If your small site publishes a bit about everything, it's hard for a reader or a search engine to tell what you're actually good at. The fix most SEO people recommend is topical authority: pick one narrow subject and cover it properly, with a main page and a set of supporting articles that link to each other.
That advice is sound, but the term is oversold. "Topical authority" is SEO-industry shorthand. Google doesn't list it as a ranking factor, there's no score for it, and nobody can tell you how many articles earn it. What follows is what Google does document, then how to pick a topic, map a cluster, link it, and check in Search Console whether it's doing anything.
What Google actually says
Three documented things sit underneath the idea.
First, Google's guide to its ranking systems says those systems mainly work at the page level, and that site-wide signals also contribute to how pages are understood. It adds that good site-wide signals don't guarantee every page ranks well. So the rest of your site matters to some degree, but each article still has to earn its place.
Second, Google's self-assessment questions for helpful, people-first content ask whether your site has a primary purpose or focus, whether the content shows first-hand expertise and depth of knowledge, and whether it describes the topic substantially and completely. A focused cluster written by someone who knows the subject answers yes to all of those.
Third, Google's link documentation says it uses links to find new pages and as a signal of what a page is about, and that every page you care about should be linked from at least one other page on your site.
None of this promises that a small site will outrank a big publisher. It does explain why one well-covered subject tends to be a better bet than forty unrelated posts, particularly for specific questions that a general site only answers in a paragraph.
How narrow should the topic be?
Narrow enough that you could list every question a beginner would ask and realistically answer all of them. Try this:
- Write the topic as one sentence that names an audience and a situation. "Gardening" fails. "Growing vegetables in containers on a balcony" passes.
- List every question someone in that situation would have. Use your own knowledge first, then Google's autocomplete and "People also ask" boxes, then forums where people in that situation ask for help.
- Count the questions that deserve their own article. If you have fewer than about eight, the topic may be too thin to build a site section around. If you have more than you could write well in the next six months, cut the topic down until the list is manageable.
- Check that you know the subject, or are willing to learn it by doing it. Thin articles on a topic you've only read about are the weak point of most clusters.
Those numbers are my rule of thumb for scoping a project, and Google publishes no threshold. If you want help deciding which questions are worth an article, the guide to keyword research based on realistic opportunity covers that step.
Mapping the pillar page and supporting articles
A cluster has two kinds of page.
The pillar page is the overview. It covers the whole topic at a level a newcomer can follow, gives a short answer on each subtopic, and links out to the article that handles that subtopic in detail. It's a useful page in its own right, and it's the page you'd send a friend who asked where to start.
Supporting articles each answer one question fully. If two planned articles would answer the same search, merge them. If one article is trying to answer three different searches, split it.
Here's an illustration of a small cluster map. It's a made-up example to show the shape, with no real site or results behind it.
| Page | Question it answers | Links to |
|---|---|---|
| Pillar: container vegetable gardening on a balcony | Where do I start? | Every article below |
| Choosing container sizes | How big a pot does each crop need? | Pillar, potting mix, tomatoes |
| Potting mix | What soil goes in a container? | Pillar, container sizes, watering |
| Watering | How often do I water pots? | Pillar, potting mix, low-light balconies |
| Low-light balconies | What grows with four hours of sun? | Pillar, watering |
| Tomatoes in pots | Which varieties work, and how? | Pillar, container sizes, watering |
Write the map before you write the articles. It shows you gaps, such as pests or winter care in this example, and overlaps before you've spent time on them. Then schedule the pieces; a content calendar built from your keyword list keeps the cluster from stalling halfway.
Publishing order matters less than people think. I'd write the three or four most specific articles first, because they're easier to do well, then the pillar once you know what it has to summarize.