Why Manual Keyword Research Will NEVER Let You Down

Using “Keyword Difficulty” metrics to do your niche research is usually a bad idea.

Let me rephrase that. Relying mostly on KD scores is a bad idea. 

It's natural that you're going to use these metrics as part of your research, but time and time again I see people choosing niches based solely on keyword scores and search volumes alone. Sometimes I wonder if they even pay attention to what the keywords mean.

It's no wonder that their websites never amount to much; they've completely missed the point of how to analyze competition in a niche.​

In this article we're going to look at how keyword research fits into niche selection, how you use keywords for choosing article topics and titles, and most importantly, the manual strategies for analyzing your competition.

When it comes down to it, if you put the extra time and effort into following the tips in this post, you'll be much better at niche selection.​

What You'll Learn Today

• How keyword research and competitiveness fits into niche research
• How to manually check competitiveness of keywords and niches as a whole
• How different site types should approach keyword research
• check How to really come up with article titles and ideas

My Beef With Keyword Difficulty Scores

They go by different names depending on the platform, but they are essentially the same thing. KC, KD, the name doesn't matter.

For those not familiar, a keyword difficulty score is something that keyword search tools use to help you analyze how easy it is the rank a site for a particular keyword. On the surface, they work quite well. You'll see that something like “Electric Shaver” is a lot more competitive than something like “Green protein powder reviews”, and rightly so.

However, you really do need to do manual analysis of each keyword, because the KD score can only tell you half of the picture.

When teaching Perrin how to use LongTailPro to do keyword research in his famous “Niche Site Project 2“, Spencer Haws explained that you should use KC (The LongTailPro keyword score) as a filter to eliminate the more competitive keywords. A filter, not a yes/no decision maker.

I fully recommend you read that post and the one before it, because you'll learn a ton about niche research just from them.​

​You can see here that Spencer still analyzed each keyword manually and gave Perrin feedback on the individual results:

This first example is huge. It's a common mistake that people make and highlights exactly my issue with just looking at keyword difficulty scores. You might find a KC 19 keyword which you think is amazing, and jump straight into the niche, only to find out you are competing with eCommerce stores and other sites that you just won't beat.

At the time same, I have found that some people won't touch a keyword if it has a few eCommerce stores in it. In my experience, I've found that you CAN rank among them, so the key is to finding search terms that have a mixture of eCommerce and affiliate/authority sites on page 1 (or even zero eCommerce)​

What I like about this example is that again, Spencer looks at page 1 and looks at the bigger picture. There are 2 or 3 tough sites, but there is clearly room for a new site on page 1.

He also looks at how worthwhile the keyword is. Do you want to just rank position 4 or 5 for a keyword with 1,600 searches a month? You can probably do better.

When I do keyword research, I often find terms where I know I can rank on page 1, but ranking in the top 3 spots would be a huge task. If the keyword only has a few hundred or thousand searches per month, I just don't see the point in going for it, unless the niche as a whole is still promising.​

Using Keywords And Keyword Difficulty For Niche Research

The vast majority of your time researching a niche should be spent doing keyword research, unless you are not planning on relying on Google traffic.

The competitiveness of the SERPs is usually my sole decision maker when deciding whether or not to enter a niche. That said, I'm not talking about finding a few keywords with “best” in them that have a keyword score lower than 30.

Instead, I analyze page 1 and sometimes 2, and do that for a multitude of keywords. What I'm looking for is evidence that there are a few (but not too many) affiliate sites ranking for a multiple of keywords. We call these definite wins. I'm not afraid of finding a handful of eCommerce sites ranking well, and I'm also not afraid of huge authority sites either, but I want to make sure the following:

• There is evidence from Google that affiliate sites can rank for the main keywords in this niche
• There aren't too many affiliate sites ranking (I don't really want to compete with dozens of people who are doing the same link building as me)
• If there are strong authority sites/eCommerce sites present, they are not always in the top 3 positions.
• The search volume is worth the effort. By this I mean, if I think it will be tough to rank in the top 3 positions for a keyword, but the volume is 20,000 per month, it might still be worth it. 

This is also how we train our keyword researchers to pick niches for our Done For You sites. Sometimes a customer will query why we've picked a niche with eCommerce terms or some high keyword difficulty scores, and the answer is that we looked manually at the SERPs and determined it was still a viable niche.

It's worth noting that the first few times you do this, it will be quite frustrating, and you will need discipline. There are plenty of times when I've rushed through niche research and almost entered a niche, only to look at it again later and realize it's not worth it. 

Break your keyword research up into several sessions, so you get at least one chance to look at it with fresh eyes.​

How To Do Manual Analysis

Up until this point, you might have been thinking that this all sounds logical enough, but wondering how to actually do the manual analysis.

