High-quality content and pages represent the base of every website. They are one of the most important factors that are driving traffic. Regardless of the quality, they will not rank high without using the right keywords. For that reason, we prepared a list showcasing some of the best free keyword research tools.
We’ve also made a list of the best paid keyword research tools in the past. But, we understand that many of you are SEOs on a budget. Free doesn’t always have to mean low-quality, as the following tools will prove. They are all performing great, and have the potential to boost your rankings considerably.
Updated for 2026. This article started life as a 2019 round-up, and both the tools and the landscape around them have moved on. Google renamed AdWords to Google Ads back in 2018, Keyword Planner now shows most accounts wide volume ranges instead of exact numbers, Universal Analytics has been retired in favour of GA4, and AI Overviews answer a growing share of queries directly on the results page. Two of the three original picks have also stopped being dependable: the Hall Analysis page that hosted the permutation generator is gone and the domain now redirects elsewhere, and Keyword Sh!tter has been unreachable for long stretches. We have kept the original write-ups below, because the methods behind them are still exactly right, and added the checks, the step-by-step workflow and the free sources we would actually reach for today.
What makes a free keyword tool worth your time
Every free tool trades something away — usually volume accuracy, export limits or freshness. Before you build a content plan on top of one, run it through these six checks:
- Where the data comes from. Autocomplete scraping, clickstream panels and Google’s own advertising platform produce very different numbers. Know which one you are reading.
- Ideas versus demand. Some tools only list phrases. Others tell you how often people actually search for them. You need both, and rarely from the same free source.
- Country and language coverage. A tool that is dependable for the US can be thin for Romania, Hungary or the Nordics. Always set the market before you judge the output.
- Export options. If you cannot pull a CSV out, you cannot cluster, deduplicate or hand the list to a writer.
- Daily limits. Most free tiers cap how many searches you can run per day. Plan a research session around that cap instead of hitting it halfway through.
- Stability. Free scrapers appear and disappear. Never let one of them be the only step in your process.
Google Trends
A step up from the classic keyword planner, the Google Trends app represents one of the best free keyword research tools out there. Many other free tools just give you raw keywords to choose from. Google Trends provides you with essential data about the popularity of a keyword.
The tool is great for planning your content ahead. By inserting a keyword you can generate periodic graphs illustrating the popularity trends the keyword experiences. Such statistics are great for content creation. Many keywords are seasonal. They will spike in popularity each year around the same time. You can create specific content around them in that period.
This strategy is great if you want to capitalize off seasonal audiences and traffic. The Google Trends tool doesn’t stop here. Besides the popularity analysis, the tool also provides you with other related queries. The tool only suggests keywords that are popular at the time of your search. This way you can get both content ideas but also create articles that are in demand.
How to use it, step by step
- Open Google Trends and set the country before anything else. Worldwide data will hide national patterns almost every time.
- Set the timeframe to the past five years. A five-year window shows whether a topic is seasonal, growing or fading; a twelve-month window often cannot tell the difference.
- Add up to four comparison terms. Comparing your candidate against a term you already rank for is the fastest way to get a sense of scale.
- Scroll to Related queries and switch between Top and Rising. Top shows established demand, Rising shows what is gaining ground — that is where unclaimed content opportunities live.
- Check Interest by subregion if you sell locally. Demand is rarely spread evenly across a country.
- Export the CSV and keep it with your brief so the writer can see the same timing you did.
A worked example. Take a query with an obvious season, such as ski accommodation. On a five-year view the curve climbs from late autumn, peaks in the depths of winter and collapses in spring, repeating every year. The publishing rule that follows is simple: put the page live six to eight weeks before the climb starts, so it has time to be crawled, indexed and to pick up a few links before demand arrives. Publishing at the peak means competing with pages that have a season of behavioural data behind them.
Two caveats worth knowing. First, Trends does not report search volume — the numbers on the chart are a relative index from 0 to 100, where 100 is the highest point in the range you selected. A term can score 100 and still be searched only a handful of times a month. Second, low-volume queries produce noisy or empty graphs, so use it for topics and categories rather than for long-tail phrases. Where a word is ambiguous, use the category filter and the search-type selector (Web, Image, News, Shopping, YouTube) to separate the meanings before you draw a conclusion.
AdWord & SEO Keyword Permutation Generator
This tool is quite simple in concept but very effective when it comes to keyword research. You can input multiple words or lists of words and the tool will generate all possible permutations. Depending on the number of words you introduce, you can get thousands of new keyword ideas.
