Keyword Research for Beginners: Find Terms You Can Actually Rank For

Most beginners don’t fail at keyword research because they pick bad tools — they fail because they chase terms that were never winnable in the first place. This guide shows you how to use free tools, read volume and difficulty numbers without being fooled by them, and walk out of a single evening with a real keyword list you can start writing from.

Updated August 2026 · 14 min read

A few years ago I helped a friend launch a blog about backyard vegetable gardening. She had written eleven articles before anyone looked at keywords, and every single one targeted phrases like “gardening tips” and “how to grow tomatoes.” Eleven articles, six months, almost no traffic. The problem wasn’t her writing — it was that she was competing against university extension programs, major seed companies, and sites with fifteen years of authority. When we redid her keyword list with one rule — only target terms where a small site already ranks on page one — her next ten articles started pulling in visitors within weeks. Same writer, same effort, completely different outcome. That’s what this guide is about: not finding the biggest keywords, but finding the ones you can actually win.

What keyword research actually is (and what it isn’t)

Keyword research sounds like a technical chore, but it’s really just structured eavesdropping. People type things into Google millions of times a day, and keyword research is the process of finding out exactly which words they use, how often, and how hard it would be for your page to show up when they do.

What it is not is a hunt for the highest-volume phrases in your niche. That instinct — “go where the people are” — is the single most expensive mistake beginners make. A keyword with 40,000 monthly searches is useless to you if the first page is locked up by Wikipedia, Amazon, and three government sites. You could write the best article on the internet and still sit on page four, which is functionally the same as not existing. Research from multiple click-through studies over the years has consistently shown that the vast majority of clicks go to the top few results, and page two gets a rounding error’s worth of traffic.

So reframe the goal. You’re not looking for the terms everyone searches. You’re looking for the terms where you specifically have a realistic shot at the top five results. That shift in framing is 80% of what separates keyword research that works from keyword research that wastes six months of your life.

Why “rankable” beats “popular” for a new site

Google’s job is to show searchers the most trustworthy, useful answer. When your site is new, Google has no reason to trust it yet — no track record, few or no external links pointing to it, thin content library. Google’s own SEO starter guide makes clear that earning visibility is about demonstrating helpful, reliable content over time, not about tricks. For competitive head terms, you’re up against sites that have spent years building that trust.

Here’s the mental model I use. Imagine every keyword’s first page as a ten-seat table. For “gardening tips,” the seats are taken by sites with millions of monthly visitors. For “how far apart to plant cherry tomatoes in raised beds,” the seats are often taken by forum threads, a Reddit post from 2019, and a small blog or two. If small, imperfect pages already hold seats at the table, that means Google is willing to seat small sites there — which means there’s a seat available for you.

Don’t ask “how many people search this?” Ask “who is currently ranking, and can I write something clearly better than them?”

This is also why volume matters less than you think. An article that ranks #2 for a 300-search-a-month keyword brings you roughly 60–90 visits a month. Ten articles like that bring you 600–900 monthly visits. That’s a real audience, built from keywords most competitors consider too small to bother with. Meanwhile, the beginner who swung for “gardening tips” is still at zero.

The free tools you actually need

You can spend hundreds of dollars a month on keyword tools, and at some point that may make sense. But for building your first list, free tools are genuinely enough — I still use most of these weekly. Here’s the stack:

One more thing about Keyword Planner: you don’t need to run ads to use it. Create a Google Ads account, skip the campaign setup (you can pause the dummy campaign immediately), and the tool is yours. The ranged volumes annoy people, but for a beginner’s purposes, knowing a keyword gets “100–1K” searches instead of “10–100” is usually enough resolution to make a good decision.

Step one: brainstorm five to ten seed topics

A seed is a broad topic word you’ll expand later — not a keyword you’ll target directly. Open a spreadsheet (Google Sheets is fine) and spend twenty minutes listing seeds. For my friend’s gardening blog, the seeds were things like: tomatoes, compost, raised beds, pests, watering, container gardening, seed starting, harvesting.

Where do seeds come from? Three places, in order of importance:

• Your own knowledge. What do people ask you about? What confused you when you started? If you’ve ever explained something to a friend, that’s a seed.
• Communities. Spend fifteen minutes in the subreddits, Facebook groups, or forums for your niche. Note the questions that come up every single week — recurring questions are recurring searches.
• Competitor navigation menus. Look at how established sites in your niche organize their categories. Their category names are professionally validated seed topics, handed to you for free.

