When I’m not listening to power ballads or watching Ghostbusters for the 100th time, I think I’m getting a little cynical in my old age.
I’m sure having worked in SEO/digital for so long is in absolutely no way a contributing factor here… but it may crop up on occasion.
There’s a great SEO community out there which share lots of practical tips and ideas every day, but it can be the broader industry news that can cause some headaches.
Take some of the big SEO studies (top ranking factors revealed etc.). Sure, there’s sometimes an underlying product/tool promotion but that’s ok (it’s a fantastic selling point if it was key to getting that data or arriving at that insight) – it’s that some of the recommendations or bigger statements made can lean heavily on the implication that correlation = causation. It’s our job to analyse these and translate into what it means for our clients in their specific situation while avoiding knee-jerk reactions from headlines.
It’s also harder to believe Google isn’t trying to squeeze as much paid revenue as possible from a search result after each update. It is not a positive character trait, I admit, but I’m working on it!
Having said all that, I figured I should call out something good when I see it. Balance!
Sistrix released a great study last week on Click Through Rate (CTR) data, having analysed over 80 million keywords and associated mobile search results.
The methodology, the data used and the necessary caveats are clear as they make some useful observations having a large set of data available.
Do go check it out for yourself, but some of the key takeaways were:
- CTR varies a lot between SERP elements. CTR for 1st position ranging from 13.7% to 46.9% depending on the layout
- Googles Ads presence drops 1st position CTR by 10 percentage points, on average
- Featured Snippets knowledge panel presence reduces organic clicks (as the study was on mobile results, knowledge panels had a bigger impact)
- Shopping ad presence reduces the organic CTR (1st position CTR seen as low as 13.7%).
- Niche/tail queries tend to remain mostly 100% organic, while the bigger searches have more variation in the search layout.
Some of that might not come as a huge surprise, but nice to see it validated across such large data sets.
OK, now we know. What can you do with that information?
At the very least, this information adds a few considerations across your SEO workflow, but here are some areas you can take practical steps.
Tag up/categorise those keywords!
I tend to go a bit overboard with tagging up my keyword set because I like having the flexibility to analyse the data in different ways. Now I have another excuse for it.
Your “types” of tags could expand to include:
- Your usual business topic or product/service categories
- Audience segments
- SERP layout
- Keyword intent/context
With any of the well-known ranking tools you can work backwards by running checks on your keywords and then filtering by SERP appearance and tag up accordingly from there.
If you’re at the start of an SEO engagement, it’s important to scope your keyword research appropriately. Think about how you’ll be using this list. If you’re going to go big, then the segmenting of the list becomes even more important.
Build your own CTR curve(s)
The study shows a general pattern across a large set of data, but each site will be unique – the sector you operate in, your goals, the competition, the quality of your meta data (and your products, services and reputation of course)
So, start building your own CTR curve. Here’s a good place to get started. You just need Search Console data to get going.
If you want to go all-in, you may wish to build several of them based on different scenarios such as
- Mobile and desktop results
- SERP layouts
- Keyword & audience intent
Consider CTR differences when forecasting
SEO Forecasting pros and cons (or “cons and cons” depending on your viewpoint) is an entire debate in itself which I’m going to sidestep completely here.
Time spent vs. “accuracy” or robustness of a forecast is a consideration here – as with all digital planning, you just need enough reliable data available to help set the direction of travel and weigh up where and how you direct your efforts.
You may already choose to apply different conversion rates by content or product sections of the site; applying the new CTR curves would complement that.
At the very least, if you’re unable to account for it, add CTR differences to the list of data caveats and considerations that would typically accompany a forecast (alongside assuming similar seasonality patterns, timely implementation of recommendations etc.).
Know your SERP layout, the audience intent and optimise for it.
Assuming you’re now working with a representative portfolio you’re happy with and you’ve identified where you need to focus, it’s time to get to work.
Whether it’s targeting a featured snippet or other layout, you can start planning your necessary technical (i.e. do you have the necessary mark-up in place) and content updates (is it meeting the needs of the query/user?).
Things are constantly changing here, so keep an eye out for industry updates – for example, featured snippets saw an update last month.
Paid Media & SEO – keep testing and pick your battles
This study commented on ad presence and that shopping CTR is an eye-opener, but it doesn’t consider all elements such as
- Branded keywords
- Paid and organic presence in the same SERPs
Should you stop bidding for topics you already rank well for? Should you put less “effort” into areas that are dominated by paid results?
Well, I’m going to sidestep this one too and say the classics “it depends” and “test and learn”! (I’m a chicken as well as a cynic).
Our Paid Media team are a very clever (and, as I’m often told, award-winning) bunch. The type of data available is always an eye-opener (especially given the limited nature of organic data available from Google). If you have the opportunity to test scenarios out, do it!
On a less CTR-related and more keyword portfolio management note, I’m particularly interested in the opportunities for the iterative optimisation Dynamic Search Ads and associated organic landing pages.
Hey, maybe I’m mellowing after all? I hope these tips provide a good starting point but if I’ve omitted a point that’s important to you or if you have any questions, get in touch!