I'm going to tell you this. Sometimes it's going to come down to experience and instincts, but for the most part, the keyword tools you use will actually help you with manual analysis too.​

Tools like Long Tail Pro, SeCockpit, or even the Ahref's keyword tool will all let you do an analysis of the sites currently ranking on page 1 for that keyword. This is usually how they generate their own difficulty score too.​

What you want to ask yourself are the following questions:

1. Are the sites on page 1 eCommerce sites? If so, how many? – You want there to be at least 3 or 4 sites that are not eCommerce​ sites, such as Amazon or a Shopify type store.
2. Are the sites huge magazine/authority sites? The same rule applies to eCommerce sites. Is there room for an affiliate site on page 1? How about at the top of page 1? – Over time you get good at recognizing the big authority sites in your niche, but a general rule of thumb is that they have massive amounts of content, huge amounts of backlinks, have a variety of different writers (sometimes), and blog on a number of different, albeit related, topics. They also usually have very high Domain Authority, 50 or above.
3. What kind of Domain Authority and Majestic Trust Flow do these sites have? If the majority have a DA of 40 or higher, you might want to look elsewhere. And yes, DA is just as relevant today as ever for this part of niche research.
4. Do you see the same sites appearing for many different keywords? If there are a couple of small niche sites that are appearing for lots of keywords, and are near the top, then there's no reason you couldn't do the same. If you keep seeing the same huge authority sites occupying the top 3 or 5 spots though, you may not have space.

Two Examples Using Manual Analysis

Below we see two examples of seemingly similar keywords found using SeCockpit. At this point we are not interested in search volume or keyword difficulty, I just want to show you what the manual analysis shows:

Let me just walk you through this keyword, what you are looking at, and why I wouldn't try to target it.

​1.) Even though there are lots of green squares and the average Page Authority isn't that high, Domain Authority is huge. Every site apart from 2 is a huge site with very strong DA. The only exception is a site which is in position 10 and probably won't climb much higher, and a site in position 5 which is using PBN's to rank and hiding its link profile. It looks like it has weak DA, but it doesn't.

2.) This keyword is dominated by huge sites like Reviews.com, consumerreports.org, thesweethome.com, amazon.com, bestreviews.com, and toptenreviews.com. 1 or 2 of them wouldn't be an issue, but all of them together? No thanks.

3.) The sites are ranking based off their authority, rather than on-page or off-page SEO. This means that you could optimize your site better than them and eventually beat them, but you'd have to spend a long time building up your own authority. For a keyword with 27,100 searches, it *might* be worth it if there are lot's of other keywords in the niche you can target in the meantime. ​

Why bother, when there are other ideas out there?

While this niche isn't as good as it used to be, since lots of people entered it as copycats of my LuxuryShaves site, this example still shows what a more appealing keyword looks like. 

1.) As well as low PA, most sites also have low DA.

2.) Most sites are not huge authority sites like in the first example. They are mostly niche sites, and some of them are quite new. When I first ranked Luxuryshaves for this search term, at least 4 of the sites there now didn't even exist.

3.) There are lots of other long-tail keywords that these sites also rank for, showing that the niche is “niche site” friendly.​

Note: SeCockpit uses DA and PA still whereas LTP uses Majestic Trust Flow instead. Both are equally useful for checking out the strength of sites like this. Trust Flow tends to rely too much on backlinks and is a better metric for PBNs.

When Different Site Types Affect Your Decisions

Let's face it. If you're planning to build out a niche site with around 30 pages, or an authority site that starts with 50+ and adds more content every week, you're going to have a different strategy.

You'll be approaching link building differently, you'll be targeting more or fewer keywords, and you'll be able to rank differently.

An authority site with huge amounts of content is going to be able to rank for more keywords. You might even want to tackle “best electric toothbrush”, because on a long enough timeline, an authority site can rank for anything. You'll also be able to get a lot more white-hat links, which means you don't have to worry about overdoing it with the gray hat.​

As well as volume of links, authority sites are going to have so much more content and more in-depth articles, that they won't necessarily be worried about seeing the big authority sites on page 1. It just means you'll have to spend longer outranking them.​

So the summary of this section is that if you want to build a niche dominating authority site, you can afford to target slightly harder keywords.​

Using Manual Analysis For Individual Articles

Now if you already have a site up and running, you may be thinking that this article hasn't been of much use to you. However, you can still use the methodology outlined above to help you choose better keywords for your posts.

There will be many situations where you have a topic in mind, but aren't' sure what the best keyword to use for it would be, and other situations where you are doing keyword research to come up with new niche ideas. In both these instances, manual analysis is going to be a huge benefit to you.

At HPD I still use manual analysis when deciding on keywords from time to time, but at this stage I am confident of ranking for most terms.​

Before moving onto the advanced techniques, I think it's important to stress how important keyword research is to your site. It's literally the concrete on which you lay the foundation of your business – at least at the beginning. 

Of course you can always start off doing keyword research wrong, pivot and then correct the path of your site but that's exactly why we offer Keyword Packs now. We noticed a lot of people trying to replicate our process, instead we thought we'd just sell the keywords to people so they could at least start off on the right foot.​

More Advanced Techniques

As well as using keyword tools, you can also use things like Semrush to get ideas for articles and a general overview of a niche.

By checking out 2 or 3 sites in a niche and reverse-engineering their best keywords, you'll be able to get a feel for the niche and consider whether there are a lot of keywords a site like yours (whether it's new or old) could also rank for.

However, for the meantime I recommend you just focus on mastering the manual analysis techniques mentioned above, before getting yourself overwhelmed.​

Humans Will Always Be Better Than Keyword Research Tools

A lot of “keyword difficulty” scores are calculated by using similar methods and data to what I taught you to do in the manual section. That said, there is still plenty of room for improvement. You also need to be aware that there are factors keyword tools can't pick up that well, such as the type of site ranking, their backlink profiles, and ​sometimes their on-page SEO.

By all means use KD scores as a filter to speed up your keyword research processes, but make your final decisions based on your own analysis.​