You’ve done it. You’ve collected every potential keyword under the sun on your chosen research topic. You’ve used AdWords, you’ve looked at competitors and you’ve been checking out industry forums. Hopefully, you would have used my previous blog as a guide to finding as many keywords as humanly possible.
what you’re researching and how specific you were with your initial keyword
research you might be left with lots of irrelevant keywords, this is because
the tools we use aren’t perfect, and they don’t know your company or your
clients as well as you do.
This means it’s
time to remove all the unnecessary codswallop from your list.
Now this is the
real challenge, that can often be rather time-consuming and manual but in order
to get a good understanding of the search volume surrounding the
product/service you want to target it’s a necessary step.
So, buckle up
and let us dive into phase two headfirst.
Step 1: General admin
Paste all of
your keywords that you’ve collected into a spreadsheet. Keep all the data that
comes with the keywords that AdWords spits out, as this might come in handy
with future analysis, however, for the moment we’ll just be focusing on the
search volume and the keywords.
Create a filter
on the top row, this will be used to search through the dataset throughout.
keywords with a search volume of 0.
Step 2: Remove duplicates
This is an
obvious step but can often be overlooked which leads to rather disastrous
implications. Imagine leaving two of the same keywords with large search
volumes in a search landscape document could drastically impact your future
In Excel it’s
dataset, Data > Data Tools > Remove Duplicates
Step 3: Remove irrelevant
First of all,
you need to remove the branded keywords from the dataset. Often when using AdWords,
you may find the keywords it spits out is full of brands and other company
names. These are phrases that more often than not you won’t rank for and so
you’ll want to avoid targeting.
through your dataset making a mental note of any other brand you see. Search
for these brands in the search bar and remove all of these from your
Step 4: Remove irrelevant
Quite often a
few of the usual tools you might use for keyword analysis include locations
that are irrelevant. The same scenario as the branded keywords skim through and
remove any you won’t be looking to target.
keywords by a-z can help you stop a large chunk of irrelevant keywords in one
go, helping to speed up the process.
Step 5: Remove irrelevant years
There’s no use
in ranking for top iPhone X deals 2016 this year. It’s 2019, get with the programme
Step 6: Arrange by character length
are just 1 word tend to be very generic and can often be extremely hard to rank
for. It’s worth removing these from your list as realistically targeting them
is going to be a waste of time.
temporary column next to the keyword column insert the following formula
find the number of words in the keywords. Next, order the values in this column
and delete any that have just one keyword straight away. This will get rid of a
load in one go.
also go the opposite way and check for keywords that are too long and delete
Step 7: Correct search intent
looking for informational or commercial keywords, or perhaps both? If you are
looking for these get rid of the unnecessary keywords.
might not need the other keywords, they might come in handy later. Instead of
deleting them, copy them into another document. This could save you time during
future keyword research.
Step 8: Get manual
you should have drastically hacked away at your long list of keywords from your
research. Now its time to go through manually and check the words that are
scrolling through and checking that the words will provide useful for your
needs. Unless you’re a complete master of your subject you’ll need to check
lots of these keywords, a simple google to check the SERPs will help you along
having two tabs on your screen and a good playlist makes the process more
you have it, the ultimate guide to filtering and removing unnecessary keywords
from your keyword research. Try these techniques to speed up this sometimes-time-consuming
process so that you are left with a lovely list of highly relevant keywords.