How to Find Low Competition Keywords With Reddit and Semrush

How to find low competition keywords gets much easier when you stop guessing what people search for and look at a website that already ranks for millions of searches. My method uses Reddit and Semrush together: Reddit ranks for a huge range of long-tail searches, and Semrush’s filters narrow that list down to easy keywords you could write about, or even build a new website around. Reddit has become far more visible in Google, and I use that as a starting point. I don’t assume a keyword is easy just because Reddit ranks for it; Reddit gives me a huge list of topics to investigate, and Semrush helps me narrow it.

Affiliate disclosure: links marked “affiliate link” earn me a commission if you buy, at no extra cost to you. How I review tools. I recorded this in January 2024, so the numbers are what Semrush showed then. For my verdict on Semrush itself, read my Semrush review.

Prefer to watch? Here is the walkthrough.

• 01:10 Low keyword difficulty
• 01:33 Informational keywords
• 01:46 Long-tail keywords
• 02:15 Reddit’s top 10 rankings
• 02:39 Modifier words
• 03:09 Finding the niches

How much does each filter narrow Reddit’s keywords?

Why start low competition keyword research with Reddit?

Because it ranks for so many searches you would never think of yourself. When I entered reddit.com into Semrush’s Organic Research, it showed about 129 million ranking keywords in the US database. That’s far too many to read, but it’s a gold mine of topics, and the all-time graph shows how fast Reddit’s search visibility has grown.

I open Competitive Research, then Organic Research, enter reddit.com and click View all keywords to reach the Positions report. From there, the job is to turn Reddit’s keyword list into a manageable set of content ideas.

How do I filter Reddit keywords for very easy difficulty?

I set Keyword Difficulty to Very easy, which is a score of 0 to 14; Easy is an option if you want a wider list. That took Reddit’s 129 million keywords down to about 15 million.

It’s a big reduction, but 15 million keywords still don’t tell me what to write. A low difficulty score is a useful signal, not a decision, so I narrow the list further by the kind of content I want to create and the length of the search.

How do I find informational long-tail keywords on Reddit?

With the intent and word count filters. I set Intent to Informational, because those are the searches I plan blog posts, guides and answers around. Then I add a word count of more than five words, to get past short, broad searches and into specific long-tail ones. That left about 1.1 million keywords.

Then I limit Reddit’s position to the top 10, which left about 392,000.

To be clear about this step: Reddit being in the top 10 doesn’t make a keyword low competition. I still check the difficulty score and the actual results. The position filter just gives me a focused set of keywords where Reddit is already visible, which in my experience tends to go with lower difficulty scores than average.

Which words help me find content and review ideas?

Modifier words such as best, review, top, versus and compare. Once the main filters are in place, I type one into the keyword filter to see what people search for around it. I don’t treat every phrase with “best” as a buying keyword; they are just ways to explore. In fact, the top “best” results in my test were sayings like “first is the worst, second is the best”, which is exactly why you read the list rather than trust the word.

Adding “best” left about 3,200 keywords, all still Very easy, informational, longer than five words and with Reddit in the top 10. I sort that list by search volume, not by Reddit’s traffic, because I want to know which searches have the most demand. Topics as different as dog food for hunting dogs, organising Pokémon cards and the best bait for a mousetrap appeared, and that variety is why I like starting from a huge website: it shows me subjects I would never have typed into a keyword tool.

How do I turn one interesting keyword into a possible niche?

Find the root word, then add and remove modifiers around it. Suppose “best bait for mousetrap” catches my attention. I open Advanced filters and add a condition to include keywords containing trap. Now I’m not limited to the first idea; I can see whether related searches exist and whether there’s enough variety for more than one useful piece of content.

I did the same with shipping. With “best” still active, a condition for ship showed searches about shipping Pokémon cards and Funko Pops. When I removed “best” but kept “ship”, the list grew to about 2,357 keywords, with all the other ways people phrase shipping questions.

That’s the part I find most useful: start with a specific phrase, find its root word, then add or remove modifiers to see the topic around it. It tells me whether I’ve found a single article idea or a subject worth researching further.

Can I use question words instead of product words?

Yes, and for informational content they often work better. Replacing “best” with how to brought up about 45,000 opportunities under the same filters, full of direct questions to answer. I also try review, compare and what: “compare” returned a much smaller list, while “what” opened about 31,000.

I keep testing words because there’s no single magic filter. The goal is to find the language people use for problems and questions in subjects I can cover well. When a phrase looks promising, I check it one by one: does the topic fit my website, can I answer the search properly, and are there related questions that would help me build useful content?

Is there a way to find low competition keywords without Reddit?

Yes: use the same filters on any seed keyword in the Keyword Magic Tool. Set Keyword Difficulty to Very easy and word count to five or more, then read the list. To prove it works for anything, I tested it with random words: “beer” left more than 5,000 keywords with an average difficulty of 7, and “idea” with a minimum volume of 200 found “ideas for 50th birthday cake” at about 2,400 searches and a difficulty of 12. My Semrush keyword research guide covers that tool in full.

If you don’t have Semrush, you can still find low competition ideas for free with my ChatGPT prompts for keyword research or Google Keyword Planner, then check the results pages yourself. The same questions decide whether a keyword is worth it.

What is my process for choosing a low competition keyword?

1. Enter reddit.com in Semrush Organic Research.
2. Filter for Very easy difficulty and informational intent.
3. Keep phrases longer than five words, with Reddit in the top 10.
4. Add a word such as “best” or “how to”, and sort by volume.
5. Use Advanced filters to explore the root topic of anything interesting.
6. Check each keyword and its results page before deciding to write.

I wouldn’t publish an article just because Semrush shows a low score. The filters find the hidden possibilities; the final choice depends on whether I can create content that genuinely answers the search. If you’re new to how rankings work in the first place, start with what SEO is.

Try Semrush free here (affiliate link).

FAQ

What counts as a low competition keyword in Semrush?

Semrush’s Very easy range is a Keyword Difficulty of 0 to 14, and Easy is 15 to 29. Treat those as a starting point, and check the ranking pages yourself before you commit.

Does Reddit ranking for a keyword mean it’s easy to rank for?

No. It narrows the list to searches where Reddit is visible, which can mean weaker competition, but you still need to check the difficulty score and the actual results.

Should I sort by search volume or by traffic?

By search volume. Traffic is what Reddit receives from the keyword; volume is how many people search for it, which is what matters for your own page.