How to Do Keyword Research for a New Website: Where to Start

Most new websites make the same mistake before they publish a single page: they guess at what their audience searches for, rather than looking it up. The result is a content plan built on assumptions — and months of effort that generates almost no organic traffic. Keyword research changes that. It turns a blank slate into a structured content roadmap with a realistic path to ranking.

This article explains exactly how to do keyword research for a new website — including the steps most guides skip, such as how to prioritise when you have no authority yet and how to avoid targeting terms you cannot realistically compete for in the short term.

If you're looking for expert help in this area, explore how Indexed's SEO audit and strategy can drive measurable results for your business.

Why New Websites Need a Different Keyword Strategy

Keyword research guides written for established sites tend to focus on volume and competition metrics. That approach is sound when your domain already carries authority — but it is actively misleading for a new website with zero backlinks, no indexed content, and no ranking history.

Google and other search engines use a combination of relevance signals and authority signals to determine which pages to surface. A new site has relevance (you can create relevant content immediately) but no authority. That asymmetry shapes every keyword decision you make in the first 12 months.

What Domain Authority Means in Practice

Domain authority is not a single score — it is shorthand for how much trust a search engine extends to your domain based on the quantity and quality of sites linking to it. A site launched last month competing for a keyword dominated by ten-year-old domains with hundreds of backlinks is unlikely to rank in the near term, regardless of content quality. The keyword selection process must account for this from the outset.

The Realistic Timeframe

Expect six to twelve months before carefully chosen keywords produce consistent organic traffic for a new site. That is not a reason to delay — it is a reason to start with a clear plan rather than hoping broad, high-volume terms will somehow yield early wins.

Build Your Seed Keyword List Before Opening Any Tool

The most important step in the keyword research process happens before you open a tool. You need a seed list — a set of broad topics that define the core subject matter of your site. Tools amplify thinking; they do not replace it.

How to Generate Seeds

Start by writing down every service, product, or topic your site covers. Then ask: what would someone type into Google if they needed what we offer but did not know our name? That question shifts your thinking from brand-out to audience-in — which is the only perspective that matters for organic search.

Practical sources for seed keywords include:

• Sales conversations — the exact language prospects use when describing a problem
• Customer support tickets — questions that appear repeatedly
• Competitor site navigation — the categories and services they organise around
• Amazon, Reddit, and niche forums — community language often surfaces terms that formal tools miss

Aim for 15 to 30 seed terms. You are not trying to be exhaustive at this stage — you are building the starting points that tools will expand.

How to Do Keyword Research Using the Right Tools

Once you have seeds, the next step is expansion and qualification. Several tools serve this well at different budget points.

Free Starting Points

Google Search Console is free and, once your site has any indexed pages, will show you what queries are already generating impressions — even before clicks arrive. For a brand-new site with no content yet, pair it with Google Keyword Planner (free with a Google Ads account) to get volume estimates for your seeds and related terms.

Google Autocomplete and People Also Ask are underused by new site owners. Type a seed term into Google and note every autocomplete suggestion — those are real queries typed by real users. The People Also Ask boxes reveal related questions that often make excellent long-form content targets.

Paid Tools Worth the Investment

If budget allows, Ahrefs and Semrush both offer keyword difficulty scores alongside volume estimates. For a new website, keyword difficulty is the metric that matters most. Filter for low-difficulty terms first — these are the keywords where a well-structured new page can realistically earn a position within months rather than years.

A practical workflow: enter each seed term into your chosen tool, export the keyword suggestions, then filter to show only those with a difficulty score below 20 (on Ahrefs' scale) or equivalent. You will likely still have hundreds of candidates — which brings you to the prioritisation step.

Prioritise by Commercial Proximity, Not Just Volume

This is the section most keyword research guides either skip or compress into a single sentence. For a new website, prioritisation is arguably more important than discovery — because publishing content on the wrong keywords wastes the limited authority budget you are building month by month.

The Four Intent Types and What They Mean for New Sites

Search intent is conventionally divided into four categories:

• Informational — the user wants to learn something (e.g. "how does VAT work for freelancers")
• Navigational — the user wants a specific site or brand
• Commercial investigation — the user is comparing options before buying (e.g. "best project management software for agencies")
• Transactional — the user is ready to act (e.g. "buy project management software")

New sites often make one of two errors: they target only transactional keywords (too competitive, too little content context) or only informational keywords (low commercial proximity, slow revenue impact). The right balance depends on your business model, but a useful starting point is to map your keyword list so that roughly 60% is informational, 30% is commercial investigation, and 10% is transactional. This builds topical authority while maintaining a path toward conversion traffic.

