Content Strategy & Research

Keyword & Topic Research

Know exactly which questions deserve a page, which deserve a refresh, and which deserve neither.

Search volume alone doesn't tell you what to write. Keyword & Topic Research maps query demand to the audience behind it, so every topic decision comes with intent, business relevance, and a clear next step attached. You walk away with clusters that carry an intent, a business-fit score, and a next action, checked against what you have already published.

Map demand into topics
Keyword clusters and search-result patterns being sorted into a prioritized topic map

Some of the 500+ brands we've worked with

See all references
  • Hepsiburada
  • Capital Dergisi
  • QNB Finansfaktoring
  • Duru
  • Sina Pırlanta
  • Kale
  • Yolcu360
  • Amazon
  • BMW
  • Shell
  • Hyundai
  • PepsiCo
  • Red Bull
  • Decathlon
  • MediaMarkt
  • Bayer
  • Sanofi
  • EY
  • KPMG
  • GE
  • 3M
  • Domino’s
  • Lexus
  • Trendyol

How we run it

How the demand map gets built

Each step turns raw query data into a decision a planning meeting can use.

  1. Pull the raw demand signal

    Query datasets, result-page patterns, and whatever first-party search and site data you already have, gathered before any judgment gets applied. A strategist checks that the pulled dataset's sources are reliable before anyone builds on top of it.

    A raw demand dataset with sources attached.

  2. Check it against what already exists

    Every query gets checked against your current URL inventory, so we aren't recommending a page for something you already rank for. A strategist resolves any ambiguous overlap between a query and an existing page, since a surface-level match can still be a false one.

    A demand set filtered against existing coverage.

  3. Cluster by real intent

    Queries that look similar but mean different things get split apart. Queries that mean the same thing get grouped, even if the wording differs. A strategist decides where an ambiguous query actually splits, since automatic clustering can still miss the difference.

    Intent-based clusters.

  4. Score for business fit and confidence

    We score each cluster for business relevance and evidence confidence alongside demand. Volume alone would rank the list badly. A strategist weighs the business relevance the data can't see on its own before a cluster earns a high score.

    A prioritized topic map with confidence and business-fit scores.

  5. Assign the next action

    Every cluster gets a clear call: new page, section addition, refresh, or leave alone. That turns raw data into a decision the roadmap can use. Before a cluster reaches the roadmap, a strategist approves its final call: new page, refresh, or leave alone.

    A demand map with an explicit next action per cluster.

Agents can sort the query set. A strategist sets the priority.

Agents are useful for the heavy sorting: matching thousands of queries to your URL inventory and drafting an initial set of clusters. Search volume still says nothing on its own about whether a topic deserves a page. A Zeo strategist reads the intent, weighs the business relevance, and makes the final call. Weak or ambiguous evidence remains marked for review on the roadmap.

What you get

What the keyword map prioritizes

A ranked list only helps if the reasoning behind it survives contact with a skeptical stakeholder.

  • a ranked opportunity map

    Prioritized keyword and topic demand map

    Clusters, intent, audience job, and a next action for each, organized so a planning meeting can use it directly.

  • a ledger of link entries

    Evidence and source register

    Where every number came from, so a skeptical stakeholder can check it themselves.

  • a topic cluster map with linked nodes

    Page-boundary notes

    Which queries share a page and which need their own, so the next brief does not accidentally create two pages competing for the same result.

We call it done when: every cluster has an intent, a business-fit score, and a next action, and the map has been checked against what is already published.

Who it fits

When this earns its place

This decides what's worth writing about. It doesn't write the brief or the page itself.

A good fit when

  • A large keyword export exists, but nobody can tell which queries deserve a new page, a refresh, or no action at all.
  • SEO and content keep disagreeing about priority because nobody has checked demand against what already exists.
  • Every roadmap topic needs a reason a skeptical stakeholder can inspect, but the current priority is backed only by search-volume rows.

Better handled as other work when

  • Search volume is being treated as proof of audience demand, so high-volume queries move forward without anyone checking the intent behind them.
  • The plan gives every keyword its own page. Several phrases share one intent while ambiguous queries split into different needs, so the URL boundaries will be wrong.
  • A specific ranking has to be promised before the topic is approved, although the page, site, and competition still determine the eventual position.

People who write and ship this content

Some of these consultants write the pages themselves. Others plan distribution or check how a piece performs once it's live, so the team below covers more than one stage of the work.

Content we've produced on this topic

Articleby Metehan Urhan6 Reasons To Focus on Long Tail KeywordsVideoby Ozan KetenciAnahtar Kelime Analizi: Büyük Verinin Anlamlandırılması - Ozan KetenciVideoby Ozan KetenciAnahtar Kelime Listelerini Kullanıcı Eğilimlerine Göre Hesaplamak - Ozan KetenciArticleby Samet ÖzsüleymanHow to Analyze Long-tail Words in R

Tools we use

Tools behind this work

  • Ahrefs

    the primary volume pull: thousands of keyword ideas, clustered, in one query

  • Google Keyword Planner

    a second, Google-sourced volume number to check the primary pull against

  • Keyword Cupid

    clusters by real search intent, using which URL Google already ranks

  • Airtable

    the demand map itself: cluster, score, and assigned action in one table

  • Google Search Console

    checks a candidate topic against what the site already ranks for

Start the work

Sitting on keyword data nobody's turned into a plan?

An existing export is enough to begin. We add intent, compare it with current coverage, and give each cluster a next action.
Talk to a Content Marketing specialist
Two people shaking hands on the start of the work

Before we start

Frequently Asked Questions

Do you just hand us a keyword list?

You get clusters organized by intent and the value of the decision they support. A bare keyword list would leave the prioritization work unfinished.

What tools do you pull data from?

We use whatever gives us reliable first-party and query data for your market. The source can vary. We are not attached to one vendor. Any estimate that serves as a proxy is labeled that way.

How do you handle queries that could mean two different things?

Ambiguous queries are separated when the search results point to different needs, since combining those intents on one page usually serves neither searcher well.

Can this tell us if a topic will rank?

This research shows whether demand and business fit justify the investment. It cannot promise where the finished page will rank because topic choice is only one factor.