Free Keyword Research Tools and Sources
The most useful free sources are not software at all. Several of them report what people actually typed rather than estimating how often. For a small business covering a handful of subjects, they will get you further than a subscription.
The Best Sources Cost Nothing
You arrived expecting a list of free products with their limits explained. You will not find one, because we name no keyword tools anywhere on this site. What follows instead is the sources we actually use. Most are not software and several are more trustworthy than anything a paid product reports.
Why we name nothing. A programme position.
Recommendations date, products change hands and naming one implies an endorsement we cannot maintain. Our tools guide sets out the full reasoning.
Why that costs you nothing here. The sources are better.
The list below is what an experienced person would actually reach for. Products come and go and these do not, which is the advantage of describing sources rather than software.
The distinction that matters. Records against estimates.
Some sources report what people genuinely typed. Others model how often. The first kind is evidence and the second is inference. Free sources are disproportionately the first kind.
What free sources will not give you. Volume.
No count of how many people searched. Block seven covers that properly. It matters less than most businesses expect.
Who this suits. Most of our clients.
A local business with a handful of services and a defined area. Below the scale where a subscription earns anything.
What it takes. An afternoon.
Not a licence and not a course. Attention, plus knowing what each source is actually telling you.
The Search Box Itself
Start typing a phrase and suggestions appear underneath. Those are drawn from searches people actually make, which means they are a record rather than a projection. No paid figure can claim that. This costs nothing.
What they prove. Real usage.
A suggestion exists because people type it. That is direct evidence of demand, without any modelling in between.
How to work them properly. Add letters.
Type your service, then a space, then each letter of the alphabet in turn. Each one produces a different set. The exercise surfaces phrasings no tool would have suggested.
The variation worth doing. Question words.
Type your service preceded by how, why, when, can and should. That produces the question demand around a subject in minutes.
What they will not tell you. How many.
Suggestions show that something is searched and never how often. Ordering may reflect popularity loosely and it is not a ranking you should rely on.
The local caution. They reflect you.
Suggestions are influenced by where you are and what you have searched before. Useful for local phrasings and misleading if you assume everybody sees yours.
What to do with them. Write them down.
Collect the phrasings, group them by situation and keep them. That is the raw material for the clustering exercise.
The Rest Of The Results Page
Once you have searched, the page itself is a research document. Most people read the first few results and leave. Everything below and around them is telling you what else people looked for. All of it is free.
Related searches. At the foot.
A block of other searches people made around the same subject. Reliable material and frequently the source of the best specific phrasings.
The question panel. Expandable.
Questions shown mid page that open when clicked. Opening one usually produces more, which makes it an unusually deep source for a free one.
Why the questions matter most. They reveal confusion.
What people do not understand about a subject is exactly what a good page should address. Our question keywords guide covers working them properly.
What the results themselves tell you. Intent.
Whether guides or service pages dominate tells you what kind of page belongs there. That single reading is worth more than most paid columns.
What the features tell you. More again.
A map appearing says location decides it. A direct answer at the top says people want a fact rather than a page, which changes whether the subject is worth a page at all.
The discipline. Look, do not just search.
Two minutes reading a results page properly is the highest value free research available. Almost nobody does it deliberately.
Your Own Search Reporting
Google Search Console is free, already connected on most sites and it reports what your own pages actually appeared for. That makes it the only source in this list that describes your particular business rather than the market generally.
What it shows. Real searches, real placements.
Which searches your site appeared for, how often and whether anybody chose you. Records rather than estimates, on your own site.
Its research value. Two findings.
Searches you never targeted, which reveal demand you did not know about. And searches where you appear without being found, which are the cheapest improvements available.
What it cannot do. Show what you never appeared for.
Entirely new subjects are invisible here, which is why the sources above still matter for discovery.
The limitation worth knowing. Rare searches withheld.
Uncommon queries are held back to protect privacy, which underrepresents exactly the small specific searches a local business depends on.
When it becomes your main source. Sooner than expected.
