SEO 2019 - The Quickest Path to Ranking Success - Google RankBrain, User Behaviour & Satisfying Search Intent
It's not often I find the time to write enough blog posts for the Assertive website, let alone LinkedIn, but, today I made the time instead of finding excuses (we've all been there).
Given that SEO as a service or "offering" is like a hot potato right now with the shift in peoples shopping & buying behaviours, there has never been a time more important to rank online for your target keywords.
The online world has become awash with articles promising quick fix tips for SEO along with methods guaranteed to increase your rankings and even "SEO hacks" which have popped up all over the place (literally, I was finding them in Taboola ads on mainstream news websites). But, with millions of SEO guides, tips and tricks out there, what really works and what can you do to make 2019 the year of good rankings and online fortune?
Well, I am about to let you in on a little secret.
SEO in 2019 is changing.
Why?
Well, quite simply Google is changing the way people search, the way people interact with search results and also where people search. In fact, it's a well known fact that as Google's paid advertising platform (Google Ads) has become more expensive and for that reason, it has driven advertisers to use platforms like Facebook, Twitter, LinkedIn and Even Quora for ads - not that these platforms weren't being used before, it's just that their use has increased exponentially.
But what about businesses that just want to rank in Organic search?
Well, here's the thing.
GOOGLE Wants you to use GOOGLE, but, Organic isn't profitable for them so organic listings have sort of fallen by the way-side.
For many searches, Google now returns more ads (typically 4 above the fold) and, those ads now have all sorts of expanded text, additional links and contact details, pushing organic well below the fold, especially on laptop devices or devices with a rectangular aspect ratio.
We've seen a significant increase in the use of structured data above the fold (people also ask), knowledge box, map listings and even tools / images / Google shopping listings + more, all of which push organic results down.
Does that mean Organic isn't of value?
Well, no. Organic is still of significant value, the issue is, you literally need to rank more prominently on page 1 to stand any chance of hoovering up those impressions into clicks. Because for many niches, organic is below the fold, ranking in the top spots has become even more critical.
In fact, studies have shown that typical organic drop-off rates on average have dropped by as much as 30% as users are more drawn in to clicking on paid results, map listings + other in-search results.
So, How do I make my SEO strategy work in 2019?
If I could give website owners / marketing managers / CEO's any advice in regards to SEO, it would be to go back to the drawing board to think about your target audience and what they would expect to see if they were to use a website.
By this, I am talking about USER EXPERIENCE.
And yes, USER EXPERIENCE is literally what it says on the tin.
SEO in 2019 is about POSITIVE USER EXPERIENCE
Now, I know many SEO's will probably roll their eyes, and, going by data, many will advocate that Google does NOT collect this data or use it to evaluate, but our data says otherwise.
Before I go all-out explaining Google & Potential Behaviour Evaluation / Components - here are the things you MUST include as part of your SEO strategy >
- Make your landing pages fit your target audience as best as possible - literally make it your key goal to ensure that for your target keywords, your landing pages will satisfy a majority of the users who visit it - remember you MUST do this at DEVICE level - landing pages need to be optimised for desktop and mobile - a lot of the time, mobile variants should be more condensed and to the point.
- Satisfy user intent - your landing pages on your website should go as far as possible as to end the users journey - what I mean by this is that if a user lands on your website, you want them to exit your website WITHOUT returning back to Google.
- Focus on leveraging session duration / bounce rate reduction - niche, keyword and landing page dependent you need to realistically ask yourself how long a page should keep someone engaged and then look at the true data in Google analytics / Adobe analytics. If your page is an article with lots of content + it has a poor average time on page i.e. 30 seconds, then it is clear that the content needs sorting. If you have high bounce rates where you are expecting people to navigate through your website then that'll also need sorting - remember to look at everything from a relativity point of view. If your page SHOULD have better session duration + lower bounce rates then work to do that.