A Concrete Example

Suppose you are launching a site for a UK-based HR consultancy. The term "HR software" has high volume and high difficulty — it is dominated by software vendors with enormous domain authority. "HR policies for small businesses UK" has lower volume but far lower competition and directly matches the intent of a small business owner who might become a client. That second term goes to the top of your list.

Map Keywords to a Content Architecture Before You Write Anything

Individual keywords are not a content plan. Before you publish, map your prioritised keyword list to a site structure — this is where keyword research becomes genuinely strategic rather than just a list of terms.

Pillar and Cluster Logic

Group your keywords by topic. Each major topic becomes a pillar page — a comprehensive piece that targets a broader term. Supporting cluster pages target more specific, lower-difficulty variations and link back to the pillar. This architecture signals topical depth to search engines and concentrates internal link equity on your most important pages.

For the HR consultancy example: a pillar on "HR support for small businesses" could be supported by clusters on "employee handbook templates UK," "redundancy process UK small business," and "HR compliance checklist." Each cluster page addresses a specific long-tail query while reinforcing the pillar's authority.

Avoiding Cannibalisation from Day One

Keyword cannibalisation — where multiple pages compete for the same term — is far easier to prevent than to fix. When mapping your keywords, ensure each term is assigned to exactly one page. If two terms are closely related (e.g. "HR policies UK" and "HR policy templates UK"), decide whether they belong on the same page or warrant separate treatment based on whether the search intent is genuinely different.

Validate Keywords Against the Search Results Page Before You Commit

No tool replaces looking at the actual search results for a keyword you are considering targeting. Open an incognito browser window and search for each of your priority terms. What you see tells you several things that metrics alone cannot.

First, look at who is ranking. If positions one through ten are occupied exclusively by well-funded national brands or government sites, that is a signal to reconsider — regardless of what a difficulty score says. Second, look at the content format. If results are dominated by video, a written article may not rank well regardless of quality. If results show a mix of blog posts and tool pages, that is a more accessible environment for a new site.

Third, look at the freshness of the ranking content. A results page full of articles published in 2019 with no recent updates is an opportunity: a well-researched, current piece can displace outdated content faster than it can displace a recently refreshed competitor.

FAQ

How many keywords should a new website target initially?

Focus on building depth before breadth. For most new sites, targeting 10 to 20 carefully chosen keywords in the first six months — and creating genuinely useful content for each — will outperform a strategy that scatters effort across 100 terms. Search engines reward topical concentration, particularly for domains without established authority.

Should a new website focus on long-tail keywords?

Yes, in the early stages. Long-tail keywords — phrases of three or more words with specific intent — tend to have lower competition and attract visitors who are further along in their decision-making process. They are also far more likely to result in a page one position within a reasonable timeframe for a new domain. As your site builds authority through content and backlinks, you can compete for broader, higher-volume terms.

Do I need a paid tool to do keyword research for a new website?

Not to start. Google Keyword Planner, Google Autocomplete, and People Also Ask can produce a solid initial keyword list at no cost. Paid tools like Ahrefs or Semrush add precision — particularly around keyword difficulty and competitor gap analysis — but they are an accelerant, not a prerequisite. If budget is limited, start free and invest in a paid tool once you have content live and results to analyse.

How often should keyword research be revisited?

At a minimum, revisit your keyword list every six months. Search behaviour shifts, new competitors enter the landscape, and your own domain authority grows — all of which change which terms are worth targeting. Once Google Search Console is showing data for your site, use it to identify queries where you are ranking on pages two or three: these are often faster wins than starting from scratch on new terms.

What to Do This Week

If your site is new or has not yet launched, here are concrete first steps you can take immediately:

• Write your seed list today. Set aside 30 minutes and write every topic your site covers in plain language. Do not open a tool yet.
• Run each seed through Google Autocomplete. Note every suggestion and any People Also Ask questions. This costs nothing and typically surfaces 20 to 40 additional keyword ideas per seed.
• Set up Google Search Console before you publish a single page. Verify your domain now so that indexing and query data begins accumulating from day one.
• Filter your keyword list by difficulty. Whether using a free or paid tool, your first filter should be difficulty — not volume. Remove any keyword with a difficulty score above 25 from your immediate priority list and place it in a future pipeline.
• Map your top 10 keywords to a site structure. Decide which term is a pillar and which are clusters before you write your first brief. This decision is easier to make on a whiteboard than to undo after 20 pages are published.

Related Reading

• Ultimate Guide to Keyword Research
• What Is Topical Authority and How Do You Build It?
• Content SEO: How to Create Content That Ranks, Converts, and Gets Cited by AI
• Why Does SEO Still Matter in 2026?
• What Are Common SEO Mistakes?