What Is Keyword Research?

Keyword research is the process of finding, evaluating, and prioritizing the terms and phrases people actually use when searching, so that content and site structure can be built around real demand instead of guesswork. It sits at the start of most SEO and content work because it answers a question that’s easy to get wrong by intuition alone: what does your audience actually type or say when they’re looking for what you offer?

Done well, keyword research isn’t a one-time project. Search behavior shifts as products, terminology, and the search engines themselves change, so the process is better treated as an ongoing input to content planning rather than a single research phase that gets filed away.

The Core Process

Most keyword research follows a similar sequence, regardless of which tools are involved:

1. Start with seed terms. Identify the core topics, products, or services your content needs to cover, in plain language, before touching any tool.
2. Expand the list. Use keyword tools, autocomplete, “People also ask” boxes, competitor content, and customer language (support tickets, reviews, forum posts) to generate variations and related terms you wouldn’t have thought of on your own.
3. Cluster by intent. Group the expanded list by what the searcher is actually trying to do, not just by shared words. Terms that look similar can have very different intent (“running shoes” vs. “running shoes review” vs. “buy running shoes online”).
4. Filter by relevance, competition, and volume. Remove terms that don’t genuinely fit your content or business, then weigh the remaining terms against realistic ranking difficulty and the value of ranking, not volume alone.
5. Map keywords to content. Assign clusters to specific pages (existing or planned), checking for overlap with content you already have to avoid creating two pages that compete for the same intent.

Search Intent Should Drive Prioritization, Not Volume Alone

The most common mistake in keyword research is sorting a keyword list by search volume and working down the list. Volume tells you how often a term is searched; it tells you nothing about whether the people searching it are a good fit for what you’re offering, or whether they’re ready to act.

A keyword with lower volume but clear transactional or commercial-investigation intent is often worth more than a higher-volume informational term, especially for a business page rather than a blog post. As an illustrative example only, not a measured statistic: a thousand visits from a clearly transactional keyword could plausibly outperform ten thousand visits from a broadly informational keyword in terms of actual revenue impact, simply because the two audiences are at very different points in their decision process. That ratio is a hypothetical to illustrate the principle, not a benchmark from real data; the actual gap depends entirely on your specific business, offer, and audience, and should be measured in your own conversion data.

Before committing to a keyword, it’s also worth checking the search engine results page (SERP) itself. A term might have attractive volume and moderate keyword-difficulty scores in a tool, but if the results page is dominated by a SERP feature your content can’t realistically compete for (a shopping carousel, a video pack, an answer already fully resolved by an AI Overview), the practical opportunity may be smaller than the metrics suggest.

This check matters more than it used to for broad informational keywords specifically. Ahrefs’ own click-through-rate research found that when an AI Overview appears on an informational query, the click-through rate to the top-ranking organic result drops sharply compared to similar queries without an Overview (Ahrefs, “AI Overviews Reduce Clicks,” updated analysis). That doesn’t mean those keywords should be dropped from a content plan entirely, being cited as a source within an AI Overview can still drive some traffic and visibility, but it does mean a high-volume, broadly informational keyword may be worth meaningfully less organic traffic than its volume number alone suggests, and that gap should factor into prioritization alongside intent and difficulty.

Common Mistakes

Mistake	Why it hurts
Sorting purely by volume	Ignores whether the searcher's intent matches what you offer
Ignoring intent mismatch	A page built for the wrong intent rarely ranks well regardless of on-page optimization
Treating keyword-difficulty scores as absolute	Difficulty metrics are estimates from each tool's own model, not a guarantee of how hard ranking will actually be for your specific site
Ignoring SERP features and competition	A keyword can look easy in a tool and still be effectively unwinnable on the actual results page
One-time research	Search demand and phrasing shift; lists go stale
Keyword-only thinking, ignoring topical coverage	Modern ranking systems reward pages that thoroughly cover a topic, not pages stuffed with a target phrase

Tools Commonly Used for Keyword Research

No single tool covers every part of the process well. Most practitioners combine a few, depending on budget and need.

• Google Keyword Planner: Built for Google Ads, but commonly used for SEO too. Provides volume ranges and related-term suggestions pulled directly from Google’s own ad auction data, which makes it a useful sanity check even though its volume figures are bucketed rather than exact for lower-spend accounts.
• Ahrefs Keywords Explorer: A large keyword database with difficulty scoring, search volume, clustering, and SERP-level competitive analysis.
• Semrush Keyword Magic Tool: Similar in scope to Ahrefs, with strong keyword clustering and grouping features for organizing large keyword sets by topic.
• Moz Keyword Explorer: Keyword difficulty, volume, and SERP analysis, often used alongside Moz’s broader link and domain-authority metrics.
• Ubersuggest: A lower-cost, more accessible entry point for basic volume and related-keyword data, useful for smaller sites or early-stage research.
• Google Search Console: Not a discovery tool in the traditional sense, but invaluable for keyword research because it shows the actual queries already generating impressions and clicks for your site, including long-tail variations you wouldn’t have thought to search for.
• Google Trends: Shows relative interest in a term over time and by geography. Useful for spotting seasonality, rising topics, and regional differences in terminology, though it reports relative interest rather than absolute search volume.

