Featured Snippets Fluctuating Prevalence
As part of our efforts to stay up-to-date with all the developments in search, we keep a close eye on not only our client’s rankings and their competitor’s, but also on anything Google adds, removes or changes on its organic search results pages. In this blog, we will be specifically talking about featured snippets: the coveted position “0” of organic search results.
Search engine results pages
Search engine results pages today work entirely different to what they were 10 years ago, hell, even three or four years ago – they are constantly evolving. Google’s main goal – which their entire business model is built upon – is the endeavour to provide the best and most relevant results to users of its search engine as quickly as possible. That’s what a (good) search engine should do.
As part of the natural evolution of the universal search engine, Google looked at their search results pages and saw that they had all of this abundant whitespace: an ever-increasing aspect to their SERPs, which increases in proportion to the up curve in the natural progression of increasing screen sizes over time. Google realised they could use this whitespace to transmit information directly via their search results pages without the user ever needing to click through to a website.
Once this idea was realised, it was only a matter of time until Google began to augment their search results pages with a whole host of ‘features’; rich snippets, carousels, Google My Business listings, the knowledge graph – and the featured snippet:
The industry hasn’t looked back since: search engine results pages have never been better. These augmentations help Google fulfil their goal to provide the most relevant information as fast as possible to its users and provide another area for us SEO’s to strategise and optimise.
However, featured snippets present themselves as another challenge that SEOs need to account for. If a competitor has a featured snippet, then they are effectively taking away a substantial amount of valuable clicks from everyone else included on the first page of that search result. In a study conducted by Ahrefs in 2017, it was found that, when a featured snippet is present on a search result page, the percentage of clicks to the first result drops from 26% to under 20%.
In early 2017 we noticed that for a particular client of ours, the percentage of keywords that returned a featured snippet in which we were tracking for them was quickly increasing.
The following screenshots are from Moz – one of the tools we use to track featured snippet usage.
At the start of the year, the percentage of keywords we tracked that returned a featured snippet in the search results stood at around 17% (69 out of 410 keywords).
This figure rose slightly at the start of March and plateaued until the end of August. As we can see, from August 24th to November 2nd, the usage of featured snippets increases substantially: from 25.37% to 49.76%, to be exact.
It is worth mentioning that the changes we had seen up until this point where enough to considerably change the organic results that our client and their competitors were experiencing, but the changes to the SERPs were yet to let-up.
Within a week, featured snippet usage for these 410 tracked keywords jumped from 43.66% to 59.02%, and within two weeks after this, featured snippet peaked at an astonishing 67.37%.
Now, at this point, you might be thinking why this is significant enough for us to write a blog about. In itself, the fact that the usage of featured snippets has increased so massively over the last year is, I think, in itself something to discuss.
However, another layer of intrigue exists to this, whereby the validity of these featured snippets is called-into-question.
How are featured snippets ‘won’
Traditionally, featured snippets are designed to answer a user’s question quickly. For example, “how far away is the sun?” is a question with only one answer. It is quite unarguable that Google’s search results page should generate a featured snippet for this query, immediately providing the answer of “93 million miles” without additional clicks needed to be added to the user journey.
The keywords we are tracking for this client are very competitive – and none of which are question-based. For any one of these keywords, there are at least 10 websites who are competing for these in the SERPs; all of which are strong, authoritative websites with great content. For Google to choose one of these websites over another is a little biased, to say the least. A featured snippet is somewhat of a stamp of authority: Google is promoting your content to the top of its SERPs, and puts it into a little box because it is (or at least should be) better than anything else available on the web.
When Google promotes one company via a featured snippet for these types of keywords – “purchase” keywords – the technology should be questioned, I think.
This may be sour grapes on our part, but it unfairly promotes the lucky website over the other nine websites in the search results pages.
But, the algorithm which is making the decisions on featured snippets is just that: an algorithm – a machine. It is a complex beast, programmed by humans, and it will inevitably make mistakes. So this begs the question: is the desired outcome for these featured snippets to appear for both question-based queries and “purchase”-intent keywords alike, or is this outcome an unforeseen mistake?
So, going back to the results we discussed earlier – it was alarming to see that Google rolled out featured snippets to over two-thirds of the 410 keywords that we were tracking.
We tried everything to optimise our client’s site so that they could steal some of this digital real-estate; we altered their header tags, we tested different page structures, we added <li> tags to various elements, added content, removed content and more. Alas, none of this really worked.
Ultimately, we were led to believe that we simply could not force Google’s hand in giving our client these valuable featured snippets. There was no rhyme or reason behind the featured snippets which their competitors had won. So, we had to postpone this activity as there were other important things which we needed to focus on.
However, in February 2018, we saw a huge change in the search engine results pages. Over three weeks, featured snippet usage had dropped from the 67% on February 1st, to around 29% – recorded on February 22nd.
Since then, featured snippets usage has risen very slightly to 34% – which is still practically half the rate of which they were in use just a handful of months beforehand.
I’m actually pretty happy with this: from the beginning, I believed these featured snippet results to be misplaced and unfairly (and blindly) biased to the select few competitors who, seemingly by a stroke of luck, managed to get themselves these a coveted position “0” rank.
What’s next for featured snippets?
As I said before, we keep a close eye on as many search engine developments as we can, including everything around featured snippets. Recently, I have scoured the internet to find anything, even a murmur, around the change in usage of featured snippets which another business may have experienced over the past few months, and I’ve come up short.
The only thing I could find relevant to this was a small discussion about their volatility which was taking place mid-November, 2017. As documented here, Gary Illyes, a Google employee, discusses featured snippets and the fact that they are under constant redevelopment:
Featured snippets, well the feature, is under active development and I think it’s a very volatile feature right now because it’s under active development, people are coming up with new ideas on how to improve the featured snippets, the featured snippets algorithms, so even if you get the featured snippet today, you might lose it in 24 hours because we changed something slightly in how we want to trigger featured snippets or what conditions a result has to satisfy to get the featured snippet or to be featured.
Due to the constant improvements Google makes to its algorithms, it would seem then that, for whatever reason, the algorithm responsible for adding featured snippets to the search results no longer deems many of the 410 keywords we track for this client applicable for featured snippets as it did a few months ago.
If this is true, I think this is great news, because we don’t have to worry about unfairly slanted featured snippets which are returned for “purchase” keywords aimed at those at the bottom of the marketing funnel, as opposed to question-based keywords or informationally-driven queries which rightfully deserve a featured snippet.
We will continue to monitor the frequency of which featured snippets are returned, and indeed the topic of featured snippets in general, as it is almost guaranteed that their prevalence will continue to change long into the future.
Stay tuned for updates and check out our ultimate guide for search in 2018 for more information.