Most keyword research advice ends with a pricing page. This post does not. Here is how to do keyword research for free, start to finish, using tools that cost nothing and data that comes straight from search engines.
You can run this whole process today. It takes about two hours the first time and 30 minutes a month after that. At the end you will have a spreadsheet of keywords, an intent label for each one, a rough difficulty read, and a page mapped to each. That is the entire job.
The free stack that actually works
You need five things. All free.
- Google Search itself. Autocomplete, People Also Ask, and Related Searches at the bottom of the page. This is the biggest free keyword database on earth and nobody uses it properly.
- Bing Webmaster Tools. Free account. It has a Keyword Research tool that gives actual monthly numbers, not ranges. Bing traffic is smaller than Google, but the relative demand between two terms is usually the same.
- Google Keyword Planner. Free inside a Google Ads account. You do not have to run an ad to open it. You get volume as ranges like 100 to 1K unless you are actively spending.
- Google Search Console. Free. This is the only tool on earth that tells you what people searched before landing on your site. Nothing paid beats it.
- Google Trends. Free. Compare up to five terms and see direction over 12 months.
That is it. Add a spreadsheet and you have everything. Free keyword research is not a worse version of paid research. It is the same research with more manual clicking.
Step 1: Build a seed list from your own business
Do not start in a tool. Start with a blank sheet and write 15 to 25 seed terms. Pull them from four places.
- What you sell. Write the plain-English name of every service or product. Not your internal name. The name a customer would say out loud.
- Sales call language. Go through your last 10 enquiries. Copy the exact words people used to describe their problem. “My site looks fine but nobody calls” becomes seeds like website not converting and low conversion rate website.
- Competitor navigation. Open three competitors. Read their top nav and footer. Those menu items are seed keywords they already decided were worth money.
- Support and FAQ. Every question you answer twice a week is a keyword.
Twenty seeds is plenty. Quality beats quantity here because step 2 multiplies everything by roughly 30.
Step 2: Expand seeds into hundreds of real queries
Take seed one. Type it into Google and do not press enter. Write down every autocomplete suggestion. Those suggestions are real queries real people typed. No estimation involved.
Then run what people call alphabet soup. Type your seed plus a space plus “a”, note the suggestions, then “b”, then “c”, all the way through. Twenty-six passes per seed sounds tedious. It takes about 90 seconds and it is the single highest quality free keyword source that exists.
Then add modifiers in front and behind:
- how, what, why, when, can, should
- best, cheap, free, top, alternative
- vs, or, versus, compared to
- for small business, for beginners, near me, without
Now press enter. Screenshot or copy the People Also Ask box. Click one question to expand it, and more appear. Keep clicking and you can pull 20 plus questions from a single search. Scroll to the bottom for Related Searches and take all eight.
Repeat on YouTube autocomplete and Bing autocomplete. They use different data. YouTube skews toward how-to and tutorial phrasing, which is gold for video and blog content.
One more source most people skip. Search site:reddit.com “your topic” and read the thread titles. Reddit titles are written by customers, not marketers. They are the cleanest source of real problem language you can get.
Twenty seeds through this process gets you 400 to 700 raw keywords. Dump them all in one column. Delete duplicates with the remove duplicates function.
Step 3: Get volume numbers without paying
Now you need to know which of those 500 terms anyone actually searches.
Bing Webmaster Tools first. Create a free account, verify your site, open Keyword Research. Paste in a term and it returns related keywords with monthly search numbers. The numbers are Bing numbers, so they are smaller than Google. A useful rough conversion is to multiply by somewhere between 5 and 10 for a Google estimate, but honestly the ranking of terms against each other matters more than the absolute number.
Then Google Keyword Planner. Sign up for a Google Ads account and skip the campaign creation. Open Tools, then Keyword Planner, then Get search volume and forecasts. Paste up to 1,000 keywords at once. Download the CSV. You get ranges, not exact figures, which is fine. You are sorting into buckets, not building a financial model.
One warning. Keyword Planner groups close variants together and reports them as one number. That is why “keyword research tools free” and “free keyword research tools” often show identical volume. Treat them as one target, not two pages.
Google Trends last. Take your top 10 terms and compare them five at a time over 12 months. Kill anything on a clear downward slope. Keep anything rising even if volume looks small today.
Cut anything with no measurable volume in either tool unless it is an exact description of what you sell. Long tail terms with a volume of 10 a month can still be your best page if the person searching it is ready to buy.
Step 4: Read the SERP to judge intent and difficulty
A volume number tells you nothing about whether you can win. The search results page tells you everything. This is the step people skip and it is the step that decides whether your page earns traffic.
Open your target term in an incognito window. Look at the top 10 and answer four questions.
- What format ranks? If the top 10 are all listicles, your service page will not rank. If they are all product pages, your blog post will not rank. Match the format or lose.
