How Do I Find Keywords a New Website Can Rank For?

A new website has no history with Google: no backlinks worth much, no years of trust signals, and often no existing traffic to analyze. If you try to compete for the same broad terms an established competitor already ranks for, you're unlikely to see movement for a long time. The real answer to "how do I find keywords a new website can rank for" isn't a magic keyword list — it's a method for finding the specific searches where a brand-new site has a realistic shot, and then narrowing to those first.

This article walks through that method: where to source keyword ideas without paid tools, how to judge which ones are actually reachable for a site with no authority yet, and how to turn that shortlist into a working content plan.

Why "high volume" keywords are the wrong starting point

When people research keywords for the first time, the instinct is to look for the terms with the biggest search numbers. For a new site, this is usually the wrong filter. High-volume, generic terms are almost always dominated by sites that have been publishing and earning links for years. A new domain competing head-on for those terms is competing against pages Google already trusts — not because the new content is worse, but because trust is a track record, not a toggle.

The more useful question for a new site isn't "how many people search this?" It's "how many sites like mine currently rank on page one for this, and could a genuinely useful, specific answer plausibly outrank the weakest of them?" That reframing is what separates keyword research that produces early wins from keyword research that produces a list nobody ever ranks for.

Start with what your actual customers ask, not what a tool suggests

Before opening any keyword tool, write down the real questions your customers ask you — in emails, in calls, at the counter, in support tickets. These are keywords in disguise. A dentist's front desk hears "how long does teeth whitening last," "what counts as a dental emergency," and "what's included in the cost of an implant" every week. Those exact phrases are searches. This source is free, requires no tool, and reflects genuine intent because it comes from people who already tried to become customers.

This matters more for a new site than for an established one, because a new site's advantage isn't authority — it's specificity. You can answer a narrow question more thoroughly than a large competitor who's spread across a hundred topics.

Free ways to source keyword ideas without paid tools

You don't need a paid subscription to build a solid first keyword list. Several free, publicly available sources work well together:

Search autocomplete. Type your core topic into a search engine and note what it suggests as you type. These suggestions are drawn from real queries other people have typed.
"People also ask" and related searches. The boxes that appear on a results page after you search a term show closely related questions and phrases people search around that topic.
Community forums and Q&A sites. Places where your customers already ask questions in their own words — using the exact phrasing they'd use in a search bar, not the phrasing a marketer would choose.
Your own site search and support history. If your site has any search bar or your team logs customer questions, that data is a direct line to real phrasing.
Competitor blog and FAQ titles. Not to copy the content, but to see which questions a competitor decided were worth answering — a signal about what their audience searches for.

Combine several of these sources rather than relying on one. Autocomplete alone tends to surface obvious, competitive terms; forums and support logs tend to surface the longer, more specific ones that are easier for a new site to compete on.

Judge reachability, not just volume

Once you have a list of candidate keywords, the next step is deciding which ones a new site can realistically target first. Since no independent ranking-difficulty data can be cited here without a paid tool, the practical approach is a manual check you can run yourself:

Search the exact phrase in an incognito or private browser window (to avoid personalized results).
Look at who ranks on page one. Are they large, well-known publishers and national brands, or are they small local businesses, forum threads, and thin pages?
Open the top three results. Do they actually answer the question well, or are they thin, outdated, or only tangentially related? A weak page one is an opening for a new site with a genuinely better answer.
Note the phrasing. Long, specific phrases ("what to do about a chipped tooth before you can see a dentist") are typically less contested than short, broad ones ("dental care").

Search results shift, and what looks weak today can change.

Group keywords into topics your customers actually have

Once you have a working list, group similar phrases together by the underlying question rather than treating each keyword as a separate article. "How long does teeth whitening last," "does teeth whitening damage enamel," and "how often can you whiten your teeth" are different phrasings of overlapping curiosity about the same decision — whether and how to whiten teeth safely. One well-structured article can often address a cluster like this, while a second article covers a genuinely distinct decision, like choosing between Invisalign and braces.

This kind of grouping does two things for a new site. First, it stops you publishing five thin articles that all compete with each other for the same intent. Second, it means each article can be more thorough, which matters more for a new domain trying to earn trust than for an established one.

A simple framework for prioritizing what to write first

For a new site, this is a reasonable order of operations — treat it as a starting recommendation to adapt, not a fixed rule:

Tier 1 — Long-tail, specific, low apparent competition. Questions your customers actually ask, phrased naturally, where page one shows thin or off-topic results. These are your best early bets.
Tier 2 — Moderate specificity, mixed competition. Slightly broader questions where page one has a mix of strong and weak results. Worth targeting once Tier 1 topics are published and you have some initial traction.
Tier 3 — Broad, high-volume, high-competition terms. Save these. They're worth targeting eventually, but rarely first, since a brand-new domain is unlikely to outrank established, trusted pages for them early on.

Write down which tier each keyword falls into before you commit to a content calendar. It's a lightweight exercise, but it prevents the common mistake of spending the first months of a new site's life writing content that has almost no realistic path to ranking.

What to check before you commit to a topic

Before finalizing a topic, confirm three things:

You can answer it from facts you actually know. A new site's strongest asset is specific, accurate information about the exact business — its services, pricing, process, and policies. Generic advice that could apply to any business in the category won't out-compete a page one that's dominated by strong competitors.
The phrasing matches how customers actually search, not how you'd describe the topic internally. A clinic might call it "restorative treatment planning"; a patient searches "what to do about a chipped tooth."
The page one you observed genuinely has room. If every result is a well-established, comprehensive resource, that's a signal to deprioritize the term for now rather than force it.

How this connects to ongoing keyword research, not a one-time list

Keyword research for a new site isn't a task you finish once. As pages start to rank — even on page two or three — the site's own Google Search Console data becomes a new, better source of keyword ideas: the actual queries people used to find pages that are already live. Revisiting that data periodically, alongside the same customer-question and search-autocomplete sources described above, keeps the keyword list current as the site earns more authority and can reasonably compete for slightly broader terms.

Where Pentra fits into this process

Everything above is a manual method you can run yourself with no paid subscription. If you'd rather not do it by hand for every article, this is close to what Pentra automates for a new website: it reads the site to build a business profile the owner confirms, researches the specific searches customers are making that the site doesn't yet answer, and writes each article from those confirmed facts rather than generic claims. Every draft goes through an independent fact check before it's allowed to publish, and once an article is live, Pentra opens the real page to confirm it published correctly — then reports the resulting clicks and positions from the site's own Google Search Console data. It also revisits pages sitting close to page one so a new site's early wins keep improving rather than stalling. The owner sets the pace, from Autopilot to Review First, at anywhere from 1 to 21 articles a week, and a free plan covers 3 articles a month with no credit card required.

Try Pentra

Frequently Asked Questions

Do I need a paid keyword tool to research keywords for a new website? No. Search autocomplete, related-search boxes, community forums, and your own customer questions are all free sources that can produce a usable first keyword list, especially for a new site targeting specific, lower-competition phrases.

How many keywords should a new website target before publishing? There's no fixed number that fits every site. A more useful approach is grouping your list into topic clusters (see the grouping section above) and starting with the cluster that scores highest on the specificity and weak-competition checks, rather than aiming for a target count.

Should a new website ever target broad, high-volume keywords? It can, eventually, but broad terms are usually the hardest for a domain with no track record to win early. Treat them as a longer-term goal once more specific topics have started to rank and the site has built some initial traction.

How Do I Find Keywords a New Website Can Rank For?