Recently Long-Tail Pro changed from using Moz metrics to Majestic metrics. There was a bit of fuss among various marketing groups about how KC had changed.
Some keywords were now KC 40 instead of 30, and others were reversed. People were wondering if their golden nugget keywords were suddenly not so golden. Was KC 40 the new 30? Or did the new metrics show more insight? Were the SERPS less easy after all?
Several days later, LTP released a statement saying that some of the KC calculations were now off, but they would be getting them adjusted.
Personally, while all this was going on, I really didn't care too much about the change, and felt that many people were missing the point.
Don't get me wrong, LTP is a fantastic keyword tool, and I agree that Majestic is a better crawler than Moz; I do tend to use Trust Flow more than DA/PA as a sign of a site's authority. I think it was good to switch to Majestic.
However, I also know that competitive metrics are impossibly difficult to get right, and that KC scores (regardless of the tool used) should only be used as a starting point.
Never choose a keyword or niche based on KC alone. Just don't do it.
I constantly see people posting questions in Facebook groups along these lines:
“I've found a niche where 5 keywords only have KC 30 or lower, but the search volume is only 500. Should I go for this?”
Or worse, I see a screenshot of a LongTailPro research list with keywords blanked out and a bunch of KC scores revealed, and I'm asked if this is a good niche.
The problem with this, is that it's fundamentally impossible to advise on a niche based on KC scores and search volume alone.
Without being able to see the keywords, analyze the sites on page 1, and weigh up things such as buyer intent, keyword type, and commission level, it's just not possible to determine whether a niche might be good to enter.
Even then, it's only an educated guess.
Why Manual Will Always Be Better
Keyword Tools have come a long way in the last few years, and Long Tail Pro has always stood out to me as a tool that understands the fundamentals of keyword research. If you followed Spencer and Perrin through the second niche site project, you'll have learned exactly how to analyze SERPS for their competitiveness.
KC is a tool that helps you narrow down this search, make no mistake about it. By using KC as a filter, you can save yourself a few hours worth of research…but it really doesn't stop there.
The same is true for SECockpit, the keyword tool we use for our own niche research (it's cloud based, which is essential for how we work).
SECockpit also has a keyword difficulty metric, but we never pick a keyword without looking at page 1, and here's why.
Despite keyword difficulty being based on the metrics of the sites on page 1, it's still not going to be 100% accurate. I've seen SERPS with a low KC score that I know will be hard to rank for, and I've ranked sites without building a single link for keywords that have a high KC score.
Equally, if you've done as many searches using LTP and other tools as I have, you'll have seen many a situation where KC tells one picture, but the analysis you do yourself by looking at page 1 tells you a different picture.
I can't list every example here, but to help you understand, let's talk about some things you need to consider aside from KC.
Occasionally, a customer will ask us why we have recommended a keyword that appears to have a high competition, and the answer is that we analyzed the SERP and determined that it was still rankable, despite the high score. This is precisely why we manually check every keyword in our sites and keyword packs.
Types of site in the SERPS
Sometimes you might find keywords where the sites on page 1 aren't very optimized, and have low metric scores. Despite this, you can't necessarily outrank them. A good example of this is when the search is for an e-Commerce term, such as “Product type”. You might see a KC score as low as 15, a bunch of unoptimized or low DA/PA sites, and think you've got your gold nugget keyword.
However, if you're trying to rank an affiliate site in there, you'll find that more often than not, Google just isn't going to put you there.
The same is true when you've got a bunch of sites that are ranking on their DA alone. For example, the keyword “Best safety razors” doesn't look that high KC-wise, but the sites on page 1 are all strong authority sites that are going to prove tricky to overhaul (believe me, I tried!).
This works in your favour as well though. I've seen some keywords with pretty high KC scores where forums are ranking, and I know that I can rank among them. In cases like this, I actually tend to ignore KC.
The point I'm making here is that KC might be giving you some green lights when it comes to competition, but you really need to analyze the SERPS themselves and learn how to really gauge competition manually.
Get Educated and Get Practicing
To learn more about how to do this, I suggest you start out with some of the posts Spencer and Perrin made on this subject.
Visit the overview page for Niche Site Project 2, and check out the first 4-5 “Core” posts.
Even though this project is a couple of years old now, the fundamentals haven't changed when it comes to analyzing SERPS.