What is your KW research process

Dec 17, 2015
I'm interested in people's processes or tips for doing keyword research. I've been thinking recently this is somewhere I need to improve. Not necessarily in terms of quality research but streamlining. My approach is very random with loads of bits of paper with scribblings on and loads of tabs open mainly in ahrefs. I never seem to approach my research in the same way and it takes ages. Im sure I also miss some nuggets in the mayhem. What I usually end up with is good but it's just not organised and its time consuming. I's like to develop a basic process to follow. As you can probably tell from the above, I'm not very good at focusing, tend to get distracted, ramble etc. Its probably better if I lay out a couple of points I want to improve.

1. When starting a new project I try to work out roughly how many money pages I will need for the niche/site. This will generally be a primary term with decent volume and medium difficulty and then some additional related terms with lower volume and difficulty. My issue is always when to branch off and generating related terms which don't cannibalize other pages. Any tips on deciding how many pages deep you should go and avoiding overlapping terms or wasting time creating too many pages?

e.g Say I'm going into the widget niche and I've decided I want to target 'red widgets', 'green widgets' and 'black widgets' as my initial research suggests these are good terms with decent volume. Is there a process you use to decide how far you go? Do you go
Widgets > Green widgets > Best green widgets
Green widgets reviews
Where to buy green widgets

Or do you go
Green widgets > Best green widgets > Best green widgets for Moms
Best budget green widgets
Best green widgets 2019
Green widgets reviews > Brand 1 widget review
Brand 2 widget review
Brand 3 widget review
Where to buy green widgets > Store review 1
Store review 2

There must be a threshold where the decision is made not to go too far or you feel you can rank all terms without additional sub pages?

2. Any tips on presenting your research for your writers? I seem to end up giving them broad instructions and editing quite a lot. Seems pointless when I can probably get better content if I provide better instructions.


If they cease to believe in u, do u even exist?
Staff member
BuSo Pro
Boot Camp
Digital Strategist
Sep 15, 2014
You are missing "Intent" due to not having a clear visual of your customers. I suggest creating 3 Customer Profiles (Day 3 - Marketing Research) around who you want to target.

The way you are doing it is simply going after keywords and volumes - and missing the bigger picture. If you had a customer profile you would be able to create better instructions for your writers since they will be able to visualize who they are writing for. You'll also have direction within your content to make the content feel like it's directly talking to the consumer's needs and addressing them. It will feel like a one-on-one conversation versus generic advice.

I'll admit Market Research is no fun. One guy I was consulting had a similar problem and I straight up asked him why he didn't take a look at the DSCC (Digital Strategy Crash Course), he said he read ALL OF them EXCEPT for Market Research out of arrogance cause he assumed he knew, well he wasted a ton of time on things that could have been avoided.

Market research is the most important thing for any business since you can identify your customers - and whether you are even in a profitable niche/industry or not.

I hate the widget examples, so I'm going to use the shoes example I use in the Keyword Research Day:

In the example we see that "womens shoes" has average Google organic results valuing products at $24.95 to $85. If you can rank for that term and generate you 20 sales a day, you'll be generating $499 to $1700. Great.

However if you did market research and understood women - there is a brand called "Christian Louboutin" which are known as "red bottoms" (cause the heels have red bottoms). If you were to take time to rank for "red bottom heels with diamonds" which cost $795 to $995 per pair - even only 10 sales a day would generate you $7950 to $9950 in revenue - PER DAY. Even a single sale could generate you more revenue than the other scenario.

Notice I didn't even mention keyword volume. Anyone can figure out that "womens shoes" is probably going to have a ton more search volume, 40,500 monthly searches - great. Tons more competition too. But you'll have a ton of window shoppers in the mix as well.

However someone typing in "red bottom heels with diamonds" is probably less of a window shopper and knows what they want. Even better "BUY red bottom heels with diamonds" or "BUY red bottom heels" (10 monthly searches) and you are in the money.

What's the search volume for "red bottom heels with diamonds" - most tools would have it at 0 searches. "Buy red bottom heels" comes in at a whopping 10 searches a month.

Traditional keyword research would literally eliminate a ton of profitable, more lucrative terms, and simply ignore the macro grand scheme of things.

With a proper customer profile you can create "Topics" that are relevant to individual users and therefore can create the support content needed.

I guess what I'm trying to say is "Topic" search is a better start after understanding your target audience, versus the low level keyword research. Honestly you might be in a fog just going after keywords and not thinking of the above level intent or target customer.

Think about this - you are selling a CrockPot, pressure cooker. A suburban mom who's husband makes $150K a year is probably a better target than a single mom with 3 kids living in a poor neighborhood. OR maybe not - if the single mom can buy a product that allows her to cook food while she is out at her job.

However the suburban mom has more access to disposable income and probably does things like throw dinner parties that the poor single mom doesn't. So you come up with an article for the suburban mom called "CrockPot Dinner Party Recipes" - you know the exact target audience.

There is a ton of nuances that you can extract once you have a visual of who you are targeting, and then relaying that customer profile to your writers will make their jobs 10x easier since they can visualize who they are writing for.

So the steps should be as follow:

1. Determine each customer profile (I suggest 3)

2. Research what makes each customer profile tick.

3. Create Category Topics that are central to each profile.

4. Research keywords within each topic and find terms consumers use (example: "red bottoms") that aren't necessarily in the lingo of keyword tools.

5. Create intent based content that around those topics with the keywords. Remember there is a general funnel to a consumer understanding a topic/niche/industry.

There are beginnings that probably never heard of "red bottoms" but got a ton of money and want to make their appearance better with expensive shoes (Awareness - Informational level), 1st level.

There are people that are researching different brands within that industry (Interest - Navigational), 2nd level.

There are people that want Christian Louboutin shoes and are in the desire stage - actively looking for website selling "red bottoms" - 3rd level.

Then there are the people ready to pull the trigger RIGHT NOW and are looking for a "sale" or "discount" or "coupon" or the final incentive cause they are looking to take Action (Transactional) - 4th level.

You can create content for each level going backwards - the 3rd and 4th being the most profitable, but the 1st and 2nd necessary since you can reach a broader audience that you remarket your brand to constantly to push them through the next level of the funnel.

6. Promote the content pieces and get feedback to improve and generate newer ideas.

7. Profit.​
Mar 18, 2019
You can also work backward from products to topics.

Let's say you're building a traditional Amazon Affiliate site. View your niche's product category on Amazon. Sort by price, from high to low. Look for products that go for a high price, and which have a large number of reviews, say 100+. This is a pretty strong indicator that buyers are fond of this product, are willing to pay a high price for it, and are enthusiastic enough to be reviewing it on Amazon.

Of course, you may want to check Fakespot or read some reviews to confirm that they aren't all fake. But assuming they're not, this product is likely worth publishing on, regardless of what ahrefs says.

Check out the competition for the product on Google. If several strong sites are already dominating, you can save the product for later. If not, then publish a solid review.

Even if the product name + review is getting zero searches according to ahrefs, you will still likely get traffic and sales on this page, based on the evidence of buyer behavior on Amazon.