Keep seeds broad but not absurd. “Tomatoes” is a good seed. “Vegetables” is so broad it will generate noise. “Purple heirloom tomato varieties for containers” is already a keyword, not a seed — save that specificity for step two.

Step two: expand each seed into a raw list

Now the fun part. Take one seed and run it through this loop:

1. Autocomplete mining. Type your seed into Google, but don’t hit enter. Note every autocomplete suggestion. Then type the seed plus a letter (“tomatoes a…”, “tomatoes b…”) and note new suggestions. Then try question starters: “why tomatoes,” “can tomatoes,” “tomatoes how.” Do this in an incognito window so your own search history doesn’t skew the suggestions.

2. People Also Ask. Now actually search the seed. The “People also ask” box usually shows four questions — click a couple and more load. These are literal questions searchers ask, which makes them some of the best article and subheading ideas you’ll find anywhere.

3. Related searches. Scroll to the bottom of the results page. The “Related searches” row hands you eight more variations.

4. Keyword Planner pass. Paste your seed into Keyword Planner’s “Discover new keywords” and export the ideas. This is where you get volume ranges attached, plus hundreds of variations you’d never have typed yourself.

Run this loop per seed and you’ll easily generate 30–80 raw candidates per seed. Don’t evaluate yet — just dump everything into your spreadsheet, one keyword per row, with the volume range in the next column. Evaluation is step three and four; mixing generation and judgment slows both down. Expect to spend 10–15 minutes per seed. For a deeper dive on stretching seeds into hundreds of specific phrases, the long-tail keywords strategy guide covers that process in detail.

Step three: read the search results before trusting any number

This is the step almost every beginner skips, and it’s the step that matters most. Before you write a single word targeting a keyword, search it and study page one like a detective. Volume tools tell you how many people search. The results page tells you whether you can win.

For each promising keyword, open the first page of results (again, incognito) and ask:

• Who’s ranking? Forums (Reddit, Quora, niche message boards), small blogs, and Pinterest pins on page one are a green light — Google is seating weak competition. Nothing but household-name brands, government sites, and Wikipedia is a red light.
• Do the titles actually match the query? If you search “how far apart to plant cherry tomatoes” and the top results are generic “tomato growing guide” pages, nobody has written the focused article. That’s an opening.
• Is the content any good? Click the top three results. Thin, outdated, badly formatted, or clearly off-target content means you can beat it with something genuinely better.
• What format is winning? If page one is all listicles, Google has decided searchers want a list. If it’s all videos or product pages, your how-to article won’t rank no matter how good it is. Match the format that’s already winning.

That last point deserves emphasis, because format mismatches kill more beginner articles than weak writing ever will. What you’re really checking is the intent behind the query — what the searcher actually wants to accomplish. The search intent explained article breaks down the four intent types and how to spot each one from the results page.

Page one of Google is the most honest keyword difficulty tool ever built — and it’s free.

Volume vs difficulty: weighing the numbers without being fooled

Now put the pieces together. You have keywords with volume ranges, and you have your own manual read on competition. How do you decide what makes the final list? A few hard-won rules:

Volume is a tiebreaker, not a target

For a new site, a keyword with 100–1K monthly searches and weak page-one competition beats a 10K keyword with strong competition every single time. The traffic you actually get is volume multiplied by your realistic click share — and your click share on an unwinnable keyword is zero. Zero times 10,000 is still zero.

Difficulty scores are guesses — verify them yourself

Every tool’s “keyword difficulty” score is an estimate, mostly based on how many links the ranking pages have. Useful as a rough filter, but I’ve seen 40-difficulty keywords with three forum threads on page one (easy win) and 15-difficulty keywords where page one is all major publishers (skip it). Never let a number substitute for actually looking at the results page as described in step three.

Beware the seasonal trap

Keyword Planner volumes are monthly averages. A keyword like “when to plant pumpkins” might average 1K searches a month — but 90% of those happen in April and May. Check anything seasonal-sounding in Google Trends before committing, so you know whether you’re writing for year-round traffic or a two-month spike.

Prefer specific over broad, even at lower volume

“Tomato plant leaves turning yellow with brown spots” (low volume) will convert a searcher into a loyal reader far better than “tomato problems” (higher volume), because the specific query has a specific, answerable need. Specific terms also tend to carry clearer intent, which makes them easier to write well.

