Keyword research isn't always fun.
But it is necessary whenever you're starting most digital marketing campaigns.
It's mandatory for some channels such as SEO and PPC.
Even for things like content marketing, I still highly recommend doing keyword research to understand your target audience better.
So, where do you start?
If you're like most marketers and business owners, you go to Google's very own keyword planner.
It's the only way you'll get any real data from Google itself.
There are many keyword research tools out there, but for the most part, they just pull data from Google's keyword planner anyway.
They may do it in a more effective way than you could on your own, but it's important to understand that they still have limitations.
Why?
Because Google will never tell you everything it knows, just bits and pieces.
So, while the keyword planner is a fine starting point for keyword research, it is not enough.
If you only use the keyword planner, you will end up missing out on many opportunities and spending your time and resources on keywords that aren't as good as they appear.
That's where this post comes in. I'm going to show you 4 different ways that Google's keyword planner can mislead you.
1. Averages don't always tell the whole story
How does the keyword planner come up with a single monthly search volume for each keyword suggestion?
It averages the previous 12-month period.
You may have known that, but do you know how this can affect your keyword research?
It can have a big impact.
Most niches do not have a consistent search volume year round as a whole.
And neither do keywords. If you hover over the little graph icon in a set of keyword results, you'll see a little graph pop-up showing you the search volumes for that keyword over time:
In the above picture, the peak search volume is about 8,000, while the minimum is around 3,000. The peak is more than double the minimum.
This isn't always a big deal, but there are two main reasons why you should be checking the monthly search volumes for individual keywords.
You miss out on emerging keywords: If you're the first one to write about a topic, you'll automatically rank #1 for its keywords most of the time.
Being at #1 gets you more links when people search for the keyword and then link to your results.
But if you do keyword research based only on the averages, you'll never be #1.
Do you see why?
New popular keywords come along every once in awhile in just about every niche.
Search volume slowly picks up steam, and often, it starts growing exponentially at some point.
It'll look something like this:
The average search volume that the keyword planner shows for that keyword is 18,100.
To me, that's not an 18,100 keyword-it's likely over 100,000 from here on out (maybe much more).
You'll probably notice an 18,100 search volume, but imagine if it was a keyword with a 900 search volume. You might skip over that, not noticing the emerging trend.
In reality, if it's just picking up speed, it could be an over 10,000 searches per month keyword.
By the time Google shows you the number big enough to grab your attention, you will have already missed your chance to be among the first by a long shot.
If you're going to skip over a keyword, check its recent search volume first.
Consistent is better: If two keywords have the same average search volume, would you prefer the search volume to be consistent or highly variable?
You'll find that some keywords in your niche are evergreen (popular all year round), while others are highly seasonal.
Here's an obvious example: snow shoveling:
Of course, searches like this one spike in the winter months and almost disappear in the summer months.
If you don't rank in the top 3 by the main winter months, you'll derive just about zero benefit from all your work until the next year.
Considering that getting to rank highly for a keyword can take weeks or months, you never know.
SEO has a long enough wait time to produce rankings, so you probably don't want to wait too long to start getting traffic (if you still have your rankings at that time).
Additionally, when you have a spike of traffic, it's much harder to split test to improve conversion rates. It's better to have consistent traffic so you can consistently test new variations.
You may not have a choice and have to target seasonal keywords, but sometimes you do.
So, check the variation in the monthly searches when you're considering which keywords to target. Invest your efforts into the most consistent ones.
2. Beware of rounding
One more thing about those averages: they're not “true averages.”
Ever noticed how all the keyword suggestions from the planner have search volumes that end in zeros?
That's because there's rounding going on behind the scenes.
But Google is not always rounding up or down to the nearest 10; instead, it groups keywords into “buckets.”
Picture a bunch of buckets in a line with a value assigned to each of them.
Google throws keywords with similar search volumes into each bucket, likely because it makes handling all the data simpler on its end.
As you get to the higher numbers, there are fewer keywords to go in the buckets, and that's when Google removes a bunch of buckets.
At low numbers of searches, most keywords are rounded to the nearest 10.
However, keywords with even a few thousand searches can be off by hundreds in either direction because the next closest bucket is far away.
At really high search volumes, the differences can be even bigger.
The keyword.io team did a great analysis of 57 billion different search terms in the keyword planner.
The x-axis represents the different buckets that Google shows.
The y-axis is the number of keywords that come back for each bucket.
As you'd expect, there are many more low search volume terms than those with huge search volumes.