- Use A/B Testing as part of your SEO strategy. Platforms like Google Optimise allow you to run A/B tests. You can set objectives/goals to improve session duration / reduce bounce rates. Work to make your website landing pages better. Experiment with different above the fold layouts, content, media and find what works best cross-device.
- Improve your CTR, and, yes, before you say it, you need to get the rankings to run CTR experiments. You can begin this exercise with rankings on page 2, as long as you have SOME impressions to work with you can begin CTR optimisation. Remember to use Google Search Console, filter out brand (Exclude it) so that you can look at your true CTR. Look at CTR by keyword and page and optimise. Aim for 2-3% + on page 1 for non-brand keywords.
- Build more relevant internal links between content. Simple yes, often overlooked - massively! Build up your internal link profile by linking relevant pages! Interlink with consistency and balance. Try to link thinking about what would benefit your end users, don't drop in links to pages that don;t have good synergy with the landing page.
- Content Pruning, Refreshing & Consolidation. Get rid of old stale pages on your website, consolidate pages that might be similar to one another (reduce any potential for cannibalisation or the need for canonical tags). Consolidating similar pages is a great way to also create new fresh content pages & reduce overall site complexity and size. Content refreshes are also a great way to leverage a new lease of life to existing content, find those old articles talking about the good old days or summer deals in 2017, merge old pages and create new pages. Also keep in mind that website organic performance can also be hampered by stale content. If your website has a lot of old content that does NOT generate many clicks or impressions in search console then get rid of them. If you have articles generating a handful of clicks every few months then chances are they have little to no value.
- Improve site loading speed / page speed. Now, before you clench your teeth or frown, site speed is NOT particularly weighted but it does have some clout. Typically weighted below 5% (but climbing) page speed is something that more people need to address with their websites and landing pages.
- Improve rendering performance, now, this one is frustrating, but, Google wants your page to load as quickly as possible, painting above the fold content almost instantly. Now Google PageSpeed and Google Lighthouse have merged into one tool, your scores will be marked down if you do not address loading speed and rendering performance. This means deferring those pesky CSS & JS files. For those of you running Wordpress, we feel your pain, but, it is doable with custom development!
- Focus on RELEVANT LINKS - yes, link building is still here and is not likely to be going anywhere fast. Focus on quality over quantity. Build links that have relevant synergy with your landing page and business. Ideally, your link profile needs to be spread out across your website, linking back with a mix of non-optimised anchor texts and title orientated links. Getting a link back to your content using an anchor text the same as your title is really effective. Don't waste time with link building short-cuts, by this I mean buying links from those "guest post specialists", half of it is recycled crap that'll likely offer you no ranking equity.
- Get earning citation - Become an authority in your niche, post on Quora, Twitter, Facebook and LinkedIn. Join groups and share your expertise. Become an authority and get people to know who you are and what your brand is. Earning citations is a great way to build brand trust.
- Use Google My Business - it's a Google platform that Google wants you to use. Get posting! Google+ might be about to disappear into the ether, but, Google my business is here to stay. Get posting on it! Not only will your content appear for brand searches, it increases quality factors associated with your business/domain.
Now, on to more complex things >
So before I go into detail, let me ask you a question of which I'll follow up with the answer.
What does Google want to serve to people who use its search engine?
The answer, unsurprisingly is.......... (Drum roll please).....
GOOD QUALITY RESULTS
Now that we've got that out of the way, let's ask another question.
What is a good quality result?
Well, easy, it's a website that satisfies the users that land on it. What we mean by satisfy is "served its purpose" - you might of heard the term "search intent"?, well, this is it.
A good quality result is a website returned in search results that does what it says on the tin, whether it's offering a holiday cottage in Cornwall or a bank account in Hong Kong, as long as the end user finishes their journey that is what is key here.
Another way of looking at it is, if Google consistently returned poor quality results then, as a search engine it would begin to lose market-share (we all know what happened to Yahoo, Ask Jeeves, Lycos and other search engines - yup, they pretty much lost their marketshare to Google).
So how does Google know is something is a good quality result?