Step four: build your one-evening keyword list

Time to assemble the final list. Work through this checklist with your spreadsheet open:

• Cut everything with strong page-one competition. If your manual check showed only big brands and authority sites, delete the row. Be ruthless — these are the keywords that waste your months.
• Cut keywords you can’t write well. If you have nothing genuine to add beyond what already ranks, the keyword isn’t for you, even if it’s winnable. Keep only topics where you can produce something clearly more helpful.
• Keep 15–25 keywords. That’s roughly three to six months of content at one article a week. A list of 200 keywords feels productive but paralyzes; a list of 20 gets written.
• Add three columns: volume range, competition note, format. Your competition note should be one honest line: “Reddit thread ranks #3” or “top result is from 2019.” Format reminds you what shape the winning article takes: list, how-to, comparison.
• Group near-duplicates. “How far apart to plant cherry tomatoes” and “cherry tomato spacing in raised beds” are the same article. One keyword becomes the primary target; the others become subheadings inside it. Never write two articles for two phrasings of one question — they’ll compete with each other.
• Rank by “write first” order. Put the clearest, most confident wins at the top — the ones where page one is visibly weak and you know exactly what you’d write. Momentum matters more than volume ordering.

When you’re done, your spreadsheet is your content calendar. Each row is an article, already vetted for winnability, with the format decided. That is the entire deliverable of keyword research, and it’s completely achievable in one focused evening.

A worked example: one evening, start to finish

Here’s exactly how this played out for the gardening blog, compressed into one session:

7:00–7:25 p.m. — Seeds. We listed eight seeds: tomatoes, compost, raised beds, container gardening, pests, watering, seed starting, harvesting.

7:25–8:30 p.m. — Expansion. We ran the loop on just four of the eight seeds (tomatoes, compost, raised beds, pests) using autocomplete, People Also Ask, related searches, and a Keyword Planner export. Result: 214 raw keywords in the spreadsheet with volume ranges. We saved the other four seeds for next month.

8:30–9:20 p.m. — Triage and SERP reading. We sorted by volume, then checked the top 35 candidates manually — literally searching each one and scanning page one. About a third died immediately (major publishers everywhere). Another third were promising. We also found gold: “tomato leaves turning purple underside” had forum threads ranking, a 100–1K volume range, and no dedicated article anywhere in the top five.

9:20–9:45 p.m. — Final list. We kept 19 keywords, grouped six duplicates, and ordered the list. First article assigned: the purple-leaves one. It ranked in the top three within about ten weeks.

Total time: under three hours, and half of that was the manual SERP reading — the part beginners are tempted to skip. Don’t skip it. That hour is where the wins are found.

Mistakes that waste your first six months

I’ve watched enough new sites stall to recognize the patterns. Avoid these:

• Chasing head terms. If the keyword is two words and everyone in your niche wants it, you will not rank for it this year. Come back to it when you have fifty articles and real authority.
• Trusting difficulty scores blindly. Treat every score as a hypothesis to verify with your own eyes on page one, not as a verdict.
• Ignoring intent mismatches. If page one for your keyword is all product pages and you’re planning a blog post, you’re not in the running. The query is commercial and Google knows it.
• Collecting keywords forever. Research feels safe; publishing feels risky. Cap your list at 25 and start writing. You can research batch two after you’ve published batch one.
• Writing before checking. Never draft an article first and “find a keyword for it” later. The keyword — and the page-one read — comes first, every time.

What to do with your list next

A keyword list earns nothing sitting in a spreadsheet. The immediate next moves: publish your first article from the top of the list, then set up Google Search Console on your site if you haven’t. Search Console is free, takes about fifteen minutes to verify, and within a few weeks it starts showing you the actual queries your pages get impressions for — including dozens of keywords no tool would ever have suggested. Our Google Search Console guide walks through the setup and the reports worth watching. Those real-world impressions become the seeds for your second research evening, and that second list will be sharper than your first because it’s built on data about your site, not the whole internet.

From there, the rhythm is simple: write from the list, watch Search Console, refresh the list every couple of months. Keyword research isn’t a project you finish — it’s a habit that gets faster and more accurate every time you do it. But it starts with one evening, one spreadsheet, and the discipline to only pick fights you can win.

Sources and further reading: Google’s SEO starter guide covers how search engines evaluate and surface content. Volume and competition figures described here reflect how free tools report ranges as of August 2026 and may vary by niche. We update this guide when the free-tool landscape materially changes.