IMVASA

Find keywords you can win

A worked example of choosing keywords a small site can actually rank for, from a seed topic to a prioritised list mapped to pages.

The situation

Brewlab is a small coffee roaster with a young website. It wants search traffic from people learning to brew coffee at home, some of whom will buy beans. Big coffee magazines rank for the obvious terms, so Brewlab needs keywords it can realistically win.

Step 1: start from your customers' words

List three to five seeds: short phrases your customers would type. Brewlab picks "cold brew", "pour over" and "coffee grinder". Avoid jargon nobody searches.

Step 2: pull ideas

Run Keyword research for each seed in the United States, using the Related source first. Each run costs about 2 credits.

Step 3: filter for what is winnable

Keep keywords that pass all of these:

  • Difficulty under 30. For an established site, raise the limit.
  • Some search volume. Even 50 searches a month is worth it for a specific, useful search.
  • An intent Brewlab can serve with a guide or a product page.

From "cold brew", that leaves searches such as "cold brew ratio", "how long to steep cold brew" and "cold brew without a filter", while "cold brew" itself is too hard for now.

Step 4: check the results pages

Open the SERP checker for the most promising keywords, or search them yourself. Brewlab looks for:

  • Beatable pages: forums, thin posts or old articles in the top ten are a good sign.
  • The format that wins: if every result is a step by step guide, Brewlab writes a step by step guide.
  • SERP features: an AI Overview that fully answers "cold brew ratio" means fewer clicks, but a clear, quotable answer can still earn a citation.

Step 5: group keywords into pages

Keywords that mean the same thing belong on one page. Brewlab groups:

PageKeywords
Cold brew ratio guidecold brew ratio, cold brew coffee to water ratio, how much coffee for cold brew
How long to steep cold brewhow long to steep cold brew, cold brew steeping time, can you steep cold brew too long
Cold brew without special equipmentcold brew without a filter, cold brew in a mason jar

One strong page per group beats several thin pages competing with each other.

Step 6: prioritise

Score each page idea on three things: combined volume, average difficulty and business value (does it lead to buying beans?). Start with pages that are easy and valuable. Put hard, high-volume terms on the list for later, once the site has more authority.

Step 7: write, link and track

  1. Write each page with the main question answered in the first two sentences. Use questions from What People Ask as headings.
  2. Link the new pages to each other and from a broader cold brew page, forming a small topic cluster.
  3. Add the main keyword of each page to Rank tracking and check weekly.

What it cost

Three keyword research runs, a few SERP checks and a What People Ask lookup: roughly 20 credits for a list that can drive months of content.