How to Find Low-Competition Keywords for a Small Website

I still remember the exact moment I realized I was doing keyword research completely wrong. It was a Tuesday night, about two years into running my first real niche site. I’d spent six weeks writing a 3,000-word guide targeting “best budget laptops.” I was proud of it. I’d covered every angle, added comparison tables, even made my own graphics. Then I checked my rankings three months later. Page 9. Position 87. A single Amazon product listing was beating me.

That article got maybe 12 visitors in six months. I’d poured hours into a keyword that was essentially a brick wall. The problem wasn’t my writing. It wasn’t my site speed. It was that I never bothered to check if I could actually win that fight.

Once I stopped chasing the shiny, high-volume keywords and started looking for the quiet ones, things changed. Within four months, a different article targeting a weirdly specific phrase started pulling in 300–400 visitors a month. Not huge, but it was real. And it kept growing. That one lesson — learning how to find low-competition keywords — did more for my site than any “SEO hack” I’d ever tried.

Here’s what I’ve figured out since then, including the tools that actually work and the mistakes that still make me cringe when I think about them.

What “Low-Competition” Actually Means (And Why It’s Not Just About Search Volume)

Most people hear “low competition” and think it means “low search volume.” That’s half right, but it’s the half that gets people into trouble.

A keyword can have 5,000 monthly searches and still be low competition if the pages ranking for it are weak. Conversely, a keyword with 200 searches can be brutally competitive if it’s dominated by Reddit threads and Quora answers that Google loves.

The real signal is the search engine results page. If the top 10 results are mostly big brands, review sites with massive backlink profiles, and pages that have been ranking for years, you’re not going to crack that. But if you see forum posts, thin content, outdated articles, or pages that don’t actually match the search intent? That’s your opening.

I learned this the hard way. I used to filter keywords purely by volume and “keyword difficulty” scores in tools. Then I’d write something, publish it, and wonder why nothing happened. The scores were lying to me — or more accurately, I was misreading them. A KD score of 15 doesn’t mean much if every page in the top 10 has 10,000 referring domains.

The Tool Stack That Actually Helped Me (And What I Stopped Using)

I’ve tried almost everything. Ahrefs, Semrush, Moz, Ubersuggest, Keywords Everywhere, AnswerThePublic, Google Keyword Planner. Here’s my honest take: you don’t need to spend $99 a month to find good low-competition keywords, especially when you’re just starting.

Keywords Everywhere is the one I still pay for. It’s a browser extension for Chrome, Firefox, and Edge that shows search volume, CPC, and competition data directly in Google search results. It’s freemium — the free version gives you keyword suggestions and on-page insights, but the paid version (starts around $15 for 100,000 credits that last a year) unlocks the volume data. I install it, type a seed keyword into Google, and immediately see related terms with their volumes right there. No switching tabs. No exporting spreadsheets.

Ubersuggest is what I recommend when someone has zero budget. Neil Patel’s tool has a free tier that gives you a handful of searches per day. The interface is straightforward — you type in a keyword, it spits out related terms, questions, and prepositions. The keyword difficulty scores are rough, but they’re good enough to filter out the obvious losers. For small businesses and freelancers, the free plan genuinely gives you enough to get moving.

AnswerThePublic used to be my go-to for question-based keywords, but their pricing has crept up over the years. These days I use a free alternative: just type your seed keyword into Google and look at the “People Also Ask” box. Those are real questions people are typing. I also use the autocomplete trick — type your keyword followed by “a,” then “b,” then “c,” and so on. Google fills in the rest. It’s tedious, but it’s free and it surfaces phrases you’d never think to search for.

Google Trends is underrated for one specific thing: spotting rising topics before they get saturated. I don’t use it for volume data (it only shows relative interest, not absolute numbers). But if I see a related query marked “Breakout” or showing a sharp upward curve, that’s a signal to pay attention. I’ve caught a couple of small trends early that way, and the keywords were almost embarrassingly easy to rank for because nobody had written about them yet.