OK, so, I could really tear this logic open, but, you'd be reading the technicalities for the next hour, and, this article would probably become very boring and too long to read, plus, I am running on limited time here, my coffee cup is almost empty.
This question is a hot topic among SEOs and, there are various answers and opinions across SEO communities.
So, here's the thing, A good quality result is a result that satisfies user intent - in that, the user has used the website/landing page, got what they wanted and finished their journey, so how could Google use this data?
Here goes:
Initial CTR for Search Result Listing
So firstly, on Google's side it can see for all the times your listing is returned organically how many people actually click it (CTR - Click through rate). This is the first quality indicator (subject to you ranking on page 1). If Google can see a good CTR it tells Google that your title and description are engaging, that it's attractive to end users.
Landing page behavioural aspect 1 - Average Time On Page to First Interaction
So, a user has clicked on your result and has landed on your page. So, what happens next? Well, this all comes down to your niche, your keyword, your landing page, the device the user is using, their location, age etc.
When a user is on your website, what can they do? what do they see above the fold? how do they engage with the page and what happens next?
Google may be looking collectively at how people engage with a page by where they scroll, click and how long they spend on the landing page. Google may also be looking at whether or not users go on to a first interaction (browse another page on the website) or whether they abandon the session.
Landing page behavioural aspect 2 - Bounce rate / Bounce to location
Assuming that a user lands on your page and decides to leave without browsing any other pages - Google will look at the bounce property i.e.
> Does the user bounce back to Google search results? if yes, it will look at this behaviour consistently among other organic results
> Does the user bounce away from the website (not returning to search results) if yes, it might be that the user got what they wanted and didn't feel the need to continue browsing - again, Google will look at this behavior.
This is where bounce rate becomes subjective - a high bounce rate is NOT an indicator of a poor quality landing page if that page satisfies and ends the user journey (most of the time). A high bounce rate might be a poor quality indicator if the user returns to Google search after a short session duration.
Landing page behavioural aspect 3 - Bounce to location + Average time on page
So this can be typically factored in if the bounce to location takes users back to search results. Google may have to ascertain why by looking at average time on page (bear in mind this process is split between DESKTOP and MOBILE devices).
> If a landing page has a high bounce rate + short average time on page then where does the user go? if they go back to Google, then, Google can look at this behaviour through results 1, 2, 3 etc. looking for a pattern. If your page performs significantly worse than surrounding results it may be an indicator of poor landing page quality
> If a landing page has a high bounce rate + short average time on page and the user abandons the website (not returning to Google) then Google may have to take into consideration other factors such as surrounding result average time on page + abandonment behaviour.
Landing page behavioural aspect 4 - Average time on page + Total Session duration
Now, I was having a chat with another highly experienced SEO in my field "Joe Williams", and, we discussed the possibility of Google looking at average time on page in combination with total session duration. If Google could see that a page was composed of a lot of content (text/media/images) but had a poor average time on page that it could be an indicator of poor quality/lack of relevance.
What was also discussed is looking at average time on page in combination with total session duration, why? well,
> If a total session duration was made up of lots of page views in a website where those page views had a very short average time on page, it could indicate that a user is trying to find something in a website and not finding it, or, that the landing pages just simply are not working.
If Google could see that on average, people visited 5 pages in their session and the total session duration is say 50 seconds, that would average out at 10 seconds per page - hardly engaging right?
> If a total session duration was made up of lots of page views where 1 page performs well and the others do not, it could be an indicator that Google is not returning the most relevant result for the end user (again, this depends on the landing page from organic performing poorly whilst another internal page does much better) - this has to tie in with traditional SEO factors such as content, content quality, keyword focus, internal anchors etc.
