Why SEO checklists fail: A takeaway from BrightonSEO
Last Friday a few of the CandidSky team took a trip to (not so sunny) Brighton to take part in BrightonSEO – the UK’s largest SEO conference – to listen to a host of presentations ranging from technical SEO to content, site speed, usability and psychology, and we weren’t disappointed.
We attended some great talks, but there was one in particular that resonated with me; “Ranking Factors Reloaded – Why Content Is The Key To Success” by Marcus Tober from Search Metrics. The reason why it struck a note with me in particular is that it involved some insightful research data which lined up with my own perception of how SEO has been changing over the past few years. In this article I’m going explain how the industry is changing and what we can do to make the most of it.
The checklist approach
All too often SEO is seen as a checklist exercise where a site needs to meet a number of absolute criteria, and the growing number of SEO tools only add to this perception.
- Keyword in title Check!
- Keyword in heading Check!
- Over 1,000 words Check!
- More links/ link quality pointing to my page than my competitors Check!
Now why am I ranking on the fourth page!
It’s awfully convenient to think SEO is that simple.
What Marcus’ presentation showed most elegantly was two things:
- That there are no absolutes that allow a checklist approach to be effective
- That every decision need to be made against the context of your users, industry and competition
So lets look at the above checklist. More specifically item number 4 “More links pointing to my page than competitors”. Over the course of the presentation, Marcus pulled up research data from an analysis of rankings in 3 verticals, and the results were particularly interesting.
For the ecommerce sector, the Search Metrics team found that there was a strong correlation between referring domains (links) and rankings – as we would expect using a classic SEO checklist.
Here you can see the number of referring domains on the Y axis and average rank across the X axis.
Next came the surprise. In the healthcare market there was NO correlation between the number of referring domains and average ranking.
Finally, there was a NEGATIVE correlation between the number of referring domains and average rank in the Finance industry.
Marcus went on to run through a number of other metrics from the “classic” SEO checklist. And the results were the same. In some sectors (particularly ecommerce) classic SEO metrics correlated with rankings, but in other, more informational sectors they didn’t apply at all.
Why we shouldn’t rely on traditional metrics
Simply put, some pages provide a better result for users than others, irrespective of the “classic” SEO ranking factors such as keyword targeting and links. It all comes down to satisfying user intent. This is more pronounced in sectors where people are looking for information rather than products or services, which is why we see such a stark difference between health and finance (informational) and ecommerce (transactional).
I wrote recently about two new usability measurements confirmed as part of Google’s algorithm; clickthrough rate and dwell time. These are the metrics search engines can use to measure content engagement and how well user intent is being satisfied, but how do we improve them to get ahead of competition and bypass traditional ranking metrics? Focus on the user.
Over the course of his presentation Marcus pulled up an example of two search results for the query “Costume Ideas”. One result (below left) fell behind with traditional ranking metrics, whilst the second (below right) would appear to be a much more competitive result based on its keyword targeting and inbound links.
In this case, the first result ranked 1st, whilst the second ranked 14th. Traditional SEO can’t account for this, but focusing on the user can. The first result contained a 60-image gallery, links to other relevant content and high social engagement. It provided inspiration for someone clearly searching for it when they typed in “Costume Ideas”. The second result contained a mere 14 costume selection, and is quite clearly aiming to sell a product rather than provide inspiration.
It’s true that the difference here will be measured in metrics like dwell time and clickthrough rate, what marketers need to focus on is offering best result for the user intent in question by understanding the desire behind it.
Rather than subscribing to a classic SEO checklist, we need to
- Optimise for users, not search engines
- Think topics and answers, not adgroups and keywords
- Benchmark against your industry, not SEO “norms”
SEO isn’t complicated any more – It’s complex
“What’s the difference?” I hear you ask. A complicated process like sending a rocket to the moon is difficult, but once you know the series of steps which need to be taken to reach your goal you can repeat them precisely, much like an SEO checklist. A complex process, on the other hand, will vary each and every time. A great example of this is building a skyscraper; it depends where you’re building it, what regulations you need to meet, the plot you’re building on and a whole host of other factors which will vary each time you build a new one. As you can imagine, a complex task requires high levels of skill and awareness, and the classic checklist approach – although successful sometimes – simply can’t guarantee the same result every time.
Credit to Marcus and the team at Search Metrics for bringing this to the fore. You can access the full presentation from BrightonSEO here.