Keyword Golden Ratio = All In Title / Search Volume

Ryuzaki

お前はもう死んでいる
Moderator
BuSo Pro
Digital Strategist
Joined
Sep 3, 2014
Messages
6,126
Likes
12,741
Degree
9
I'm going to think out loud here:

I run around all of the SEO blogs out there desperate for high-level information and entertainment regarding this topic, and rarely get it.

But a couple of months ago I landed on some young guy's blog (I can't even recall who it was) who was doing nothing but case studies that weren't going so hot. The reason I remember is he kept harping about "KGR" over and over in his updates but never mentioned what it was.

Well low and behold, today I'm looking at Jon Dykstra's blog, https://fatstacksblog.com , which is easily one of the best SEO blog out there these days in terms of motivation and being a clever marketer, and he off-hand mentioned Keyword Golden Ratio, and I immediately realized "KGR!!!"

Turns out it's a simple concept:

Keyword Golden Ratio
All in Title / Search Volume = Keyword Golden Ratio
So here's the video, by this guy Doug Cunningham (here is his site).


With that said, if you don't want to watch the video, which is long-winded, here's the gist of the entire simple concept:

LJzJXxF.png


The guy is suggesting that you pull a ton of keywords, ones that make sense in a phrase, not stuff like "running app tracker iphone best," and pulling their All in Title count and dividing it by Search Volume. This leaves you with a what is hopefully a fraction.

The lower the keyword golden ratio is, the better. This means that there are relatively fewer posts attempting to rank for that term purposefully, and thus it's likely less competitive.​

So the guy doesn't say this, but I'm assuming that this works based on two of the strongest on-page signals you can send:
  • keyword in title tag
  • keyword in H1 tag
Of course you can manage both separately but I'd venture to say 99.9% of websites that are content based don't.

In the picture you can see that he managed to find a decent keyword: "what do i need to brew beer." This keyword only gets 90 searches a month, and only 20 sites in Google's public facing index contain that phrase in the title, giving it a 0.222 keyword golden ratio value. Most of us likely have sites strong enough that publishing an article with that as the title would guarantee a top 3 spot immediately, if not #1.

Of course this KGR metric doesn't take into account actual competition, just a sliding scale chance that it's not competitive. Those 20 posts with that phrase in the title could all be monsters with a million backlinks each.

But you could check the top 5 or top 10 quickly with something like Market Samurai, which I'll do right now...

YnksOCV.png


#9 says it has it in the title but it's not an exact match, it's a broad match. Looking at the referring domains to the page, I'd venture to say that if you actually targeted this phrase, which none of them are doing, #2 is yours, if not #1 eventually (probably not worth it).

But what is likely worth it is ranking for a ton of one-time searches that have no volume reported. Long-tails and similar searches can probably be picked up and dominated by the sole fact of targeting the parent keyword. And this is just above being a commercial keyword. It's ripe for dropping some cookies for sure.

This Doug guy is doing this with some fairly low volume keywords, where it's the safest to make assumptions. I don't think this golden ratio is going to scale into higher volume very well. He uses his cut off point at around 0.25 KGR. If we went for higher volume keywords, I suspect we need to push that down to 0.10 or ever lower.

Anyways, if your site has a blog, this is likely a way to come up with a TON of content ideas to pass off to your writers. You could mass export keywords, calculate the KGR, filter for values equal to or less than 0.25 KGR, then delete rows that have non-sensical keyword phrases. Then just pass those off to the writers. You could blow a few hours on the keyword mining and likely boost your traffic by a decent amount pretty quickly once they're all published. For CPM and CPC info sites, this is straight cash money that will pay itself off eventually, probably within a year, I don't know. Depends on how much you pay your writers.
_______________

Anyways, I'm not saying we should all jump at concepts like these, but it was interesting enough to share. It's certainly nothing new, but giving it a jazzy name like Backlinko does means this guy will end up getting links and traffic for the term, which is a great example of marketing.

But as far as applying the concept itself, sites that have covered everything under the sun on their niches may be able to mine out some extra traffic without moving on to clickbait nonsense like most end up doing.

Have you guys ever employed something like this?

I know my site is ready to dominate searches like these and have been talking about it in my case study. I think @Kevin is exploiting this tactic, maybe not with the Keyword Golden Ratio. How are you measuring it, Kevin?

I'll definitely move to this approach one day, along with sniping out keywords from lesser sites after they do all the work for me. Looking at terms that forums rank for is a piece of cake for this type of attack too. I yearn for the day when I'm begging for topics to cover instead of being flooded and drowning.
 
I think this is the kind of thing you throw in the mix when you can't decide between a set of target keywords. The KGR will end up being the tie breaker, so to speak.

I'm not aware of a tool that will spit out bulk-data for the "allintitle" operator. If there's something out there that does that, it would make this more viable as a keyword research metric.
 
TBH I was reading through, expecting you to "poo poo" Doug's KGR, which I have been concentrating on using recently to try to boost traffic to my site (despite not yet generating much content, so I can't comment on it's effectiveness). However, it does make sense and if you stick to <250 local monthly searches (as he recommends) then I am sure it will work - certainly his case study of adding KGR content in bulk to a site seemed to help him boost his earnings.

