Built by us, for us: we wanted to measure our own sites in search and AI – without an expensive subscription.

Start for free

Guide

Keyword Research for Beginners: Finding Topics

Find topics your customers search for, and learn how to use search intent, volume and difficulty for your website.

Keyword research for beginners is mainly about finding topics and search queries that fit your own website and the questions your target audience is asking. The key things are search intent, suitable content and clear prioritisation for your work on visibility in Google.

You are not just looking at individual words, but at what people want to achieve with a search. This helps you find topics that fit your site and can be developed sensibly over time.

For later analysis, a [Search Console overview and a [dashboard can help you look at search queries and content together. For the technical foundation, site audit and rank tracking are also useful.

Search intent

Before you gather terms, clarify the search intent. This means the reason why someone enters a search query: the person wants to find information, compare something or buy something, or commission a service.

For beginners, this is the most important filter. Not every keyword fits every page. A topic is only useful if it suits the page and the content expected. So always ask yourself: what question lies behind the term, and what answer can your website give?

Real customer questions are extremely helpful. Collect wording from conversations, emails, contact forms and common objections. Often, the words customers themselves use are the best ones. This gives you topic material that is closer to practice than a pure list of individual search terms.

Also think in topic clusters. One page can cover a main topic and address several related questions. This helps you organise content logically and expand it in a targeted way later.

Use your own website structure as a guide. Ask yourself:

  • Which pages already have a clear purpose?
  • Where is content still missing?
  • Which questions can sensibly be added?

A simple test is: would a person searching for this term expect a guide page, a product page, a category page or a local information page?

For categorising similar terms, an analysis of competitor pages can also help. If you then sort topics further, keyword research is the next step.

Sources of ideas

Start with your own website. Category pages, service pages, blog posts, FAQ sections, internal search and contact enquiries often provide the best starting points for new topics.

Another important starting point is Search Console. There you can see clicks, impressions, click-through rate and average position for your verified own website. Average position is an average across impressions and not one fixed rank.

Pay particular attention to search phrases that are already generating impressions but do not yet match your current page exactly. These often reveal new subtopics, better page titles or additional sections.

It is also helpful to compare this with the existing content:

  • Which terms are already covered by your current pages?
  • Which questions are still unanswered?
  • Where are there overlaps or duplicates?

Additional ideas come from general sources such as search suggestions, related searches, industry terms, specialist forums and customer feedback. Such sources often show how people really phrase things.

If you analyse your own pages, a dashboard can help you see different data sources in one place. In SerpDash, such functions are being developed around rank tracking, Search Console data and other modules; individual functions are still under development.

Volume and difficulty as rough guides

Search volume and difficulty are only rough pointers. They help you assess a topic, but they do not replace checking search intent and fit with your website.

Search volume simply describes how often a term is searched for. That is useful, but it still does not say whether the query really suits your offer.

Difficulty is an estimate of how competitive a topic may be. Such values are always model-based and depend on the method used to calculate them.

For selection, it is therefore important to ask:

  • Does the topic suit your target audience?
  • Can the question be answered sensibly on your website?
  • Are there already pieces of content that fit?
  • Might a related subtopic be the clearer starting point?

Especially for your own project, it makes sense to read the numbers only as guidance. A lower search volume can still be valuable if the search query matches your service or content exactly.

A technical note helps with classification: a site audit checks a website against rules such as status codes, titles, meta descriptions, headings, canonical tags, robots directives, internal links and loading time. Such findings vary in importance; in practice, serious errors are usually fixed first.

For ongoing analysis, SerpDash from ballmüller solutions, Berlin, can be used; it is being used first for the publisher's own websites, and some functions are still under development.

Choosing topics

First choose topics with a clear search intent and a direct connection to your offer. This makes later content planning more straightforward and helps avoid wasted effort. Not every keyword deserves its own page straight away; what matters is that the question fits the service, product or information on your website.

The usual sensible order is: first the topics that build on existing pages, then the topics that close gaps in your structure. That way, step by step, you create a coherent topic landscape in which new content does not sit next to each other at random.

So check for each topic which format fits best:

  • a new page, if the question is not yet properly answered anywhere
  • a revision, if an existing page fits the topic but is still too brief
  • an addition, if an important aspect is missing from existing content

A site audit also helps with the technical basis. It crawls pages and checks them against rules such as status codes, titles, meta descriptions, headings, canonical tags, robots directives, internal links and loading time. Not all technical findings are equally important; in practice, serious problems are fixed first because they can affect a page's usability and findability especially strongly.

Keep rankings in view with the necessary caution as well. A position is not a fixed value. It can change depending on location, device, language, personalisation and time of day. Rank tracking therefore always shows only the status for a selected configuration at a particular point in time.

So do not compare only individual numbers, but above all trends. More important than a single value is often how pages and search queries develop over time. That makes it easier to see which topics are holding up and which still need work.

If you want to bundle this into a workflow, SerpDash from ballmüller solutions, Berlin, can be used for that; it is being used first on the publisher's own websites, and some functions are still under development.

Frequently asked questions about keyword research

What kind of topics are suitable for getting started?

It is best to start with topics that clearly fit your offer and have an understandable search intent. For beginners, questions that your target audience is already asking are often a good choice, and that can be answered on an existing page or with a new, clearly defined page.

A topic is practical if you already have knowledge, services or content relating to it. That makes it easier to build a page properly and expand it later.

Is it enough to look only at search volume?

No. Search volume is only a rough indication of how often a term is searched for. It does not say whether the topic fits your website or whether the search intent matches your content.

For selection, more matters than a single number:

  • Does the topic fit your offer?
  • What question lies behind the search?
  • Are there already suitable contents on your website?

How should I deal with several similar search terms?

If several search terms are very close to one another, first check whether they really represent different questions. Often, a group of similar terms belongs to one shared topic.

You can then decide:

  • one main topic with additional sub-questions on one page
  • several pages with clearly separated focuses
  • a revision, if the content only differs slightly

The important thing is that pages do not resemble each other too closely or answer the same question twice.

Can I use Search Console data for choosing topics?

Yes, that is very helpful. In Search Console you can see clicks, impressions, click-through rate and average position for your verified own website. Average position here is an average across impressions and not a fixed rank.

For topic selection, search queries that are already generating impressions but still do not fit the current page well are particularly interesting. These often lead to new subtopics, better titles or additional sections.

What about AI searches?

AI searches are still a new area. They consist of answers presented with language models and source links, but their behaviour is not fully documented.

For that reason, you should treat statements about visibility there with caution. What looks like a good source is more an observation or assumption than a fixed rule. When assessing this, continue to use search intent, fit with the content and the development of your own website.

SerpDash from ballmüller solutions, Berlin, brings together rank tracking, Search Console data, site audit and visibility in AI searches in one dashboard; some functions are still under development.

Next step

Turn your collection of ideas into a small prioritisation:

  • first the search intent
  • then the fit with the offer
  • then the technical feasibility
  • only after that, the numerical values

If you want to check your website regularly, tools for Search Console, rank tracking and site audit help keep developments clearly organised. That way, you can see topics, search queries and technical notes in a logical order.

SerpDash from ballmüller solutions, Berlin, is being developed for this purpose; it is first being used on the publisher's own websites, and some functions are still under development.

This lets you keep track of your topics, search queries and technical foundation in one workflow, without overloading keyword research.

All guides