Landing page behavioural aspect 5 - Exit rate + location
Say users coming in through a website show a trend where they exit, this could be an indicator of quality, why? well, if, say for example you run a holiday cottage agency and you have a lot of incoming traffic through your Cornwall cottages page, Google might be able to piece together search intent satisfaction by looking at >
> Are the typical exit pages on the website part of an analytics goal? i.e. booking complete? If Google sees people exiting the website through a funnel it could be that the website is satisfying the users search intent. If a user books a cottage, why would they return to Google? it;s unlikely they would, so if Google can see a high website exit rate at say a booking confirmation page (which could be a Google Analytics goal) then that alone could be a strong indicator for Google.
> What proportion of visitors follow the same exit path?
Exit rates on pages paint how a page performs in relation to other pages on the website. Funneling exit rates away from pages which should clearly retain and engage users is a great way to leverage your SEO strategy. You must keep in mind that landing page performance all comes down to search intent + landing page type.
Landing page behavioural aspect 6 - Scroll rate depth
I had to throw this in, controversial yes, topical yes, verifiable, not yet, but, hear me out. Remember, Google doesn't need to have data from EVERYONE, it can use partial data collection also. It might even be that additional data could influence ranking benefits, but, I need to be careful with what I say.
So, scroll rate depth, Google Tag Manager allows website owners to deploy a trigger to capture scroll rate depth (See google tag manager adds native scroll depth trigger) now, the fact that this can be added means that Google can collect this data, and, looking at how Google does it - we know that traditional GA UA tracking implementation allows for Google to be able to get this data regardless of tag implementation.
So why does this data matter?
Well, again, it's a behavioural component. If Google can ascertain how far people are scrolling, it could, potentially combine this data with other behavioural data such as average time on page / session duration as well as segmented device performance.
Not just scrolling, click logic is also possible.
The more data Google gets on how people browse, click scroll and their overall user journeys, it can begin to see the bigger picture away from traditional SEO (links etc.).
Pages that contain a comprehensive amount of information, let;s say an article or long-form blog - now, if that content/page has poor session duration + poor scroll performance, that too could spell quality issues / lack of relevancy.
Landing page behavioural aspect 7 - Events
Now, another controversial point, but, I believe entirely feasible for Google to use this data. EVENTS. Google analytics (traditional tracking) and GTM can track ALL click events. Whilst this data may only be viewable AFTER you implement triggers, the underlying data is still generated.
Events can include where and what you click, forms you may trigger and even the rendering/playback of media.
Events are actually a fantastic way for Google to understand what a user is doing on a page and how they engage with a page.
This data could allow Google to interpret more about how users interact with a page from playing a video to using a tool such as an in page calculator.
It's important to understand that Google is a search engine and that search engine relies on an algorithm. That algorithm is completely variable down to keyword level, niche and average position.
Humans by default interpret content independently, so what I might like in a landing page might not be what other people like. If humans do not have a unified view of content / landing pages then how would a computerised algorithm from Google know what to return?
Well, we could rely on traditional elements such as link building / domain authority (based on an old school voting principle / popularity principle) or we could rely on keywords in content, keyword association, keyword semantics etc. BUT, this is 2019. Google has access to tens of billions of websites and billions of devices, so why wouldn't it use big data as part of its algorithm?
Well, it does. RankBrain is a fundamental development for Google. AI is already acting upon search results, and, as time goes on, user behaviour data will fuel AI decisions. If Google can collectively see sites that engage, retain and serve their purpose, maybe it'll forget about backlinks.
Whilst some traditional SEO elements can still be beneficial in 2019, it;s worth following the trend/direction that Google is heading in to make sure your SEO strategy not only works in 2019, but is also futureproof!
Written by Daniel Carter of the Assertive Agency
Interim Programme Manager | Digital Strategy & Delivery | Leading Business Change & Transformation | Consultant | Project Director | CEng CITP MIIM MIoD MBCS MCIM MCIET #ono
4yHi Daniel Carter . Interesting article on a complex subject. So similar advice for 2020 or has it all changed again? #seo #digitalmarketing
Proofreader | Author | Get the respect your work deserves!
5yI can't use Google My Business because you have to pick a category and they don't have any categories that even remotely resemble what I do. I'm a proofreader and there's no proofreader or editor category.