I know that you can use Scrapebox to get allintitle: results, but you need a handful of private proxies to use too.
 
I heard about this a while ago, and I have started posting some KGR articles in the last couple of months.

These are all keywords that are KGR compliant -

bY7L8m9.png


As you can see it's been kind of a hit and miss. However, the keyword that is ranking at 3 also has a short tail keyword with a 5400 search volume, it is a buying keyword, and that keyword is ranking at 11. All of these articles are less than 2 months old and no backlinks have been built to any of them.

I think if you were to do this at scale you could uncover some real gems, but doing it to scale is very time-consuming, hence why I have only done about 10 in the last 2 months.
 
I don't think I do this. Most of what I do is pick a basket of keywords that is "led" by a captain (the one I'd prefer to rank for over all else). What I notice is that I rank for longtails immediately, those pull traffic and give signals for mid tails, and then eventually I get the 'head term' (not a true head term, but in the context of my topical bucket, it is).
 
I don't think I do this. Most of what I do is pick a basket of keywords that is "led" by a captain (the one I'd prefer to rank for over all else). What I notice is that I rank for longtails immediately, those pull traffic and give signals for mid tails, and then eventually I get the 'head term' (not a true head term, but in the context of my topical bucket, it is).

This is how I do it too and it's one of the reasons pbn's are "dead". It's harder to move individual keywords now than it is to get organic traffic around a topic. Yeah, I totally just changed the subject. Sorry.
 
I do something pretty similar but use the intitle:”” inanchor:”” search rather than just intitle. I will build out a similar spreadsheet when I get some time to do the search volume but though to try it out.
 
So after reading this, I decided to implement the process for my next potential site. I started with 24314 keywords and ended with 300 definites and 2141 highly likely keywords.

Rather than just throwing a bunch of cash at the content for the new site I have decided to order three test articles with one per keyword. I received the first one last night and published it, the next two should be this week. I will post and update on how the articles do.
 
I heard about this a while ago, and I have started posting some KGR articles in the last couple of months.

These are all keywords that are KGR compliant -

bY7L8m9.png


As you can see it's been kind of a hit and miss. However, the keyword that is ranking at 3 also has a short tail keyword with a 5400 search volume, it is a buying keyword, and that keyword is ranking at 11. All of these articles are less than 2 months old and no backlinks have been built to any of them.

I think if you were to do this at scale you could uncover some real gems, but doing it to scale is very time-consuming, hence why I have only done about 10 in the last 2 months.

This one is Serplab
 
You guys who are trying this, are you using broad match or phrase match when using allintitle?

For example:
broad - allintitle:keyword golden ratio​
vs.
phrase - allintitle:"keyword golden ratio"​
 
I've also been using broad match, so it doesn't matter what order the words are as long as they are somewhere in the title.

However, I must have used phrase match sometimes without realizing it because when I went back to check the keywords that were not performing as well as others, the number of results were a lot higher than I had originally recorded. So if I had originally used broad match I would not have used the keyword, and this could be the reason that they are not performing as well.
 
You guys who are trying this, are you using broad match or phrase match when using allintitle?

For example:
broad - allintitle:keyword golden ratio​
vs.
phrase - allintitle:"keyword golden ratio"​
To me, broad match is the only one that makes sense. We know that Google is moving away from keywords and towards focusing purely on the intent of the search, a la RankBrain.

By using the phrase parameter you're completely disregarding what Google believes the intent is, which makes no sense because 99% of searches are done without quotation marks and therefore RankBrain is factored in, very heavily.
 
You guys who are trying this, are you using broad match or phrase match when using allintitle?

For example:
broad - allintitle:keyword golden ratio​
vs.
phrase - allintitle:"keyword golden ratio"​

In the past, I have always used phrase match but I may switch over to broad match as I believe Google has massively upped their synonym game recently.
 
I hate how manual the process is for something so simple and I want to use this as a tool to optimize drafted articles before they hit the web.

Pulled Serpstat's API and a Google scraper, and voila, a one-click "is it gold" button.

kLV44MU.png


I don't think Serpstat is super accurate or comprehensive though. Don't think it will be very useful.
 
Last edited:
Nice one @doublethinker - I also started to make my own tool (just some crude PHP) but soon fell foul of Google banning my IP for suspicious behaviour - are you running this via some proxies too?

I was copy/pasting a CSV of keywords and search volumes for mine. Then I can just use the tool I trust the most for that.
 
Quora and Yahoo Answers are going to send goons to your house if you step into their territory.

That's a hint at one way to take down these kinds of terms at scale without paying for a ton of crappy content and spending forever editing and posting. Seed these questions and topics on answer sites with high metrics, plant your own answer with your link, and vote it to the top. Traffic leak it. Or create your own Q&A site for your vertical and let the users have their own leaks on your site as long as they write the content for your seeded questions.
 
Any of you folks focusing on reigning in relevancy by focusing titles on your specific terms using any NLP tools or processes?
 
Back