Every paid keyword tool sells you two things: a list of phrases, and a judgement about whether you can win them. You can get both free, and the free version of the second one is often more honest than the paid number.
Step 1: get the phrases from autocomplete
YouTube's search box completes what you type using what real people actually search. That is not an estimate or a model. It is a record.
- Type your topic and note every completion.
- Then type your topic followed by each letter of the alphabet, one at a time. Each letter returns a different set.
- Then put question words in front: how, what, why, best, is.
Twenty minutes of this produces more genuine long-tail phrases than most people know what to do with. Phrases with four or more words are the ones worth writing down, because they tell you exactly what the viewer wants.
Our Chrome extension automates this step in its keyword tool, showing the suggestions as clickable chips, but the manual version costs nothing but time and works identically.
Step 2: judge competition from the results page
This is the step paid tools charge for, and it is the step you can do better yourself, because the evidence is right there.
Search the phrase and look at the first ten results. Ask four questions:
How big are the channels ranking? If every result is a channel with millions of subscribers, the topic is settled. If channels under 100,000 subscribers are holding positions, the door is open.
Are any of them outperforming their size? A video with more views than its channel has subscribers means the topic pulled in viewers who did not know the creator. That is a topic being discovered through search and suggested, which is exactly the kind you can win.
How old are the top results? If the best answers are three years old, there is room for a current one. If they are all from last month, you are late to something competitive.
Do the results actually answer the phrase? Sometimes the top videos are loosely related because nothing better exists. That gap is the opportunity.
Our extension scores this automatically, showing an overall score with interest and competition broken out, plus a flag on any small channel that is ranking. But the four questions above are the whole method, and you can run them by eye.
Step 3: check the search volume claim carefully
Here is the part worth being cynical about. No tool outside Google has YouTube's search volume data. Every "monthly searches" number you see in a YouTube SEO tool is modelled, usually from Google web search data or from clickstream panels, and then presented with a confident-looking precision it has not earned.
We deliberately do not publish a search volume figure, because we would be making it up. What we show instead is what the top results are actually doing: median views per day, how big the channels are, how fresh the videos are. That is measured rather than modelled, and it answers the real question better anyway.
If a tool shows you "8,100 monthly searches" for a phrase, treat it as a rough ordering signal between phrases, not as a forecast of your views.
Step 4: pick topics where interest and openness overlap
The topics worth making are where demand is real and the competition is beatable. High interest and high competition is a waste of a week. Low interest and low competition is a video nobody watches.
The sweet spot is usually a specific long-tail phrase inside a popular subject. Not "photography tips", which is settled, but the particular question people ask that nobody has answered properly this year.
Step 5: write the title for the phrase, not the tags
Once you have the phrase, put it near the front of the title, in natural language. Put it again in the first two lines of the description, where it is visible before the fold. Say it out loud in the first thirty seconds of the video, because the transcript is indexed.
Tags come last and matter least. See do YouTube tags still matter for why.
What this method cannot do
It cannot tell you whether a topic will trend next month. It cannot see search volume. It cannot predict your retention, which is the thing that actually decides whether you keep the position after YouTube tests your video.
What it can do is stop you making videos about settled topics, and point you at the ones where a smaller channel is already proving the door is open. That is most of the value, and it costs nothing.