction in page engagement from readers; you don’t want that.
LSI has evolved in a big way; a lot of the work is done for you. Many guides suggest including 3-5 LSI phrases in content, don’t do it for the sake of it. The post I mentioned above ranks for 200+ keywords; I think I targeted 2 or 3 on that occasion.
Let’s Get Searching…
You can see from the image below, Google will automatically offer some suggestions which will give us a clue as to what the intent behind the search might be.
Note down what you find as you go (they want to do research, they are looking for a tool etc.)
Twitter, AdWords and Facebook advertising all suddenly become possibilities for content topics, longer-tail keywords and search intents. But for me it’s keyword targeting for SEO that is of interest. Most likely, readers are looking for guidelines on keyword density (a highly outdated strategy and explained at the end of this article) are/or targeting methods.
Google backs this up by serving the intents of multiple queries. This includes results that focus on SEO best practices but also some related to Twitter:
Repeat the process for closely matched terms in your cluster; they might be a better option. I know the term “research” is going to bring up a lot of tool related suggestions, as I use the Google tool most days. I have an easier route to market. Although I expect the oddity on this occasion, this research helps me uncover oddities for phrases where I might not be so wise.
“Keyword research” gets 9,900 searches globally, whereas “keyword targeting” gets just 210. In terms of competition they have 4,000,000 and 101,000. I’ll fry the smaller fish.
This post is highly related to keyword strategy (which popped up using when I searched keyword targeting in Google’s Keyword Planner). Google would have most likely picked that up as a LSI term. The term has 320 search volume and 170,000 results, not dissimilar to “targeting” mentioned above.
And again, it hints at “for SEO”; a term that would naturally appear here. The search results produce similar content to keyword targeting. Again, obvious in this case, but a process worth executing for each keyword, just in case search engines view the intent in a different way to what you do.
So working in keyword variations that incorporate targeting, strategy & SEO will suit this article. Of course, I’ll be steering clear of “research”, given most people are actually looking for the Google Keyword Tool and 4,000,000 results is a lot of competition to deal with.
You could go as far as targeting a keyword for each section, maybe you already do. But, why would someone dig through a 2,000 word article that “covers” their intent, when another focuses on it?
Understanding the User
Phrase dependent, you may be able to ask and answer lots of different questions about the user; to help further target the content. How qualified is the Googler? What’s their income? What decision do they want to make? What do they need to learn? Those sorts of things…
My assumption here, the readers know the basics (“targeting”, “strategy” and “for SEO” rather than “how do I pick keywords?” or “why are keywords so important?”). They are webmasters and business owners looking for some keyword strategies that will aid their SEO efforts.
This whole process is actually quite helpful to determine if your target audience is likely to be the person searching your keyword. I.e. It’s great ranking for a keyword, but if your buyer is not the one searching it, what’s the point?
Define your focus early on. I’m not focusing on Twitter, Google’s Tool or any of the other possible intents; I’ll let someone else do that. I have one focus for this entire article; keyword targeting for content writing and SEO. Therefore, I focus my comparisons towards results that are trying to service this intent – to see what’s working and what’s not.
Compare the top 10 results. Compare anything you can. Content type (video, infographic, blog post)? Length? Layout/structure? Images? References that add value? The key points/takeaways they provide the reader?
These answers will inform you on the intent that Google thinks the searcher has and which results it thinks best serve them (which can be combined with the intent you think searchers have). Be objective, as some keywords may not have been serviced correctly; Google has to display something! Always question whether the content displayed is actually best serving the intent.
You will notice many of the top results are leading websites. Don’t be fooled into thinking the ranking is simply because of their authority. My findings even researching this post, is that the same sites keep appearing in my results for a reason; because they are the best (in general) results available.
Now, put yourself in shoes of the Googler. Will they come away from top 10 results feeling like they are still uneducated? What reasons can be attributed to the intent not being fully served? Were they expecting a “next step” and not just an explanation? What else could be added to educate the user?
Are some pieces of content actually overeducating users? A “definitive guide” when users just want one piece of that puzzle? On the other end of the scale, content fluff – content for the sake of content?