More importantly, you can see the differences in bucket sizes.
At first, the difference is small (10 > 20 > 30 > 40 > 50 > 70).
However, that difference quickly increases (720 > 880 > 1,000 > 1,300 > 1,600).
Why is this a big deal? The obvious reason is because you want accurate search volumes.
A more common reason is because it makes comparing similar keywords incredibly difficult.
Say you have two keywords in your results:
- “Keyword 1” – 1,000 searches per month
- “Keyword 2” – 1,300 searches per month
The second keyword is obviously way better, right?
In reality, the first keyword might have 1,149 searches per month while the second 1,151.
Essentially, they're identical.
Or if the two keywords were both showing 1,000 searches per month, it's possible that one actually has 1,149 while the other 941 (about a 20% difference).
The higher the search volumes are, the less certain you can be about the actual number of searches.
Which means that when you're trying to decide which keyword to go after based on such data, it's likely that you'll make wrong decisions.
What can you do about this? The unfortunate part is that there's nothing you can really do to fix the problem.
The best thing you can do is mitigate the issue by not putting all your eggs in one basket.
Focus on long-tail keywords when possible, and only once you start getting some real data in Google Analytics and webmaster tools should you heavily invest in any particular keywords.
3. Misspellings and variations affect search volume a lot
This particular quirk of the keyword planner doesn't lie to you, but you need to be aware of it, or your keyword analysis will be incorrect.
When you search for a keyword, Google will show you different results based on the specific variation you enter.
For example, if you search for the TV show “brooklyn nine nine,” you'll see these numbers as the top results:
I chose this example because there are multiple variations that mean the exact same thing from the searcher's perspective.
- Brooklyn 99
- Brooklyn ninenine
- Brooklyn nine-nine
When you enter these other variations, you get different sets of results:
This is strange because if you type them into Google itself, it knows what you mean when you type any variation.
Why this is important: If you're comparing the search volumes for different keywords, you need to make sure that you're considering all variations.
Try out different misspellings and see if they have any search volumes (they won't show up unless you type in the exact misspelling).
Then, add all the search volumes of the variations together to get a more accurate representation of the overall search volume for your main term.
Keep in mind that you're adding rounded search volumes together (point #2 in this post). This means that with each term you add, your figure becomes less accurate (but still more accurate than if you ignore the variations).
4. Did you know that Google hides keywords?
It's understandable that Google doesn't want to hand over all its data to SEOs.
But not all limitations of the keyword planner are designed on purpose; some just exist due to the way the tool works.
The most important one is that Google won't show you all the keywords you want to see.
It's not malicious in any way, but it really impacts your keyword research.
For example, let's say you typed in “wooden decks”:
I've filtered down the results to only closely related ones (that have “wooden” and “decks” in them).
That's all of the results I got.
But when I searched for “how to build a wooden deck,” I got a search volume of 210:
Even in the full original results, that keyword was not there.
In addition, “how to build wooden decks” has another 10 searches per month.
No, these aren't big keywords, but they illustrate the point that there are obviously related keywords that won't show up when you search for terms.
The solution? Again, there's no concrete solution. The best you can do is enter as many seed terms as you can and include several variations.
Additionally, use a tool such as Keywordtool.io to get keywords from other sources, and then run those through the keyword planner to get exact search volumes.
The data is there; it's just hidden until you find the keyword from other sources.
Should you abandon the keyword planner?
These are some pretty big limitations, which begs the question in this heading.
I don't think you should abandon the keyword planner. Why? Because the data, while not perfectly accurate, is still the only real data you can get from Google.
However, I think as a way to discover keywords in the first place, it has extreme limitations.
There are many great alternative keyword research tools out there that are worth the few dollars they cost to use.
What they typically do is extract a bunch of seed keywords from different sources and then run those through the keyword tool for you. Then, they return to you a more complete set of keyword results than you'd get if you used the planner yourself.
You could do all of this yourself, but it will take you a ton of extra time, which is just not worth it in most cases.
Conclusion
Google's keyword planner is a great tool, which should be used by all marketers and business owners for keyword research.
However, it has limitations.
I've shown you the 4 main limitations of the tool and what you should do to mitigate their negative effects.
Go forward with your keyword research in the future, but keep this post in mind. Don't use the keyword planner as your sole tool for keyword research, or you'll miss out on a lot of great opportunities.
If you have any questions about any of these concepts, just leave me a comment below. Also, if you love a particular keyword research tool, share it with everyone.
No comments:
Post a Comment