You spent two hours exporting keywords. You filtered by volume, picked the ones with “low difficulty,” and wrote a post. Three months later, zero traffic.
This isn’t bad luck. It’s a broken process.
Most keyword research fails because you treat it like a lottery. You look at a number, guess, and move on. But the difference between a keyword that works and one that doesn’t isn’t volume. It’s fit. Fit with your site, your audience, and the actual search results.
Here’s a practical checklist that fixes that.
1. Start with a seed, not a wish
Don’t open a tool and type “best” + your industry. That gives you generic data for generic terms. Instead, list 5-10 seed phrases that describe what your actual customers say when they have a problem.
If you run a plumbing site, a seed like “water pressure low in shower” beats “plumbing tips” every time.
2. Expand with real user language
Use Google Search Console, forum threads, or even customer support tickets. Look for the exact phrases people type when they’re stuck. Tools help, but they show you what tools think people search. Real conversations show you what people actually search.
Take those phrases and paste them into a keyword tool. Keep the messy, long-tail ones. They’re gold.
3. Separate relevance from volume
High volume means nothing if the keyword doesn’t match what you offer. A keyword like “best SEO tools” gets 5,000 searches a month. But if you sell a single niche SEO plugin, that traffic will bounce.
Create a relevance score: 1 (tangential) to 5 (direct match). Discard anything below a 3.
4. Analyze the SERP, not the number
This is where most people stop too early. They look at keyword difficulty and move on. Instead, open the search results and answer three questions:
- Are the top results from big brands or small sites?
- Do they use listicles, guides, or product pages?
- Is there a featured snippet or “People also ask” section?
If the top 10 results are all from Forbes or Amazon, your chances are low. If you see a small blog ranking with a practical guide, that’s a signal you can compete.
5. Check alignment with your site strength
Know your domain’s authority relative to the keywords you want. If you have a DR 30 site, don’t target terms where every result has a DR 70. That’s not “low difficulty.” That’s a trap.
Instead, look for keywords where the average DR of the top 10 results is within 10-15 points of yours.
6. Identify search intent in the results
Look at the top-ranking pages. Do they answer a question, compare products, or sell something? Match your content to that intent.
If the top results are all “how to” guides, writing a “best of” list won’t rank. You’ll be fighting against what Google already decided users want.
7. Prioritize by multiple signals
Don’t rely on one metric. Create a simple scoring system:
| Signal | Weight |
|---|---|
| Relevance | 40% |
| SERP opportunity | 30% |
| Domain fit | 20% |
| Volume (minimum threshold) | 10% |
Score each keyword. Pick the top 10-20 that score highest. Ignore the rest.
8. Store and tag for reuse
Don’t dump keywords into a spreadsheet and forget them. Tag them by content type (guide, list, comparison), stage of intent (awareness, consideration, decision), and priority (high, medium, low). This makes it easy to pull relevant keywords when you plan your next post.
What this looks like in practice
You run a site about home fermentation.
Seed phrase: “how to make kombucha at home”
You expand with real user language from a fermentation forum: “kombucha scoby mold,” “kombucha first fermentation smells like vinegar,” “kombucha carbonation without plastic bottles.”
SERP analysis shows that “kombucha scoby mold” has results from small blogs and forums. No big brand. Your site has DR 25, and the top results have DR 20-30. Relevance score: 5.
You add it to your priority list. Write a guide. Three months later, it ranks #4 and brings 200 visitors per month.
That’s the checklist in action.
Common mistakes that waste your time
- Picking keywords before checking the SERP. The SERP is reality. Everything else is theory.
- Ignoring intent. A keyword with high volume and wrong intent gets you traffic that leaves.
- Trusting difficulty scores blindly. They’re estimates. The SERP tells you the truth.
- Not tagging keywords. You end up redoing research every time you write a new post.
Final practical takeaway
Keyword research isn’t about finding the highest volume term. It’s about finding the term where your site has the best chance to earn traffic, keep it, and convert it.
Use this checklist every time. Spend 20 minutes on the SERP analysis alone. That’s where the real opportunity lives.
FAQ
Q: How long should I spend on keyword research for one post?
A: Plan for 30-45 minutes. The first few times it will take longer. The SERP analysis alone should get 15-20 minutes.
Q: What if I have a brand new site with low authority?
A: Focus on long-tail terms with low competition in the SERP. Look for queries where the top results are from personal blogs or small sites. Avoid terms where authority sites dominate.
Q: Should I include the keyword in the title and H1?
A: Yes, but only if it reads naturally. Forcing a keyword into a bad title hurts click-through rates more than it helps rankings.
Q: How many keywords should I target per post?
A: One primary keyword and 2-3 closely related secondary keywords. Trying to target too many dilutes the focus.
