Disclosure: this post contains affiliate links. We may earn a commission at no extra cost to you.
New websites rarely fail because the writing is bad. They fail because they target keywords they can't win. A brand-new blog won't outrank huge sites for "best laptops" or "email marketing", however good the article is.
The fix is to target low-competition keywords: specific searches with real demand where the current results are weak. This guide shows a free, repeatable method to find them, and a 5-point check to confirm you can rank.
In short: find specific, question-style searches through Google autocomplete, forums and free tools; then check the first page of Google for weak results (forums, thin or outdated posts, small sites). If at least two of the top 10 are weak, you can probably compete.
What makes a keyword "low competition"?
A low-competition keyword usually has:
- A specific, longer phrase (3+ words): "best free CRM for a cleaning business" vs "CRM"
- Clear intent: you can tell exactly what the searcher wants
- Weak results on page one: forum threads, outdated posts, thin content or small sites
- Some search demand: even 50 to 500 searches a month is worth it; dozens of these add up
Keyword difficulty (KD) scores from tools are a useful shortcut, but they're estimates. Always look at the actual results.
Rough KD guide for a new site: 0 to 20 = start here · 20 to 35 = possible with a great post · 35+ = save for later, once your site has some authority.
Free tools you need
- Google search itself (autocomplete, "People also ask", related searches)
- Google Search Console: once your site is live, it shows which searches you already appear for
- Ahrefs Webmaster Tools / free keyword generator: free keyword ideas and difficulty estimates
- Google Keyword Planner: free with a Google Ads account (no need to run ads); gives volume ranges
- A spreadsheet: to track and prioritise ideas
Method 1: Google autocomplete and "People also ask"
- Open a private/incognito window and go to Google.
- Type your topic followed by a space, and note every suggestion.
- Try modifiers: "best ___ for", "how to ___", "___ vs", "___ for beginners", "___ for small business", "is ___ worth it".
- Use the alphabet trick: "crm for a", "crm for b"... each letter gives new ideas.
- Search a few of the best ideas and copy the questions from "People also ask" and "Related searches" at the bottom.
These are real searches people make. Questions in "People also ask" are also excellent H2s and FAQ entries.
Method 2: Forums and communities
Where real customers ask questions, there are keywords nobody has written a good answer for.
- Search Google for `site:reddit.com your topic` or `your topic forum`.
- Look for questions that keep coming up, with lots of replies but no single great answer.
- Note the exact wording people use. That's how they'll search too.
- Check Facebook groups, industry forums and the questions your own customers email you.
Bonus: if a forum thread ranks on page one for a question, that's a strong sign the keyword is winnable (see the SERP check below).
Method 3: Free keyword tools with filters
- Enter a broad topic into a free keyword tool.
- Filter for keywords with low difficulty (under 20 to 25) and some volume.
- Prefer question keywords and "for audience" keywords.
- Export or copy the best 20 to 30 into your spreadsheet.
Paid shortcut: Semrush's Keyword Magic Tool can filter thousands of ideas by difficulty, intent and volume in seconds. A free trial is enough for one big research session. 👉 Try Semrush free (affiliate link)
The 5-point SERP check: can you actually rank?
Tools estimate. Google's first page tells the truth. Search the keyword and check the top 10:
| Check | Good sign for you |
|---|---|
| 1. Forums or Q&A sites (Reddit, Quora) | Ranking in the top 10 |
| 2. Content quality | Thin, outdated (2+ years) or doesn't really answer the question |
| 3. Site size | Small blogs or niche sites, not only giant brands |
| 4. Exact match | Few results use the exact phrase in the title |
| 5. Intent fit | You can write something clearly better and more specific |
Rule of thumb: if 2 or more of the top 10 results tick these boxes, the keyword is worth targeting.
Prioritise by intent and money
Not every low-competition keyword is worth the same. Score each one in your spreadsheet:
| Column | What to write |
|---|---|
| Keyword | The exact phrase |
| Est. volume | From Keyword Planner or a free tool |
| Difficulty | Tool score + your SERP check (Low/Med/High) |
| Intent | Informational, commercial ("best", "vs", "review") or transactional |
| Monetisation (1-10) | Could this post recommend a product or lead to a sale? |
| Priority | High intent + low difficulty first |
Commercial keywords ("best X for Y", "X vs Y", "X alternatives") earn the most from affiliate links. Informational keywords ("how to") build traffic, links and email subscribers. You need both.
Common mistakes
- Chasing volume. 10,000 searches you'll never rank for are worth less than 200 you can win.
- Trusting difficulty scores blindly. Always do the SERP check.
- Ignoring intent. If the top results are product pages and you write a blog post (or vice versa), you won't rank.
- One keyword per post only. Each post should also answer related questions; you'll rank for dozens of variations.
- Writing for the keyword, not the reader. Use the keyword naturally; never stuff it.
Your 60-minute keyword research session
- 10 min: pick 3 topics your customers care about.
- 15 min: collect ideas from autocomplete, "People also ask" and forums.
- 15 min: run them through a free keyword tool; keep anything under ~25 difficulty.
- 15 min: do the 5-point SERP check on your top 10.
- 5 min: score and pick your next 5 posts.
Free: get 50 ChatGPT prompts for your business, sent straight to your inbox. Get the free prompts ()
Once you've picked your keywords, write the post with our AI + human workflow and promote it while you wait for Google, using the tactics in how to get your first 1,000 visitors.
FAQ
What is a good keyword difficulty for a new website?
Aim for under 20 to 25 on most tools for your first few months, and always confirm with a manual look at the search results.
Is Semrush worth it for beginners?
It's powerful but not essential at the start. Free tools plus a manual SERP check are enough for your first 30 posts. A free trial is useful for one big research session.
How many keywords should one blog post target?
One main keyword, plus the related questions and variations people search for on the same topic.
Can I rank for keywords with zero search volume?
Tools often under-report small, specific searches. If real people ask the question in forums, it's often worth writing about, especially if it's close to a sale.
How long does it take to rank for a low-competition keyword?
Often 1 to 6 months on a new site, faster if you link to it from other posts on your site and promote it.