The tool I stopped using? Ahrefs. Not because it’s bad — it’s genuinely excellent. But for a small site owner watching every dollar, the $99/month minimum wasn’t justifiable when Keywords Everywhere and Ubersuggest covered 90% of what I needed. If you’re running a client agency or managing a portfolio of sites, sure. For a single small website? There are smarter ways to spend that money.

The Method I Use Now: A Step-by-Step Walkthrough

This is the exact process I run whenever I’m planning content for a new site or trying to revive a stagnant one. It takes about 45 minutes per topic cluster, and it’s saved me from writing at least a dozen dud articles.

Step 1: Start With a “Seed” That’s Already Specific

Don’t start with “fitness” or “personal finance.” Start with something like “bodyweight workouts for small apartments” or “budgeting apps for freelancers with irregular income.” The more specific your seed, the more useful your keyword suggestions will be.

I normally find my seed keywords in three spots: Search Console queries where I’m getting impressions but ranking poorly, subreddit discussions related to my topic, and the People Also Ask results on Google. “People Also Ask” boxes on Google.

Step 2: Mine Reddit and Quora for the Real Questions

This is the step most people skip, and it’s the one that’s given me the best results.

Reddit users don’t phrase things like keywords. They phrase things like humans.

A Reddit post asking “Why does my traffic drop even though I didn’t change anything?” isn’t something I’d target as a keyword. It’s a signal about a real frustration.

But it tells me exactly what someone is worried about.

I go to a relevant subreddit, sort by “Top” over the past year, and read through the titles. I’m not looking for advice. I’m looking for the way people describe their problems. Then I take those phrases and check if there’s search volume behind them.

For example, I found a thread where someone asked, “Am I wasting time writing blogs for my startup?”

If you type that exact phrase into a keyword tool, the monthly search count is probably tiny.

. But it led me to “how long does SEO take for new websites” and “SEO for early-stage startups” — both of which had decent volume and almost no good content targeting them specifically.

Quora works similarly. Search your niche, look at the questions with the most views, and note the exact wording people use.

Step 3: Run the Keyword Golden Ratio (KGR) Check

This is a technique created by Doug Cunnington that I’ve found genuinely useful for finding “hidden” keywords. The idea is simple: if a keyword has low search volume and very few pages using that exact phrase in their title tag, you can rank quickly even without backlinks.

The formula is:

KGR = Number of “allintitle:” results ÷ Monthly search volume

A score of 0.25 or lower is excellent. Between 0.26 and 1.0 is worth trying. Anything above 1.0 is probably too competitive for a new site.

Here’s how I do it in practice:

1. I take a long-tail keyword I’m considering (say, “best budget mechanical keyboard for coding”).

2. I Google allintitle:"best budget mechanical keyboard for coding" and note the number of results.

3. I check the monthly search volume in Keywords Everywhere or Ubersuggest.

4. I divide the first number by the second.

If the result is under 0.5, I’ll usually write the article. If it’s over 1.5, I’ll move on unless there’s a specific reason to believe I can beat the competition.

One caveat: KGR works best for keywords with fewer than 250 monthly searches. Above that, the allintitle count tends to be less meaningful.

Step 4: Inspect the SERP Manually

Never trust a keyword difficulty score alone. I open an incognito window, search the keyword, and look at the top 10 results. I’m asking three questions:

• Who’s ranking? If it’s all Amazon, Walmart, and massive media sites, I pass. If there are forum threads, small blogs, or pages that clearly weren’t written for that exact query, I lean in.

• What’s the content quality? If the top results are thin, outdated (check the date), or don’t actually answer the question, that’s a gap I can fill.

• What’s the search intent? Are people looking to buy something? Learn something? Compare options? If the intent doesn’t match what I can offer, I move on.

I’ve passed on keywords with KD scores of 5 because the SERP was dominated by Reddit threads with thousands of upvotes. And I’ve targeted keywords with KD scores of 30 because the top results were clearly not optimized for the query.

