Skip to content
Tutorials Center
My learning

Topics

Lesson 9 of 36 ยท Keyword Research

Keyword Research Basics: Finding What People Search For

Learn a repeatable keyword research process: seed topics, sources of ideas, expanding and cleaning lists, checking intent and choosing targets, with runnable Python helpers.

Keyword research is the work of discovering the words and phrases your audience types into search engines, and deciding which of them your site should target. It connects what people want with what you can offer. Done well, it tells you which pages to create, how to title them and which topics are worth the effort.

Keywords are the language of your audience, not of your company. A bakery may call a product "viennoiserie", but customers search for "croissants". The goal of this chapter is to find that gap and close it.

What a keyword is

A keyword (or query) is any word or phrase a person enters into a search box. Despite the name, many are full questions: "why are my tomato leaves curling". The same page can attract thousands of different queries. Keyword research picks a primary target for the page and notes the related phrases it should also answer.

A simple process

flowchart LR
    A["1. Seed topics"] --> B["2. Collect ideas"]
    B --> C["3. Clean and group"]
    C --> D["4. Check intent and competition"]
    D --> E["5. Prioritize"]
    E --> F["6. Map to pages"]

The next lessons go deeper into steps 4 to 6. This lesson covers steps 1 to 3.

Step 1: seed topics

Seeds are the broad subjects you want to be known for. Start from your products, services, expertise and the problems you solve. Write 5 to 15 short terms. For a garden shop: "tomato plants", "raised beds", "compost", "watering".

Step 2: collect ideas

Where do ideas come from?

  • Autocomplete and related searches. Type a seed into a search box and note the suggestions and the "related searches" at the bottom of the results.
  • People also ask questions on results pages.
  • Your own data. The queries report in your webmaster tools shows what you already get impressions for, which is often the best starting point.
  • Site search and customer language. What do visitors type into your internal search box? What do support emails and reviews say?
  • Competitors' pages. Their titles and headings reveal topics they target.
  • Communities. Forums, Q&A sites and social groups show real wording and real problems.
  • Keyword tools. Free and paid tools suggest related phrases and estimate demand.

Combine several sources; every source has blind spots.

Try it: expand seeds with modifiers

A quick way to generate candidates is to combine seeds with common modifiers: question words, qualifiers, audiences, places. The program below builds a list, removes duplicates and sorts it. These are candidates only; you still have to check which ones people truly search.

Python Expand seed keywords

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Notice that some outputs sound unnatural ("cheap raised bed problems"). Real search data filters these out, which is why mechanical expansion is only the first step.

Step 3: clean and group

Raw lists contain noise. Clean them by:

  1. Lowercasing and trimming whitespace.
  2. Removing exact duplicates and obvious near-duplicates (plural versus singular often belong together).
  3. Removing irrelevant or off-brand phrases. "Tomato soup" is not relevant to a shop that sells plants.
  4. Tagging each phrase with intent and topic.
Python Clean and tag a keyword list

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

The two dropped duplicates show how merging singular and plural variants keeps the list tidy. In real work you would not merge variants whose search results differ, so check the results page when it matters.

What to record about each keyword

A good keyword sheet has columns like these:

Column Why it helps
Keyword The phrase itself
Intent Informational, commercial, transactional, navigational
Monthly volume (estimate) How many searches; see volume and difficulty
Difficulty (estimate) How hard it is to rank
Current rank (if any) Where you already appear
Target page The URL that should answer it
Notes Seasonality, SERP features, competitors

Keep it in a spreadsheet; it becomes your content plan.

Common mistakes

  • Chasing only high-volume terms. Big phrases are competitive and vague; smaller specific phrases often convert better (long-tail keywords).
  • Ignoring intent. A keyword with plenty of searches is useless if you cannot match what searchers want (search intent).
  • Trusting tool numbers as exact. They are estimates. Use them to compare, not to forecast.
  • Researching once. Language, trends and competitors change; revisit your list regularly.
  • Stuffing the exact phrase everywhere. Pick a primary phrase for the title and main heading, then write naturally.

Check your understanding

Quick check ยท 4 questions

of 4 answered You scored / 4 ยท perfect!

  1. 1. Which source of keyword ideas gives you data about what you already receive impressions for?

    Search console reports list the queries that already show your pages, which is a reliable starting list.
  2. 2. Why is mechanical keyword expansion only a first step?

    You must validate candidates against real data and the live results.
  3. 3. What is a seed topic?

    Seed topics are the starting subjects from which you collect more specific ideas.
  4. 4. Which is the best way to treat search volume numbers from a tool?

    Estimates differ between tools; they are best for relative comparison.

Summary

  • Keyword research finds the language your audience uses and decides which phrases to target.
  • Start from seed topics, gather ideas from several sources, then clean, deduplicate and tag.
  • Record intent, volume, difficulty, current rank and target page in a keyword sheet.
  • Avoid chasing volume alone, ignoring intent or trusting estimates as facts.

Next: Keyword Types and Long-Tail Keywords