How to do keyword research — a step-by-step guide

Keyword research is the practice of discovering the search terms that your target audience uses when they look for products, services, information, or solutions related to your business. It involves identifying those terms, understanding how many people search for them, gauging how hard it would be to rank for them, and figuring out what the searcher actually wants when they type that query.

Think of it as market research for search engines. Just as you would study your customers before launching a product, keyword research helps you understand what your potential visitors are looking for before you create content. It is the foundation of SEO because everything else — on-page optimization, content creation, link building — depends on knowing which keywords to target in the first place. If you are new to SEO entirely, start with our SEO basics guide before diving into keyword research.

Without keyword research, SEO is a shot in the dark. With it, you have data-backed direction for every page you publish.

Seed keywords are the starting point. They are the broad, obvious terms that describe what your business does, what you sell, or what your website is about. You do not need any tools for this step — just your brain and a blank document.

Ask yourself: if someone needed what I offer and had to search for it on Google, what would they type? If you sell handmade candles, your seed keywords might be "handmade candles," "soy candles," "scented candles," "candle gifts." If you run a web design agency, they might be "web design," "website redesign," "small business website."

Talk to your customers. What words do they use to describe your product or service? Look at support tickets, sales calls, and reviews. The language your customers use is often different from the language you use internally. A software company might call their product a "customer relationship management platform," but their customers search for "simple CRM for small teams."

Write down everything that comes to mind. Do not filter or judge at this stage. You want a big, messy list of 20-50 seed terms. You will refine them later. The goal is to capture every angle someone might use to find what you offer.

Now take your seed keywords and type them into Google — but do not hit enter yet. Watch what Google suggests as you type. These autocomplete suggestions are based on actual searches that real people perform. They are free keyword ideas handed to you by Google itself.

For example, if you type "keyword research" into Google, you might see suggestions like "keyword research tool," "keyword research for SEO," "keyword research free," "keyword research template." Each of those is a real search query that people use, and each one could become a piece of content on your site.

Try adding different letters after your seed keyword. Type "keyword research a" and you might see "keyword research api" or "keyword research agency." Type "keyword research b" and you might see "keyword research beginner" or "keyword research best practices." This alphabet soup technique is simple but surprisingly effective at uncovering queries you would never have thought of on your own.

Next, actually search for your seed keywords and scroll down to the "People Also Ask" section. This box contains real questions that searchers ask about the topic. Click on a few of them and more questions appear. These are goldmines for content ideas — each question can become a section in a blog post or even a standalone article.

Also scroll to the very bottom of the search results page and look at the "Related searches" section. These are additional queries related to what you searched for. Add any relevant ones to your growing list.

You now have a long list of potential keywords. The next step is finding out how many people actually search for each one. This is called search volume — typically measured as the average number of monthly searches.

Google Keyword Planner is the most reliable free option. It is technically an advertising tool (you need a Google Ads account, but you do not need to run any ads), and it shows you search volume ranges for any keyword. The ranges are broad (like 1K-10K) unless you are running active campaigns, but they are enough to tell you whether a keyword gets meaningful traffic or almost none.

Ubersuggest offers a few free searches per day and shows estimated search volume, keyword difficulty, and related keyword ideas. It is a good starting point for beginners who want more specific numbers than Keyword Planner provides.

Google Trends does not show exact search volume, but it shows you whether interest in a topic is growing, declining, or seasonal. This is invaluable for spotting trends. A keyword with moderate volume but a strong upward trend might be more valuable than a high-volume keyword that is declining.

Other free tools worth trying include AnswerThePublic (visualizes questions people ask about a topic), KeywordTool.io (uses autocomplete data from multiple platforms), and Google Search Console (shows you queries your site already appears for, which you can then optimize for). We put together a full roundup of the best free SEO tools if you want to explore more options.

A common mistake is chasing only high-volume keywords. A keyword with 50,000 monthly searches sounds exciting, but if every major website in your industry is already ranking for it, your chances of reaching page one are slim. A keyword with 200 monthly searches that nobody else is targeting well could send you consistent, qualified traffic every month. Volume matters, but it is just one factor.