Then there is spam content. Maybe I have to attribute rankings to links (or lack of quality content for Google to list), but that’s still a gap to target (don’t model your content on these results). These sites will get slapped eventually; hard.
Check out this cringe-worthy example; it ranks page 1 for “content writing”. The rest of the article isn’t dissimilar:
This is good – for us. We can do it better (obviously)!
So alongside your normal searches (competition/volume etc) look at phrases where:
- Links alone are allowing sites to rank short term – the content is horrible – they won’t be there for long.
- The content is “quality” but quite possibly not exactly what the user wanted to find, leading them to make a completely different search – yours is the obvious replacement.
- The content includes a section “covering” the keyword – the Googler is much more likely to benefit from content where the keyword is the only focus, rather than part of an article.
So, you’ve found some opportunities. You’ve got some notes that are starting to build a profile of the Googler, the content type and what needs to be included. Processes used to find keywords, highlight opportunities and to then generate a content strategy are fairly similar…
You have to be better; it’s as simple as that. For a keyword strategy to be effective, you need to go above and beyond what’s out there. This won’t just help you gain positions; it allows you to keep them. Think of it like the guest posting boom. People jumped on the bandwagon and produced content that just about satisfied guidelines; in the masses.
Others became contributors on leading websites, because they saw valuebeyond the links; credibility by association and targeted traffic. 500 words minimum (assuming quality was constant)? No problem, they produced 2,000 their piece warranted it; to get a point across.
Those same contributors still have their credibility today, with their sites intact. The spammy guest posters? Working double time to get their de-indexed web properties back in order – submitting their reconsideration request because they got slapped for being naughty.
- Find posts ranking for your phrase and LSI terms.
- Determine which parts of the content offer substance (do those better).
- Highlight outdated snippets of information (some content will be old – explain why information is now irrelevant; see the keyword density section at the end for an example).
- Add sections you feel are missing (lack of relevant referencing, lack of detail).
- Decide which parts of the content on page 2 and beyond are holding that content back from going further (don’t make the same mistakes).
- Ask yourself lots of questions – profile the Googler – figure out if other results serve the intent and ask yourself why? Why? WHY?
- Cross reference this information with your content – have you gone the extra mile?
Extend the content with a funnel
So, you’ll notice here my keyword targeting strategy is pretty low ball; I’m not shooting for the stars.
With highly competitive phrases 1 in-depth article is not likely to be enough. Develop a content funnel. KISS Metrics, Moz and even Wikipedia (a colossal example) all do it; it works.
But what is a content funnel? Not a sales funnel, for a start (although if your clever, you can combine the two).
Kind of like a family tree. The broad term acts as the index holding page (these types of resources attract a large volume of links – if they are exceptional). You can then ensure you service the intent of the broad term, by servicing the intent of all the long-tail terms. I’ll do a post on this soon, but you get the idea…
1-5%? Forget density. Not completely, just the way most think of it; percentages. It’s not the 20th century.
The first content/SEO related result for the term I’m targeting (i.e. not Twitter/Facebook related) is over 3,000 words and the keyword appears 8 times including the title. That’s a density of 0.27%.
The #10 result is just under 2,000 words. It has 6 examples of an LSI phrase. That’s 0.3%. I think you get my point.
I actually go back and trim my articles after writing, to reduce density if it makes sense to do so. Naturally, if you’re writing to service an individual query, the phrase is going to appear more than once. If it looks unnatural and doesn’t help the flow of your content, remove it. If it adds value to a sentence, keep it.
Webinars On Demand
January 25, 2017While we’ve had access to the Internet since the early 1990s, we’ve only started to experience the full effects of its disruption on public rela...
December 07, 2016It's finally happened, social media has grown up and sold out. And it's awesome. For digital marketers, the maturing advertising options on soci...
Video is expected to account for three-quarters of all mobile traffic by 2020. But, creating powerful, effective video requires a significa...
Marketers are constantly seeking to engage with their buyers and drive actions that help buyers move rapidly through their customer lifecycl...