Onpage Optimisation Changes

[nezona]

Hi Guys,

I love the SEO world I really do, but sometimes it can be quite confusing and even after 11 years and a few clients under my belt, I still have head-scratching days and this week has been one of them.

It seem the rules surrounding onpage optimisation of keywords have changed quite a lot this year. Whilst, I understand blatantly sticking to a 3% keyword density rate for your keywords, hasn't been good practice for a while, and with RankBrain and machine learning, we have to pay attention to semantic words and phrases but it seems there is a new set of rules I haven't learnt yet.

For example I had a client I was working on and we we noticed although they were ranking quite high for a keyword phrase, it wasn't actually mentioned in the text at all and so by adding it in a place it made sense, we should lift this and other keywords. Here is what happened, within a week their main keyword moved down from 1 to about 6 and the keyword that wasn't added moved from 4th to 23rd. After scratching my head and then going to full panic mode, I calmed down and looked at competitors, they didn't mention the word in the content either and so I decided to remove the one word we added to the text. The rankings came back overnight (well after doing a fetch as Google and getting to reindex). So if keyword density now is clearly NOT a metric to go on, how do we know the sweet spot? Do we use something like Ryte and make sure we using semantics and keywords within the average of the top ten? Does what Google deems important depend on the niche?

Not a right or wrong answer here, just interested in your thoughts

Regards

Neil

[nezona]

Hi All,

I will just give you all a quick update. I am more confused then when I first started!! I am telling my team from now if it's on the first page already we don't touch the keyword. Of course other keywords may also be trying to rank for the same page on page 2, so it's going to be a tricky one,

I have just done a quick experiment on a ecommerce page the keyword was on page 6, the changes I did yesterday made it go to page 7, further tweaks made it go to page 9!! This is mental! I am using Ryte to make sure semantics are mentioned, maybe using this tool and getting the averages that each keyword and related keyword is mentioned, is not the best idea. But without a guide how do we know what to optimise for?

I will give it a week as Nigel above said and see what happens. At least I know resetting to their original values before we started for any keywords that have declined will work.

Regards

Neil

[NickW816]

No problem Neil, on-site SEO often makes me feel the same! It's easy to go from "following best SEO practices" to "over-optimization", and constant testing, checking results, and striving to always be learning what works today is essential. Best of success!

[Nigel_Carr]

Hi Neil

That's really interesting and I have seen similar things happen. Google is determining user intent from each keyword and phrase typed into search, so the results will vary considerably with each semantic search. There is also a possible a stuffing issue, not just between identical keywords but semantics that we may need to consider. If adding that single word overplayed the whole piece, then the probability is that you were trying just that bit too hard. The whole thing has become very flaky and I have seen wild movements up and down which then appear to find a level after a week or even two.

For some clients I have seen wild movements for pages I haven't even touched - literally from position 5 to 83 and then back again within a week. Whatever Google is doing is unsettling - wrong even. It's hard enough for experienced SEOs but if you are a website owner with no experience of SEO it must be rather disheartening to experience the madness and start scratching your head looking for an explanation when there really isn't one.

[TucsonAZWebDesign]

Wow, so what you're suggesting is not only does keyword density seem to be applicable, but that it varies depending upon the specific keyword?

Wow! If that indeed turns out to be true after the conclusion of your further analysis, that's sounding to potentially be a headache for the SEO world when it comes to onsite lol!

[nezona]

Hi Nicholas,

Thank you for making me feel like I’m not going completely mad! I’m running an experiment now on a client site looking at a mixture of average keyword density of top ten for a keyword + searcher intent using sematic words that are in common with the top ten.

Remeber the days, you did the onpage month 1 and then built 300 ba each month! No skill, but easier times!! Haha!!

Regards

Neil

[NickW816]

Very interesting situation Neil, it sounds like you did the right thing with looking at what the competition was doing, and then just making one change at a time, while using the Fetch & Render tool in GSC to see results faster.

While keyword density is a factor, Google is constantly trying to figure out the "searcher intent", and it is possible that adding this keyword in that seemed like a no-brainer changed Google's perspective of the page's "searcher intent". Just spitballing here of course, as there a ton of potential on-site factors at play. It sounds like you are on the right mindset though and I wouldn't totally throw keyword density out, just focus more on "content depth" and helping solve the searcher intent.