Content Marketing Glossary

  • What is Keyword Research?

What is Keyword Research?

Keyword Research

Keyword research is the process of figuring out which search terms your audience uses, which ones your business should be writing about, and which ones you can realistically rank for. For most of SEO’s history, keyword research was the first step of any content strategy: pull a keyword list, rank by search volume, assign topics. That playbook isn’t dead, but it’s been demoted. In 2026, keyword research matters less than two other inputs: authority positioning and business alignment.

Why chasing keywords in isolation stopped working

Picking individual high-volume keywords and writing one piece per keyword is the pattern most SEO tools still sell. It produces content lists, editorial calendars, and quarterly reports full of green arrows. It also produces a lot of content that ranks briefly and disappears.

The reason: ranking doesn’t come from individual pages. Ranking comes from clusters, from topical authority, from being the site Google and AI answer engines trust on a whole topic. A single keyword-optimized piece in a site with no cluster behind it wins rarely and loses quickly.

Traditional keyword research optimizes for the wrong unit. It picks keywords instead of topics. It cherry-picks queries that sound promising without asking whether the site can plausibly cover the whole pillar those queries belong to.

Chasing keywords one at a time isn’t how it works anymore. You cover the topic. Not the query.

The three-input hierarchy that replaces keyword-first

The decision about what to write shouldn’t start with “what’s high-volume and low-competition.” It should start with two questions that most tools can’t answer for you.

Authority positioning. What would position you as THE authority on this topic? Not what would rank. What would establish, over 18 months, that your site is the credible source on this subject. That’s a different shape than a keyword list.

Business alignment. What makes sense for your business to be covering? What are the topics that, if you’re known for them, produce customers and not just traffic? What’s your business even supposed to be talking about?

Keyword research. Then, on top of those two, the traditional methods. Volume, competition, SERP analysis, gap identification.

The order matters. Authority and alignment first. Keyword data second.

Here’s the hard truth about why this order matters. Keyword volume sells the strategy to the people who pay you on day one. It makes the plan look defensible in the pitch deck. Business alignment is why you still have a job six to twelve months in, when the CFO asks what the content program has produced.

What modern keyword research actually looks like

The work has shifted from “find high-volume terms” to “decompose the topical space your business should own.” Four steps.

Define the pillars your business should own. Two to five topics, narrow enough that you can plausibly dominate them in 18 months. Come out of authority positioning plus business alignment, not out of keyword tools.

For each pillar, decompose the SERP landscape. What subtopics do the ranking competitors cover? What cluster shape does the category expect to see? This is closer to search intent decomposition than classical keyword harvesting.

Build the table-stakes cluster from the decomposition. The pieces the pillar needs to exist. Each gets a target query. Each query ties back to the pillar. Keyword research, in this version, happens at the cluster level, not the isolated-page level.

Layer in differentiation queries. Once the table stakes are covered, look for queries that represent genuine information gaps in the category: things buyers search that no ranking competitor answers well. These are the pieces with the strongest payoff relative to effort, not the high-volume ones.

This is a different shape than the keyword-spreadsheet approach. The output is a cluster plan, not a list.

Where traditional keyword metrics still matter

Not everything about keyword research is dead. Three specific uses still work.

Volume as a directional signal. Within a pillar you’ve already decided to own, volume helps prioritize which pieces to ship first. Not as a selector of what to write, but as a sequencer of which piece in the already-chosen cluster to ship first.

Keyword difficulty as a reality check. If the entire target cluster is dominated by domains with 80+ DR while you’re at 30, volume doesn’t matter, you’re not getting in. Keyword difficulty catches this.

Long-tail discovery. Keyword tools still find specific long-tail queries you wouldn’t have brainstormed. Those queries feed into the differentiation layer of the cluster.

The rule: keyword metrics are useful inside a strategic frame, useless when they substitute for one.

Common keyword-research mistakes

Four patterns that keep showing up.

Building calendars from keyword lists. The editorial calendar gets assembled by walking down a ranked spreadsheet of target keywords. Produces disconnected pieces, no cluster structure, no compounding authority.

Chasing volume over relevance. Picking high-volume keywords that tangentially relate to the business. Produces traffic that doesn’t convert, which produces an eventual budget cut.

Ignoring the SERP. Keyword tools give you query strings. The SERP tells you who’s actually ranking, what they cover, and what the cluster shape needs to be. Research that stops at the keyword tool stops too early.

Treating “keyword research” as a one-time event. Keywords evolve. Queries emerge. Intents shift. Research done at launch and never revisited goes stale inside 18 months.

Keyword research in the AI-search era

Two shifts worth knowing.

First: AI Overviews change click-through economics on specific queries. A keyword that had high volume and 60% click-through rate in 2022 might have the same volume and 15% click-through in 2026 because AIO answers the query in-SERP. Volume without a click-through-rate adjustment for AIO presence is a misleading metric.

Second: ranking for citation in AI answers is a different game from ranking for clicks. Some keywords are worth targeting primarily for citation value, not traffic. Keyword research that only optimizes for traffic misses this category.

Penfriend’s approach

We built Penfriend’s Cluster product around the observation that keyword research is the scaffolding, not the foundation. Authority positioning and business alignment are the real inputs. Cluster decomposes the pillar into a table-stakes cluster from actual ranking competitors, builds the brief shape around intent decomposition, and layers differentiation on top once the foundation ranks. Penny interviews the expert to surface the angle the keyword data never finds. The keyword work happens. It’s just not where the strategy starts.

Related terms