Keyword Discovery: How to Find Keywords Worth Ranking For
SearchBlueprint · · 10 min read · scored 95/100
Keyword discovery means collecting the search terms your customers already use, then narrowing them to the ones you can realistically win. The raw material comes from four sources: seed terms, competitor gaps, your own ranking data, and customer language. The output is a short, prioritized list mapped to pages, not a spreadsheet.
What keyword discovery actually means
Keyword discovery sits at the front of every SEO campaign. Before you can rank, you need to know what people actually type, and guessing is how sites end up with pages nobody searches for. KeywordDiscovery, one of the oldest research tools in the industry, built its whole pitch on a single claim: over 80% of online transactions begin with a keyword search. Treat the exact figure as vendor marketing, but the direction is right. Search terms are recorded demand.
Discovery is a stage, not the whole job
Keyword research covers everything from finding terms to mapping them onto URLs. Discovery is the first half: casting a wide net and pulling in every candidate phrase. The second half is judgment. You filter for intent, difficulty, and business value, then assign each survivor to a page. Skip the second half and a huge export becomes decoration instead of a plan.
One phrase, several products
The phrase also names several products, which makes search results for it confusing. KeywordDiscovery.com is a standalone research database whose homepage currently shows a maintenance notice announcing a new service. Siteimprove sells a Keyword Discovery feature that lists the terms your site already ranks for, available on its SEO Advanced packages. BrightEdge folds discovery into its S3 platform, backed by a reverse index it says covers half a billion keywords. This guide covers the practice rather than any single product, and everything below works with free tools.
How keyword discovery works: four sources
Good lists come from four places. Work through all four, because each one surfaces terms the others miss.
1. Expand seed terms
Start with the words a customer would use to describe what you sell, one seed per product or service. Feed each seed into Google and harvest what comes back: autocomplete suggestions, which Google builds from real searches people run, the People Also Ask box, and the related searches at the bottom of the page. Then run the same seeds through a free keyword suggestions tool to pull long-tail variants in bulk. Long-tail phrases carry less volume but far clearer intent, and they are usually the terms a newer site can actually win.
2. Mine competitor gaps
A gap analysis lists the keywords your competitors rank for and you do not. BrightEdge draws a useful line here between gaps, meaning terms rivals already optimize for, and opportunities, meaning terms nobody in your market has claimed yet. You do not need an enterprise platform to apply the idea. Put a competitor's domain into any rank-visibility tool, export their top keywords, and subtract the ones you already cover. What remains is a map of proven demand in your market.
3. Read your own ranking data
The cheapest discoveries are queries you already rank for without trying. Google Search Console's Performance report lists every query that showed your pages, free. Sort by impressions and look for terms sitting just off page one: a small improvement to an existing page beats building a new one for a fresh term. Enterprise tools formalize exactly this. Siteimprove's version filters your existing rankings into Quick Wins, the keywords with the highest potential to bring traffic, High Impact, the big terms where you already rank well, and Winners and Losers, the keywords that jumped or dropped the most places over the last 30 days. You can replicate the whole idea with a Search Console export and an hour in a spreadsheet.
4. Capture customer language
Tools only report what other people already search. Your customers hand you phrasing no database has indexed yet: the words in reviews, support tickets, sales calls, and forum threads about your industry. When a phrase keeps appearing in tickets but shows zero volume in tools, the demand is usually just too new or too specific for the databases to have caught. Those phrases make strong FAQ and long-tail targets.
Free keyword discovery tools, compared
You can run a full discovery pass without paying for software. Two categories do the heavy lifting.
Google Keyword Planner
Google Keyword Planner is free once you set up a Google Ads account, switch it from Smart Mode to Expert Mode, and open Tools, then Planning. It suggests related terms, shows how search interest has trended over time, and estimates what advertisers bid per click. Google's own walkthrough uses "clothing store" as its example, with a suggested bid of $1.72. Remember what the tool was built for: ad budgets. Its volume data is directional and it groups close variants, so read its numbers as a ranking of relative demand, not exact counts.
Browser extensions that show volume as you search
Extensions put the data inside the results page, which suits discovery work. You see volumes while you explore autocomplete and competitor pages. Ratings below are as listed on the Chrome Web Store, which notes they are updated daily.
| Tool | What it shows | Where it works | Listed Chrome rating |
|---|---|---|---|
| Ubersuggest | Search volume and CPC for the query you just ran, free within daily data limits | Google, YouTube, and Amazon searches | 4.5 from about 1.1K ratings |
| Keyword Surfer | Keyword ideas and search volumes directly inside Google results | Google search pages | 4.2 |
| Keywords Everywhere | Volume, CPC, and competition data across 15+ sites | Google, Search Console, YouTube, Amazon, and more | 4.5 |
| Ahrefs SEO Toolbar | On-page checks, keyword data, and link metrics as an overlay | Any webpage and Google results | 4.4 |
| Detailed SEO Extension | One-click on-page SEO insights for the page you are viewing | Any webpage | 4.9 |
Ubersuggest deserves a specific note for discovery work: it surfaces suggested-keyword data on YouTube and Amazon searches as well as Google, and its listing reports around 600,000 users on the free tier.