Keyword difficulty tells you how hard it will be to rank on the first page of Google for a given keyword. Most SEO tools assign a difficulty score from 0 to 100, where higher numbers mean more competition. But even without paid tools, you can assess difficulty manually.

Search the keyword and look at the results. Who is ranking on page one? If the top 10 results are all massive, authoritative websites — think Wikipedia, Forbes, Amazon, or industry leaders with thousands of backlinks — that keyword is going to be very difficult for a smaller site to crack. On the other hand, if you see smaller blogs, forum posts, or thin content ranking on page one, there is an opportunity.

Look at the quality of the top results. Are they well-written, comprehensive, and genuinely helpful? Or are they outdated, shallow, or poorly structured? If you can honestly create something significantly better than what currently ranks, you have a shot — even against bigger sites. Google ultimately wants to show the best result, and sometimes that best result does not exist yet.

Check how many backlinks the top results have. You can use free tools like Ahrefs' free backlink checker or Moz's Link Explorer to get a rough idea. If every page on page one has hundreds of backlinks, ranking there will require a significant link-building effort. If the top results have few or no backlinks, the barrier to entry is much lower.

As a general rule: if your site is new or small, start with keywords that have a difficulty score below 30 (in tools that provide one) or where you see smaller sites already ranking. You can always go after harder keywords later as your site builds authority.

Search intent is the reason behind a search query — what the person actually wants when they type something into Google. This is arguably the most important concept in keyword research, because if your content does not match the intent behind a keyword, it will not rank — no matter how well-written it is.

There are four main types of search intent:

Informational.The searcher wants to learn something. Examples: "what is keyword research," "how to change a tire," "why is the sky blue." These queries are best served by blog posts, guides, tutorials, and explainer articles.

Navigational.The searcher wants to go to a specific website or page. Examples: "Facebook login," "Spotify download," "sourcebeam pricing." Unless you are the brand being searched for, you generally cannot (and should not try to) rank for these.

Commercial.The searcher is researching before making a purchase. They are comparing options, reading reviews, or looking for the best solution. Examples: "best analytics tool for small business," "Mailchimp vs ConvertKit," "top running shoes 2026." These queries are served by comparison articles, reviews, and "best of" lists.

Transactional.The searcher is ready to buy or take action. Examples: "buy running shoes online," "sign up for email marketing tool," "hire web designer near me." These queries are best served by product pages, landing pages, and service pages.

The easiest way to determine intent is to search the keyword yourself and look at what Google shows. If the top results are all blog posts, Google has decided the intent is informational — so create a blog post. If the top results are product pages, the intent is transactional — a blog post will not rank there. Always match the format of your content to the intent behind the keyword.

You do not need a separate page for every keyword. Many keywords are variations of the same topic and can (and should) be covered by a single page. Grouping related keywords into clusters makes your content strategy more focused and prevents you from competing against yourself in search results.

For example, "how to do keyword research," "keyword research steps," "keyword research process," and "keyword research for beginners" all describe essentially the same thing. Rather than writing four separate articles, you write one comprehensive guide that naturally covers all of those variations.

A topic cluster typically has a pillar page — a broad, comprehensive piece of content on a main topic — surrounded by cluster pages that dive deeper into subtopics. The pillar page links to the cluster pages and vice versa. This internal linking structure helps Google understand the relationship between your pages and establishes your site as an authority on that topic.

To group your keywords, go through your list and look for overlapping themes. Sort them into buckets based on the core topic they relate to. Then decide which keywords become pillar pages and which become supporting articles. A simple spreadsheet with a "cluster" or "topic group" column works perfectly for this.

You cannot target every keyword at once. You need to decide where to start, and the best starting point is keywords that combine high intent with achievable difficulty. In other words, keywords where the person searching is likely to become a customer, and where you can realistically rank.

High intent + low difficulty = start here. These are your quick wins. They might not have the highest search volume, but they bring in visitors who are ready to take action. A keyword like "affordable project management tool for freelancers" has clear buying intent, and the long-tail nature means less competition.

