Here is the brutal truth about keyword research. Most of the keywords you will ever think of are already taken. The big sites — Wikipedia, Amazon, the New York Times, Healthline, Forbes — own the obvious phrases, and they own them with thousands of backlinks you cannot replicate this year, or maybe ever. If you try to rank for "running shoes" or "home decor" or "recipes," you are fighting a war you cannot win.
But here is the good news. Roughly seventy percent of all Google searches are long-tail queries — specific, multi-word phrases that nobody is targeting. That is where you live. That is where you win. In the next fourteen minutes, I am going to walk you through the exact one-hour process that takes you from zero to one hundred rankable keywords. By the end, you will have a spreadsheet full of topics you can actually write about and actually rank for, instead of wishful thinking. Four steps. One hour. Let's go.
STEP 01 Pick the right seeds — 10 minutes
A seed is a narrow, specific topic you genuinely care about or have expertise in. "Indoor herb gardens" is a seed. "Gardening" is not — gardening is an ocean, and you will drown in it. The rule is simple: pick five seeds, no more, no fewer. Five is enough to give you breadth, few enough that you can go deep.
For each seed, brainstorm five branches. There are five reliable categories that work for almost any topic. First, specific items — for herb gardens, that would be basil, mint, rosemary. Second, common problems — yellow leaves, pests, leggy plants. Third, equipment and tools — grow lights, pots, soil mixes. Fourth, situational constraints — small apartment, no sunlight, winter conditions. Fifth, beginner mistakes — overwatering, wrong lighting, planting too early.
Five seeds times five branches gives you twenty-five sub-topics to work with. That is your raw material. Write them down in a spreadsheet before you move on.
If you cannot list ten sub-topics under a seed in sixty seconds, the seed is too narrow. If you can list fifty, it is too broad. Five seeds, five branches each, is the sweet spot.
STEP 02 Multiply with modifier columns — 15 minutes
Now we turn twenty-five branches into hundreds of long-tail phrases. Open a fresh spreadsheet and create three columns. Column A is question words: "how to," "why does," "when to," "what is," "best way," "can you." Column B is your topics from step one: basil leaves, grow lights, indoor mint, herb soil, pruning herbs. Column C is buyer signals: "turning yellow," "in apartment," "no sunlight," "for beginners," "on a budget," "without soil."
Now you combine them. "Why does basil keep turning yellow indoors." "How to grow mint in apartment with no sunlight." "Best grow lights for beginners on a budget." "What is the best herb soil without compost." Each combination is a real query that real humans type into Google. A single seed easily produces fifty to one hundred long-tail phrases this way, and most of them nobody is writing about yet.
Then validate with Google Autocomplete. Type the beginning of each phrase into Google's search box and watch what it suggests. Every suggestion is a query that real people are searching for right now — Google only suggests phrases with meaningful volume. If your phrase triggers a suggestion, you have a real keyword. If it does not, cross it off.
The keyword you would never think of is the keyword nobody else thought of either. That is exactly why you can rank for it.
STEP 03 Filter by KD and volume — 15 minutes
You now have a few hundred raw keywords. Most of them are bad. This step separates the winners from the trash. Run your full list through any keyword tool — Ahrefs, Semrush, Ubersuggest, or the free Keywords Everywhere browser extension. You only need two numbers per phrase: Keyword Difficulty, usually called KD, and Monthly Search Volume.
Here is the filter. Keep any keyword with KD under twenty and volume between one hundred and two thousand searches per month. That combination is the sweet spot — low enough competition that a brand-new site can rank, high enough volume to be worth the article you are going to write. Below one hundred searches a month, the traffic is too thin to justify the effort. Above two thousand, you have usually walked into a high-competition fight with big sites.
A KD between twenty and thirty-five is a maybe. Keep these only if the search has clear commercial intent — words like "buy," "best," "review," "versus," or a brand name. Commercial queries convert, so even a little traffic is worth the ranking fight. Anything above thirty-five in KD, drop without mercy. No matter how exciting the phrase looks, you will lose that battle in year one.
STEP 04 Run the SERP smell test — 20 minutes
Here is the secret that separates amateurs from pros. Keyword Difficulty scores from tools are estimates. They are useful as a first filter, but they are wrong often enough that you cannot trust them alone. The first page of Google itself — the SERP — tells you the real truth. For every keyword that survived step three, paste it into an incognito Google window and look at the top five organic results.
Ask four questions about what you see. First, are at least two of the top five sites with Domain Authority under thirty? You can check DA with any free Chrome extension. Small sites ranking means a new site can rank too. Second, is there a Reddit thread, a Quora answer, or a forum post in the top five? If Google is surfacing forum content, it means no great article exists yet — that is your gap. Third, are the top articles thin? Under a thousand words, generic, or clearly off-topic for the specific query? Thin content is beatable content.
Fourth and final question: does the intent match a real article you could write? If the top results are all product pages and you want to write a guide, the intent is wrong and you will not rank no matter how good your article is. Hit two or more of these positive signals, and you have a winnable keyword. Hit none, drop it.
Tools tell you the score. The SERP shows you the game film. Always watch the tape before you play the game.
That is the entire system.
Five seeds, modifier multiplication, two-number filter, four-question SERP test. One hour, start to finish. If you run this process properly, you walk away with roughly one hundred keywords that a new or mid-sized site can actually rank for. That is not a wish list. That is six months of content fully planned, with each article having a realistic path to the first page of Google.
Here is what surprises most people. The content teams winning at SEO are not writing better articles than you are. Their writing is often worse. What they do better is pick better keywords before they write a single word. They skip the fights they cannot win and focus entirely on the keywords they can own. An hour of keyword research saves you six months of wasted writing. So close this lesson, open a spreadsheet, and go run your hour. Your first hundred keywords are waiting.