Why the Same Keyword Shows Different Volume Numbers in Every Tool

Pull a single keyword in Google Keyword Planner, Ahrefs, and Semrush, and it’s common to see three different volume figures, sometimes by a wide margin. This isn’t a sign that one tool is “right” and the others are wrong; each pulls from a different mix of underlying data (Google’s own ad auction numbers, third-party clickstream panels, historical trend data) and fills the gaps with its own modeling, so the estimates diverge by design.

Google Keyword Planner has a specific quirk worth knowing for organic research: by default, it aggregates volume across phrases that contain the seed term rather than isolating the exact phrase. The number shown for “pool,” for example, can fold in searches for “swimming pool” and “pool tables” rather than reflecting that single query in isolation, which can make the figure look larger than the demand for the precise phrase actually is. Combined with the bucketed ranges shown to accounts without active ad spend, this means Keyword Planner numbers are best read as a rough order of magnitude, not a literal monthly search count.

The practical takeaway: use volume to compare keywords against each other within the same tool, and expect some disagreement when cross-checking a number in a second tool. Treat it as directional, not as a precise count of monthly searchers, and let Google Search Console’s own query data (actual impressions and clicks for terms your pages already rank for) act as the closest thing to ground truth you’ll get for free.

Reading Keyword-Difficulty Scores Correctly

Every major keyword tool publishes some version of a “keyword difficulty” score, and every tool calculates it differently, typically based on factors like the backlink profiles of currently ranking pages. These scores are useful for triage (comparing many keywords quickly) but should not be treated as a precise prediction of how hard it will be for your specific site to rank, since your own site’s authority, existing content, and topical relevance all affect the real difficulty in ways a generic score can’t fully capture. Cross-checking a promising keyword’s actual current SERP, not just its difficulty number, is a better final filter.

Where Keyword Research Fits Into Broader SEO

Keyword research output feeds directly into a few other parts of SEO work:

• Content planning: which topics get a dedicated page, and in what priority order.
• Site architecture and internal linking: how clusters of related keywords map to a logical structure of hub and supporting pages.
• On-page optimization: which terms and phrasing should naturally appear in titles, headings, and body copy, without drifting into keyword stuffing, which Google’s spam policies explicitly flag as a manipulative practice (see Google’s Spam Policies for Google Web Search).
• Measurement: tracking which queries actually drive traffic and conversions over time, which is itself a form of ongoing keyword research using Search Console data.

Frequently Asked Questions

Is keyword research a one-time task?
No. Search behavior, terminology, and competitive landscapes change over time, so keyword research works best as a recurring input to content planning rather than a single upfront project.

Which keyword research tool is best?
There isn’t a single best tool; it depends on budget and need. Google Keyword Planner and Search Console are free and directly tied to real Google data. Ahrefs and Semrush offer the most comprehensive databases and clustering features but come at a meaningful cost. Moz and Ubersuggest sit in between on price and feature depth.

Should I prioritize high-volume or high-intent keywords?
Intent should usually come first. A lower-volume keyword with clear transactional or commercial intent is often more valuable than a higher-volume informational term, particularly for pages meant to drive business outcomes rather than general traffic.

What’s the difference between keyword research and keyword strategy?
Keyword research is the discovery and analysis phase: finding and evaluating terms. Keyword strategy is what you do with that research: which terms map to which pages, how content is prioritized, and how the resulting structure avoids internal competition between pages.

How does AI-driven and conversational search change keyword research?
It shifts some of the focus from short, exact phrases toward the fuller questions and topics those phrases represent, since conversational interfaces like Google’s AI Mode accept longer, natural-language queries. The underlying research process, finding what people actually ask and clustering by intent, still applies; the keyword list just needs to account for longer, more conversational phrasing alongside shorter typed queries.

Do keyword-difficulty scores tell me how hard it will actually be to rank?
They’re a useful estimate for comparing many keywords quickly, but they’re calculated differently by every tool and don’t account for your specific site’s authority and existing content. Checking the actual current SERP for a keyword is a more reliable final check than the difficulty number alone.