High intent + high difficulty = plan for later. These are your long-term targets. You will need more content, more backlinks, and more domain authority to rank for them. Put them on your roadmap but do not start with them.

Low intent + low difficulty = fill gaps. Informational keywords that are easy to rank for can drive traffic to your site, build your authority, and introduce people to your brand — even if they do not convert immediately. These are good for building a content library over time.

Low intent + high difficulty = skip. There is little point in fighting for a keyword that is both hard to rank for and unlikely to bring you customers. Save your energy for more strategic targets.

A long-tail keyword is a longer, more specific search phrase. Instead of "analytics," it is "best analytics tool for small ecommerce store." Instead of "running shoes," it is "best running shoes for flat feet under $100."

Long-tail keywords have three advantages that make them perfect for smaller websites:

Less competition. The more specific the query, the fewer websites are trying to rank for it. "Analytics" is dominated by Google, Adobe, and every major tech company. "Best analytics tool for small ecommerce store" has far fewer competitors, and many of the pages that do rank might be mediocre — giving you room to create something better.

Higher intent. Someone searching "analytics" could want anything — a definition, a tool, a career in data science. Someone searching "best analytics tool for small ecommerce store" knows exactly what they want and is likely close to making a decision. These visitors are more likely to sign up, purchase, or take action.

Easier to match intent. With a broad keyword, it is hard to know what the searcher wants. With a long-tail keyword, the intent is right there in the query. This makes it much easier to create content that perfectly satisfies what the searcher is looking for.

A common misconception is that long-tail keywords are not worth pursuing because they have low search volume individually. But collectively, long-tail queries make up the vast majority of all searches. And a page that ranks well for one long-tail keyword usually ranks for dozens of related long-tail variations. Those small numbers add up to meaningful traffic.

One of the fastest ways to find good keywords is to look at what your competitors are already ranking for. Why start from scratch when someone else has already done the work of identifying valuable topics in your niche?

Identify your SEO competitors. These are not necessarily your business competitors — they are the websites that rank for the keywords you want to rank for. Search for a few of your target keywords and note which sites appear repeatedly. Those are your SEO competitors.

Analyze their top pages. Tools like Ubersuggest, Ahrefs (limited free version), or Semrush (limited free version) let you enter a competitor's domain and see which pages get the most organic traffic and which keywords they rank for. Look for keywords where they rank but you do not — these are your content gaps.

Look for weaknesses. Maybe a competitor ranks for a keyword with a thin, outdated page. Maybe they rank on page 2 for a valuable keyword, meaning the top results are not particularly strong. These are opportunities where you can create something better and potentially outrank them.

Do not just copy — improve. The goal is not to duplicate your competitor's content. It is to understand what topics are working in your niche and then create content that is more comprehensive, more up-to-date, or better targeted to search intent. If their article covers five tips, write one that covers ten — with more depth and better examples.

All of this research needs to live somewhere organized. A simple spreadsheet is the best tool for this. You do not need fancy software — Google Sheets or Excel works perfectly.

Create a spreadsheet with these columns:

Keyword — the exact search term. Search Volume — monthly searches (even a rough range is helpful). Difficulty — your assessment of how hard it will be to rank (low, medium, high, or a numeric score if you have one). Intent — informational, commercial, transactional, or navigational. Priority — high, medium, or low, based on the combination of intent, difficulty, and relevance to your business. Topic Cluster — which group this keyword belongs to. Status — not started, content drafted, published, or ranking.

Fill this spreadsheet with every keyword from your research. Sort by priority. This becomes your content roadmap — you always know what to write next and why. Update it regularly as you publish content and track how your pages perform for their target keywords.

Start small. Even 30-50 well-researched keywords in a spreadsheet is enough to fuel months of content creation. You do not need thousands of keywords to get started.