- Who ranks? If you see forums, Reddit threads, small blogs, or pages that clearly have not been updated in four years, the term is gettable. If it is nine national brands and Wikipedia, park it.
- How many results use the exact phrase in the title? Count them. Zero to three means most rankers are ranking accidentally and a focused page beats them. Eight or more means everyone is targeting it deliberately.
- How much page is above the first organic result? Ads, a map pack, a featured snippet and a video carousel means the click is already spoken for. The number on your spreadsheet is not the number you can capture.
Give each keyword a gettable score of 1 to 3 based on that read. This manual check is more accurate than any difficulty score because you are looking at the actual competition rather than a model of it. It just takes about 60 seconds per term.
Step 5: Mine Search Console for keywords you already rank for
If your site has any traffic at all, this step is worth more than steps 1 through 4 combined.
Open Search Console, go to Performance, set the date range to the last 6 months, and turn on Impressions, Position and CTR. Then filter to positions 8 to 25. Sort by impressions, high to low.
Those queries are terms Google already thinks you are relevant for. You are on page two or the bottom of page one. Moving position 12 to position 5 takes far less work than ranking a brand new page.
Look for three patterns:
- High impressions, no dedicated page. A blog post is picking up a query by accident. Build the real page for it.
- Decent position, terrible CTR. Position 6 with a 0.4 percent click rate means your title and description are not matching what the searcher wants. Rewrite the title tag. That is a 10 minute fix.
- Question queries. Every question in your Search Console export is a heading you should add to an existing page.
Export the whole thing to CSV and paste it into your keyword sheet as a separate tab. This is proprietary data about your site that no paid subscription can sell you. Our SEO approach starts here on every single project, before anyone writes a word of new content.
Step 6: Score, map, and stop researching
You now have a messy sheet. Turn it into a plan with four columns.
- Business value, 1 to 3. Would someone searching this buy from you within 90 days? A 3 is someone comparing providers. A 1 is a student doing homework.
- Gettable, 1 to 3. From your step 4 SERP read.
- Volume bucket. Low, medium or high. Not the raw number.
- Target page. One URL per search intent. If two keywords produce near identical results pages, they are one page, not two.
Sort by business value, then gettable. Work from the top. The first six months of most sites should be spent on 3-3 terms with low volume, because low volume plus high intent plus a weak top 10 is exactly where a small site wins.
Then stop. Research is not the work. Publishing is the work. Once you have 20 mapped keywords, close the sheet and go write. If you want the wider system this sits inside, the Cheap Marketing Playbook covers how this feeds into a content plan that is not just blogging into a void.
When a paid tool is genuinely better
We will not pretend otherwise. Free hits a ceiling in three places.
Scale. The manual SERP read is 60 seconds per keyword. At 40 keywords that is fine. At 2,000 keywords across a large ecommerce catalogue it is not a job a human should do. Paid tools return a difficulty score and a top 10 snapshot in bulk.
Competitor gap analysis. There is no free way to enter a competitor domain and pull back every keyword they rank for. You can approximate it by reading their sitemap.xml and their page titles, which honestly gets you 70 percent of the way there, but it is slower and less complete.
Backlink data. Free tools give you almost nothing here. If you are competing for terms where links are the deciding factor, you will be guessing.
If none of those three describe you, free is not a compromise. It is the correct choice. The free tools in our cheap tech stack cover most small businesses completely.
The monthly loop
Once a month, 30 minutes:
- Open Search Console, filter positions 8 to 25, export new queries.
- Add any that do not have a page to the sheet.
- Check the three pages you published last month. If any sit at position 11 to 20, add two sections answering People Also Ask questions from that SERP.
- Rewrite the title tag on any page with a click rate under 1 percent and a position under 10.
That loop compounds. Twelve months of it beats a one-off research project every time.
FAQ
Do I need a Google Ads account to use Keyword Planner?
Yes, but you do not need to spend anything. Create the account and skip through the campaign setup without launching a campaign, or pause it immediately. Keyword Planner then opens under Tools. The only difference between a free account and a spending account is that free accounts see volume as ranges such as 1K to 10K rather than exact figures. For choosing between keywords, ranges are enough.
Are free keyword research tools accurate?
No keyword tool is accurate, paid or free. Every number is an estimate built from clickstream data and modelling. What matters is consistency. If a tool tells you term A gets ten times the searches of term B, that ratio is usually right even when the raw numbers are off. Use volume to sort keywords into buckets, then let the search results page make the final call on whether to target it.
How many keywords should I target per page?
One primary intent, and as many phrasings of it as the page naturally covers. If two keywords return broadly the same top 10, Google sees them as the same question and you need one strong page, not two thin ones. If the results are different, they are separate pages. Run the check before you write, not after. It saves you from cannibalising your own rankings.
Want a second pair of eyes on what you already have? Our free audit will tell you which pages are closest to page one and what is holding them back.