When paid platforms earn their fee
Enterprise suites automate the grind: continuous gap detection, keyword buckets organized by brand, geography, or campaign, and prioritization by search volume, conversion potential, and striking distance, meaning terms close enough to page one to be worth a push. That framing comes from BrightEdge, and Siteimprove packages similar filters. Buy the automation when maintaining your lists costs real hours every week, not before. The thinking, which is the valuable part, stays the same at every price.
Filter the list: intent, winnability, value
Discovery gives you hundreds of candidates. Three filters, applied in order, get you to a shortlist worth acting on.
Match intent before anything else
Search intent is what the person wants to happen after they search. Rank a service page for an informational query and visitors bounce. Rank a blog post for a buying query and nobody converts. Read the current results for a keyword before targeting it, because the pages Google already rewards tell you what it believes the intent is.
| Intent signal | Example query | What the searcher wants | What to build |
|---|---|---|---|
| Informational: "how", "what", "why" | "how does keyword clustering work" | An explanation, not a pitch | A guide that teaches, then links to the relevant service page |
| Commercial: "best", "vs", "review" | "best rank tracker for agencies" | A comparison before choosing | A comparison page with honest trade-offs |
| Transactional: "pricing", "buy", "near me" | "seo audit pricing" | To act now | A service or pricing page with one clear next step |
| Navigational: brand names | "keyword planner login" | One specific website | Usually nothing; you cannot intercept it honestly |
Weigh winnability, not just a difficulty score
Tool difficulty scores compress link data into one number, and they miss what matters most: your site's standing on that specific topic. A newer site outranks nobody for a head term, yet it can win specific long-tail phrases within weeks. Prioritize striking-distance terms first, specific long-tail phrases second, and head terms last. The long tail is vastly larger than the handful of big terms every competitor fights over, so there is no shortage of specific phrases, only a shortage of patience to target them.
Put a value on the click
Volume tells you how many people search. It says nothing about what a visitor is worth. Cost-per-click estimates are a decent proxy: when Keyword Planner shows advertisers bidding $1.72 on a term, real businesses have decided that click pays for itself. A modest-volume term with buying intent will usually out-earn a high-volume term that mostly attracts students and job seekers.
Group discoveries into topics before you write
Raw lists produce cannibalization: two pages chasing the same term compete with each other, and neither ranks well. Cluster first. Group phrasings that mean the same thing, such as "keyword discovery tools" and "tools for discovering keywords", under one primary keyword, and give each cluster exactly one page. A free keyword clustering tool will sort a pasted list into topic groups in minutes. Then arrange the clusters into hubs and spokes: one comprehensive page for the big topic, supporting pages for each subtopic, linked both ways. Finish by writing the map down as keyword, URL, intent, and status. That one sheet turns discovery into a content calendar.
Where keyword discovery goes wrong
- Chasing volume without intent. The biggest number on the list is usually informational, fiercely competitive, and far from a sale. Filter by intent first and volume last.
- Ignoring your own data. Teams pay for tools while Search Console sits unopened, holding the exact queries where one improved page could move onto page one this month.
- Treating the export as the deliverable. A thousand keywords with no page map is trivia. Discovery only counts when every shortlisted term has a URL responsible for it.
- Writing for the tool instead of the reader. Google's Search Central documentation (developers.google.com/search) describes ranking systems built to reward helpful, people-first content. Pages stuffed with discovered phrases read badly and rarely hold their positions.
- Doing it once. Rankings move constantly; Siteimprove's Winners and Losers filter exists because 30 days is enough for meaningful movement. Revisit your list at least quarterly, and after any major product change.
FAQs
What is keyword discovery in SEO?
Keyword discovery is the stage of keyword research where you collect candidate search terms from tools, competitor rankings, your own Search Console data, and customer language, before filtering them down to targets. The goal is a complete picture of how people search for what you offer.
How is keyword discovery different from keyword research?
Discovery is the gathering half. Research is the whole process, including validating volume and intent, judging difficulty, and mapping each keyword to a page. Most people use the terms loosely, but the split matters: gathering is easy, and the filtering afterward is where campaigns are won.
Can I do keyword discovery for free?
Yes. Google Keyword Planner is free with a Google Ads account, Search Console lists the queries you already appear for, autocomplete and People Also Ask cost nothing, and extensions like Ubersuggest and Keyword Surfer show search volumes on free tiers with daily limits.
How many keywords should one page target?
One primary keyword per page, supported by its close variants used naturally. Two pages aimed at the same primary term compete against each other, so cluster similar phrasings and assign the whole cluster to a single URL.
What are striking-distance keywords?
Terms where you already rank just beyond page one. They are usually the fastest wins in a discovery pass, because the page already carries some authority and improving it takes far less effort than building a new one from zero. Search Console impressions data is the easiest place to find them.
Run one discovery pass this week
Pull your last three months of Search Console queries, pick the ten terms sitting just off page one, and map each one to the page that should own it. That single exercise teaches more about keyword discovery than any amount of tool shopping. Once your ten terms have pages, work through the full keyword research guide to take that shortlist the rest of the way, from validating volume and intent to a page-by-page plan.
Resources / References
Related reading: Advanced Keyword Research: Find Keywords Rivals Miss.
Sources
About the author
SearchBlueprint researched, wrote and scored this article with the content pipeline our customers run on their own sites. The score in the byline is what our content engine gave this page before it published. Nothing here is reviewed by hand, which is the point: it is the product's output, published where anyone can check it.