Chasing high-volume keywords. This is the most common mistake beginners make. They see a keyword with 100,000 monthly searches and build their entire strategy around it, ignoring the fact that every major website in their industry is already ranking for it. Volume without achievability is meaningless. A keyword you rank on page one for beats a keyword you rank on page eight for — even if the second one has ten times the volume.

Ignoring search intent. You find a keyword with decent volume and low difficulty, so you write a blog post. But when you search for it, the top results are all product pages. Google has decided this keyword has transactional intent, and your informational blog post will not rank — no matter how good it is. Always check intent before creating content.

Targeting keywords that are too broad. "Marketing" is not a useful keyword to target. It could mean a million things. "Email marketing for Shopify stores" is specific enough to create focused content for a defined audience.

Not updating your research. Search behavior changes over time. New competitors appear. Trends shift. A keyword that was low-difficulty six months ago might be competitive now. Revisit your keyword research regularly — do not treat it as a one-time exercise.

Optimizing for keywords you already rank for. If you already rank on page one for a keyword, your time is better spent targeting new keywords rather than over-optimizing existing pages. Check your analytics to see which keywords are already sending you traffic and focus your energy on gaps. If you use sourcebeam, you can quickly see which search terms are driving visitors to your site and identify where the opportunities are.

Creating separate pages for similar keywords. If two keywords have the same intent and the same top results, they should be covered by one page — not two. Creating multiple pages that target the same intent causes them to compete with each other in search results, a problem called keyword cannibalization.

Keyword research is not a one-time project. Your initial round of research gives you a foundation, but you should revisit and refresh it regularly. A quarterly cadence works well for most businesses.

Every quarter: review your keyword spreadsheet. Check which content you have published and how it is performing. Are you ranking for your target keywords? Are there new keywords you should add? Has the competitive landscape changed? Use Google Search Console to discover queries your site is appearing for that you did not originally target — these often reveal new opportunities.

When launching something new: if you are launching a new product, entering a new market, or starting a new content initiative, do a fresh round of keyword research specific to that area. Do not assume the keywords from your last research still apply.

When traffic drops unexpectedly: if organic traffic to a page declines, check whether the keyword landscape has changed. Maybe new competitors have entered the space, maybe search intent has shifted, or maybe Google has updated its algorithm in a way that affects your content. A quick keyword check can reveal what happened and how to respond.

The best keyword researchers treat it as an ongoing practice, not a one-time task. Each quarter, you learn more about what works, what does not, and where the next opportunities are.

Keyword research is only valuable if you turn it into action. Here is how to go from a spreadsheet of keywords to a content plan that drives traffic.

Start with your highest-priority keywords. Look at your spreadsheet and pick the keywords you marked as high priority — the ones with strong intent and achievable difficulty. These are your first pieces of content.

Map keywords to content types. Informational keywords become blog posts, guides, and tutorials — our guide on how to write blog posts that rank on Googlecovers this in depth. Commercial keywords become comparison pages and "best of" articles. Transactional keywords map to product pages, landing pages, and service pages. Let the intent dictate the format.

Create a content calendar. Decide on a realistic publishing cadence — maybe one article per week, maybe two per month. Assign keywords to dates. Having a schedule prevents you from doing all the research and then never actually writing anything, which is a surprisingly common outcome.

Outline before you write. For each piece of content, search the target keyword and study the top results. What do they cover? What do they miss? Use this to create an outline that is at least as comprehensive as the best existing result — and ideally better. Include your target keyword and related keywords naturally in headings and body text, but write for humans first. Do not force keywords where they do not belong.

Publish, then optimize. After publishing, monitor how the page performs in Google Search Console. After a few weeks, you will see which queries it is appearing for. If it is ranking on page 2 or 3 for your target keyword, look for ways to improve it — better content, more internal links, a more compelling title tag. SEO content is never "done" — it is a living asset that you refine over time based on real performance data.

The cycle is simple: research keywords, create content, measure results, refine, and repeat. Over time, this process compounds. Each piece of content builds your site's authority, which makes it easier to rank for the next keyword, which brings more traffic, which attracts more backlinks. The hardest part is getting started, but once the flywheel is spinning, it builds momentum on its own.