Once pages have been live a while it replaces external research almost entirely for planning what comes next.
Where the full treatment sits. Its own guide.
Using search reporting for research covers reading it properly for opportunities.
Interest Over Time
Google Trends shows how interest in a search rises and falls. It does something unusual for a free source. It publishes a relative index rather than a count, which means it never pretends to a precision it does not have.
What the numbers mean. Comparison, not counting.
The figures compare a search against its own peak over the period shown. They are not searches per month and they were never intended to be read that way.
Why that suits this programme. No invented precision.
A relative index makes a modest claim. Everything on this site treats volume figures carefully. A source that declines to state one is doing something right.
What it is genuinely good for. Seasons.
Seeing the shape of a year for a trade. When demand rises, when it falls and whether the pattern repeats, which decides when to publish rather than what.
The second use. Direction.
Whether interest in a subject is growing or fading over several years. Useful for deciding whether something is worth building at all.
The third use. Comparing phrasings.
Which of two ways of saying the same thing is used more. That settles vocabulary arguments quickly and without any subscription.
Its limit. Small subjects vanish.
Anything specific or local frequently shows nothing at all, for the same reason keyword tools fail there, which our local research guide covers.
The Sources That Are Not Tools At All
The most valuable research material a business owns is already inside the business. None of it requires an account and all of it is in the customer's own words. No competitor has access to any of it.
Whoever answers the phone. The best source there is.
The questions asked every week, written down as the caller said them. Demonstrated demand in real phrasing, for the price of a notepad.
Your email. Written evidence.
Enquiries already contain how people describe what they want, in writing. Reading a few months of them is a research exercise most businesses never do.
Your reviews. The words that persuade.
What customers say they were worried about and what convinced them. That vocabulary belongs on your pages.
Your staff. The complaints they hear.
Whoever is on site hears the things nobody puts in an email. What went wrong last time, what a previous firm did badly, what people wish they had known.
Community groups. Read, do not post.
Local groups where people ask for recommendations show exactly how a need gets described before anybody searches. Observe rather than participate.
The exercise worth doing. A fortnight of notes.
Every question asked, written down. That produces a content plan and it produces the vocabulary at the same time.
Why nobody does it. It looks unserious.
A notepad does not feel like research. It is the most reliable input in this entire cluster.
What Free Sources Cannot Give You
Free sources have real limits and pretending otherwise would leave you unprepared. Three things they do not provide. Knowing which of them you actually need decides whether a subscription is worth anything.
Volume figures. The obvious one.
No count of how many people search something. Paid products supply one, though it is an estimate rather than a count, which our tools guide covers at length.
Scale. The real limit.
Working through hundreds of subjects by hand is impractical. If that is your situation, a subscription buys time rather than insight and the time is worth buying.
Competitor visibility. The genuine gap.
Seeing which searches other sites appear for is not available free. That is a real capability and it is the strongest argument for paying.
What they give you instead. Certainty about words.
Every free source above reports what people typed. That is a different and frequently better thing than an estimate of how often they typed it.
How to decide. One question.
Are you researching a handful of subjects or hundreds. That answers the subscription question without any further analysis.
What we would do first. The free exercise.
Work through these sources before buying anything. If you hit a wall you will know exactly what you need a product to do, rather than buying one hopefully.
Where the rest sits. The hub.
All twenty-two guides are on the keyword research guide.
The best free
research source
is whoever
answers your
phone.
The questions customers ask every week are demonstrated demand, in their own words, at no cost. No external product has access to them. A fortnight of written notes produces a content plan and the vocabulary at the same time. It does not feel like research, which is precisely why almost nobody does it.
The free sources we actually use:
Several of these report what people genuinely typed. That is evidence rather than estimation. No paid figure can make the same claim.
Every guide.
One subject.
What the words mean, what the numbers in a keyword tool actually are, how to find the searches worth targeting, how to turn a list into a publishing plan and what to do when a page you relied on slips.