Step 5: Cross-Check With Google Trends

Before I commit to writing, I plug the keyword into Google Trends. I set the timeframe to the past 12 months and check whether interest is stable, rising, or declining. I also look at the “Related queries” section — sometimes there’s a rising query that’s even better than the one I started with.

A declining trend isn’t always a dealbreaker. Sometimes a keyword is seasonal, and if I publish at the right time, I can catch the wave. But a steep, consistent decline is usually a sign to look elsewhere.

Where I Went Wrong (And What I’d Do Differently)

Mistake 1: Trusting keyword difficulty scores blindly. I once targeted a keyword with a KD of 12, thinking it would be easy. The top 10 results were all Reddit threads from 2018 with hundreds of comments. Google loved them because they matched the “discussion” intent perfectly. My polished article never stood a chance. Now I always check the SERP before writing a single word.

Mistake 2: Ignoring search intent. I wrote a detailed tutorial for a keyword where people were clearly looking for a product comparison. The content was good, but it didn’t match what searchers wanted. It’s like bringing a screwdriver to a hammer fight. Now I ask myself: if I were searching this, what would I actually want to see?

Mistake 3: Chasing volume over fit. Early on, I’d see a keyword with 10,000 searches and get excited, even if it had nothing to do with my site’s topic. That article never ranked because Google couldn’t figure out why a site about home coffee brewing was suddenly writing about car insurance. Topical relevance matters more than raw volume.

Mistake 4: Not updating old content. I had a post that was ranking on page 2 for a decent keyword. Instead of writing something new, I updated that post — added a new section, refreshed the data, improved the internal linking. Within three weeks it jumped to position 4. Nine times out of ten, that’s less work than publishing a brand-new piece.

A Real Example (With Numbers)

Let me give you a concrete example from my own site. I run a small site about home coffee brewing. One of my best-performing articles targets the keyword “how to clean a burr grinder without disassembling it.”

Here’s what I knew before writing:

• Monthly search volume: 480

• Allintitle count: 12

• KGR: 0.025 (excellent)

• SERP: Top results were mostly forum threads and a couple of thin blog posts. One was from 2019 and clearly hadn’t been updated.

I wrote a 1,800-word guide with step-by-step photos. It took me about four hours total. Within two months, it was ranking on page 1. By month four, it was position 3. It now brings in roughly 200–250 visitors a month.

That’s not life-changing traffic. But it compounds. I have about a dozen articles like that now, and together they account for the majority of my site’s organic traffic. None of them are glamorous. All of them were relatively easy to rank for because I checked the competition before I wrote.

A Quick Note on Tools and Budget

You don’t need to spend a lot to do this well. Here’s the setup I’d recommend for someone with a small site and a tight budget:

• Keywords Everywhere (paid, but cheap) — for volume data directly in Google

• Ubersuggest free tier — for brainstorming and rough difficulty scores

• Google Search Console — for finding keywords you’re already getting impressions for

• Google Trends — for trend checks

• Reddit and Quora — for real human phrasing

• Incognito Google searches — for manual SERP inspection

That’s it. No $99/month subscriptions required. The expensive tools are great if you have the budget, but they’re not the difference between success and failure. The difference is whether you actually check the SERP before you write.

Related Post: Future of Digital Marketing: The Role of AI Marketing Tools

What I’d Tell Someone Starting Today

If you’re running a small website and you’re tired of writing content that goes nowhere, stop trying to compete for the keywords everyone else is fighting over. Look for the weird, specific, slightly awkward phrases that people actually type into Google. Check if the pages ranking for those phrases are actually good. If they’re not, you’ve found an opening.

Write something genuinely helpful. Make it the best answer to that specific question. After that, repeat the same process with your next keyword idea.

The traffic won’t come overnight. But it will come. And it’ll keep coming, because you’re building on ground you can actually hold.