Not all keywords will be ideal for targeting. Due to the sheer amount of keywords generated, some will have low search volume and popularity. The tool provides no data or charts about search volumes. You will have to use a different tool to acquire that data and decide on your best keyword.
Even if the process might take a bit of time, this tool can be very effective for local SEO. If you have a list of cities or countries you want to target, you can get all possible keywords for those areas.
Two notes before you go looking for it. The name is a fossil: permutation generators of this kind still carry the legacy “AdWord” label because they were built when Google’s advertising platform was called AdWords, which was renamed Google Ads in 2018. More importantly, the Hall Analysis page that hosted this particular generator is no longer online and the domain now redirects to an unrelated site, so treat the tool as a technique rather than a destination. The maths is not tied to any one product — combining word lists is exactly how paid search teams still build campaign keyword sets, it works just as well for organic landing pages, and you can reproduce it in a spreadsheet in a couple of minutes or in any of the permutation generators that are still running.
A worked example for local pages
Build three lists and let the tool multiply them:
| List | Example entries | What it contributes |
|---|---|---|
| 1. Core service | link building, guest posting, digital PR, outreach | What you actually sell |
| 2. Commercial modifier | agency, services, company, pricing | Buying intent |
| 3. Location | Bucharest, Cluj, Romania, Europe | Geographic targeting |
Four services multiplied by four modifiers multiplied by four locations gives you 64 candidate phrases from one click. Most will be junk. That is fine — the generator’s job is coverage, not judgement. Your job is the pass that follows: deduplicate, delete the combinations no human would type, then push what survives into a tool that has demand data (Keyword Planner, Search Console or a free tier) to see which ones are worth a page. As a rule of thumb, if a phrase would look strange spoken out loud, it does not deserve a heading.
Keyword Sh!tter
Although its name might not suggest the best things, the tool is nothing but useful. Different from both tools mentioned above, this one gives you a large number of suggestions based on your main word.
If you only want to find keyphrases including a specific word then this tool is for you. The design is utterly simplistic, and so are the stats but, the generated keywords are none other than great. It has a positive and negative filter in case you want to include or exclude all of the queries containing a specific word.
What it does under the hood is worth understanding, because it explains both the strength and the limit. Tools in this family query search autocomplete repeatedly, adding letters and words to your seed and collecting whatever comes back. The output is therefore real phrasing that real people type, including the clumsy, misspelled and unexpectedly specific versions that polished keyword databases tend to smooth away. What you do not get is volume, difficulty or intent — only language.
Use the filters from the first minute rather than cleaning up afterwards. With a seed like “link building”, a negative filter for free, jobs, salary and course strips out the job seekers and students, while a positive filter for agency and services keeps only the commercially relevant tail. Let it run for a minute or two, stop it, export, then deduplicate against keywords you already rank for so you do not commission a page that competes with one of your own.
Be aware that unofficial scrapers like this one depend on access they do not control, and this one in particular has been down or unreachable for long stretches, so do not build a process that assumes it will load. When it does not, you can reproduce most of the value by hand: type your seed into Google and read the autocomplete suggestions, then work through the People Also Ask box and the related searches at the foot of the results page. The free keyword generators offered by the big SEO platforms, and the keyword research report inside Bing Webmaster Tools, cover the same ground with a cleaner interface and are actually maintained.
Three more free sources we would add today
Google Search Console
If your site already has any traffic, Search Console is the most valuable free keyword source in existence, because it reports the queries Google actually showed you for, with clicks, impressions, click-through rate and average position, across the previous sixteen months. Nothing you buy is more relevant to your own site than that. The quickest win is the striking-distance report: in the Performance report, filter to queries with an average position between roughly eight and twenty, sort by impressions, and you have a list of terms where you are already visible but below the fold of attention. Improving an existing page for those queries is almost always cheaper than writing a new one.
The results page itself
Autocomplete, People Also Ask, related searches and the discussions and forums block are keyword research with zero setup and zero cost, and they reflect the live index rather than a database refreshed monthly. They also tell you something no keyword tool will: what format Google currently rewards for that query. If the first screen is filled with videos, comparison tables or forum threads, that is the brief.
Free tiers of the paid platforms
Most major SEO suites now give away a usable slice of their index. Ahrefs Webmaster Tools is free for verified site owners and reports backlink and keyword data for domains you control; the free keyword generators run by the big platforms return a capped list of ideas with difficulty estimates; and Bing Webmaster Tools includes a keyword research report that is badly underused. Combined with Keyword Planner — still free with a Google Ads account, though it now buckets volumes into broad ranges unless you are spending — that is already more data than most small sites will ever act on.
A free keyword research workflow that ships pages
- Write ten seeds by hand. Your services, your categories, the words clients use in emails. No tool invents these for you.
- Expand. Run the seeds through an autocomplete tool and a permutation generator. Aim for breadth; ignore quality for now.
- Deduplicate and clean. Lowercase everything, remove duplicates, strip the nonsense combinations.
- Cluster by intent. Group phrases that would be satisfied by the same page. One cluster equals one URL, not one keyword equals one URL.
- Check the SERP for the head term of each cluster. If the results are all product pages, do not write a blog post.
- Add timing. Run the head terms through Google Trends and note which clusters are seasonal and when they need to be live.
- Map to existing pages first. Search your own site before commissioning anything. Updating usually beats creating.
- Measure. Track impressions and average position per cluster in Search Console, and behaviour and conversions in GA4. Revisit after a full quarter, not after a fortnight.
How to prioritise when you have no exact volumes
Working without paid volume data is a constraint, not a blocker. Each free source answers one specific question, and stacked together they answer enough of them to make a decision:
| Question | Signal to read | Free source |
|---|---|---|
| When should this be published? | Seasonal shape over five years | Google Trends |
| How do people phrase it? | Autocomplete tail | Autocomplete-based tools, Google suggest |
| Is there buying intent? | Number of ads and shopping units on the SERP | Manual SERP check |
| Am I already close? | Impressions with average position 8–20 | Search Console |
| How hard is it? | Who ranks, and whether their pages are strong | Manual SERP check |
| Roughly how big is it? | Volume range bucket | Keyword Planner |
Common mistakes to avoid
- Chasing volume alone. A broad term with heavy demand and no intent will produce traffic that never enquires.
- Building a page per keyword. Near-identical phrases belong on the same URL. Splitting them creates cannibalisation you will have to merge later.
- Ignoring the SERP. The current results are Google telling you what it considers a correct answer. Argue with it and you lose.
- Researching once. Language shifts, products change, competitors publish. A quarterly review is the minimum.
- Treating relative indexes as volumes. Trends scores and difficulty scores are comparisons, not counts.
Frequently asked questions
Are free keyword tools accurate enough to plan a site around?
For direction, yes. For precision, no. Free tools are reliable at telling you which topics exist, how people phrase them and when demand arrives. They are unreliable at telling you a keyword gets a specific number of searches per month — and so, incidentally, are the paid ones, which model volumes rather than measure them. Decide with relative comparisons, not absolute figures.
Is Keyword Planner still worth using in 2026?
Yes, with expectations set correctly. It needs a Google Ads account, and accounts that are not spending see volumes grouped into wide ranges rather than exact numbers. It also merges close variants, so several distinct phrases can show the same figure. Use it to sort keywords into rough size buckets and to read the competition and bid columns as a proxy for commercial value, then get your nuance elsewhere.
How many keywords should one page target?
One intent, many phrasings. A well-built page can rank for dozens or hundreds of variations of the same question without any of them being written into the copy verbatim. Pick a primary phrase for the title and H1, cover the related questions honestly in the body, and let the variations follow.
How often should keyword research be repeated?
A full exercise once or twice a year is usually enough for a stable business. Between those, a monthly fifteen-minute pass over the Search Console query report catches new terms you have started appearing for, which is where the cheapest wins hide. Seasonal sectors should review a quarter before each peak, not during it.
Do AI Overviews make keyword research pointless?
No, but they change what you research for. Simple factual queries increasingly get answered on the results page, so pages built only to define a term earn fewer clicks than they once did. Queries where someone is choosing, comparing, buying or hiring still send traffic, and generative answers are themselves assembled from pages that cover a topic properly. Keyword research is now less about capturing every definition and more about identifying the questions where a human still needs to visit a site.
Conclusion
Keyword research is an essential task both for content creation but also for ranking. There might be a couple of better paid-tools than the ones mentioned above. However, these three are the best free keyword research tools right now and can be very effective if used correctly.
That verdict is the one we published in 2019. The reasoning behind it still holds even though two of the three tools have since become unreliable, because what free research actually gives you is timing, broad coverage and real phrasing, and every one of those is still obtainable without paying. The point at which you outgrow them is fairly easy to recognise: you are tracking hundreds of keywords across several markets, you need historical position data, or you are spending more time exporting and merging spreadsheets than deciding what to publish. At that stage our round-up of paid keyword research tools covers what the money actually buys. Until then, Google Trends for timing, a permutation generator for coverage, an autocomplete tool for phrasing and Search Console for feedback will take a site a very long way without a single invoice.