Keyword Research Isn't Dead — But It's Unrecognisable
If your keyword research process still looks like "find high-volume, low-competition keywords and write pages targeting each one," you're working with an outdated playbook.
Google's understanding of language has fundamentally changed. Between BERT, MUM, the Helpful Content system, and AI Overviews, Google no longer matches keywords to pages. It matches topics to intent. It understands synonyms, related concepts, and the depth of your coverage.
The result: websites that cover a topic comprehensively outrank those that target individual keywords in isolation, even if those individual pages are well-optimised.
Keyword research in 2026 is really topic research. The keywords are still important — but they serve the topic strategy, not the other way around.
Search Intent: The Foundation of Everything
Before you look at a single keyword, you need to understand what people are actually trying to accomplish when they search.
The Four Types of Intent
Informational — Learning something
- "what is schema markup"
- "how to set up Google Analytics"
- "best time to post on Instagram"
Navigational — Finding a specific site or page
- "HubSpot login"
- "Tiberius Digital contact"
- "Ahrefs pricing"
Commercial Investigation — Researching before buying
- "best digital marketing agency NZ"
- "Mailchimp vs ActiveCampaign"
- "SEO services Auckland reviews"
Transactional — Ready to take action
- "buy SEO audit"
- "hire web developer Auckland"
- "book marketing consultation"
Why Intent Matters More Than Volume
A keyword with 10,000 monthly searches and informational intent might generate zero leads. A keyword with 50 monthly searches and transactional intent might generate 10 high-value enquiries.
Example:
- "what is digital marketing" — 50,000 searches. Students, curious people. Almost no commercial value.
- "digital marketing agency for plumbers NZ" — 30 searches. Every single one is a potential client.
Chasing volume without considering intent is how businesses end up with traffic that never converts.
How to Identify Intent
Check the SERP. Google has already figured out the intent. Look at what it ranks:
- Blog posts and guides? → Informational
- Product/service pages? → Transactional
- Comparison articles? → Commercial investigation
- Brand homepages? → Navigational
If Google shows blog posts for a keyword, don't try to rank a service page. Match the format Google is already rewarding.
Topic Clusters: The Modern Architecture
Instead of isolated pages targeting isolated keywords, build topic clusters.
The Structure
Pillar page: A comprehensive, authoritative page covering a broad topic (2,000-5,000 words).
Cluster pages: Individual pages covering subtopics in depth, all linking back to the pillar.
Internal links: Every cluster page links to the pillar. The pillar links to every cluster page. Cluster pages link to each other where relevant.
Example Cluster
Pillar: "The Complete Guide to Digital Marketing for NZ Businesses"
Clusters:
- SEO for NZ businesses
- Google Ads for local businesses
- Social media marketing strategy
- Email marketing automation
- Website conversion optimisation
- Content marketing strategy
- Analytics and tracking setup
Each cluster page is a standalone, valuable piece of content. Together, they tell Google: "This website covers digital marketing comprehensively. It should be an authority on this topic."
Why Clusters Work
- Topical authority: Google rewards breadth and depth of coverage on a subject
- Internal linking: Creates strong signals about your site's content structure
- User experience: Visitors can explore related content naturally
- Ranking lift: Pillar pages benefit from the authority of all cluster pages
Studies show sites using topic clusters see up to 40% more organic traffic compared to isolated content strategies.
The Research Process
Step 1: Start With Seed Topics (Not Keywords)
Think about the broad topics your business should own.
Ask:
- What do we want to be known for?
- What questions do our customers ask before buying?
- What topics does our sales team discuss most?
- What problems do we solve?
For a digital marketing agency, seed topics might be:
- SEO
- Paid advertising
- Social media
- Web design
- Content marketing
- Analytics
- Branding
- Marketing automation
Step 2: Expand With Keyword Tools
Use tools to discover the specific queries people search within each topic.
Tools:
Ahrefs Keywords Explorer
- Search volume, keyword difficulty, click data
- "Also rank for" and "Questions" reports
- Content gap analysis
- SERP analysis
SEMrush Keyword Magic Tool
- Topic groups and subgroups
- Intent classification built in
- Competitive keyword analysis
- Keyword gap tool
Google Search Console
- Free. Shows actual queries your site appears for
- Reveals opportunities you're already close to ranking for
- Real click and impression data (not estimates)
Google's "People Also Ask" and Autocomplete
- Free intent signals
- Real queries from real people
- Use AlsoAsked.com or AnswerThePublic for bulk extraction
ChatGPT and AI tools
- Generate topic ideas and subtopics
- Identify questions your audience might ask
- Brainstorm content angles
- Not a replacement for data tools, but great for ideation
Step 3: Classify by Intent
For every keyword on your list, tag the intent:
| Keyword | Volume | Intent | Priority | |---------|--------|--------|----------| | what is SEO | 40,000 | Informational | Low | | SEO services Auckland | 400 | Transactional | High | | best SEO tools 2026 | 2,000 | Commercial | Medium | | how to improve website ranking | 3,000 | Informational | Medium | | hire SEO consultant NZ | 100 | Transactional | High |
Prioritise transactional and commercial keywords for service/product pages. Use informational keywords for blog content that feeds your funnel.
Step 4: Group Into Clusters
Organise your keywords into topic clusters:
Cluster: Local SEO
- Pillar: "Local SEO: Complete Guide to Ranking in Google Maps"
- local SEO tips (informational)
- Google Business Profile optimisation (informational)
- local SEO services NZ (transactional)
- how to rank in Google Maps (informational)
- local citation building (informational)
- best local SEO tools (commercial)
Each keyword becomes a section within a cluster page, or its own dedicated page if the topic warrants it.
Step 5: Analyse the Competition
For your highest-priority keywords:
- Who ranks on page 1?
- What type of content do they have? (length, format, depth)
- What topics do they cover that you don't?
- Where are the gaps you can fill?
- Can you create something genuinely better?
Content gap analysis: Use Ahrefs or SEMrush to find keywords your competitors rank for that you don't. These are immediate opportunities.
Step 6: Map Keywords to Pages
Create a keyword-to-page mapping:
| Target Keyword | Intent | Page | Status | |----------------|--------|------|--------| | SEO services Auckland | Transactional | /services/seo | Exists — needs optimisation | | how to improve website ranking | Informational | /blog/improve-ranking | To create | | SEO vs PPC | Commercial | /blog/seo-vs-ppc | To create |
This mapping prevents keyword cannibalisation (multiple pages competing for the same keyword) and ensures every piece of content has a clear SEO purpose.
Keyword Research for AI Search
With Google AI Overviews, ChatGPT, Perplexity, and other AI platforms reshaping how people find information, keyword research now has a new dimension.
What's Different
- AI answers conversational, long-tail queries directly
- Users ask questions in natural language, not keyword fragments
- AI platforms cite sources — being cited is the new "ranking"
- Entity-based search means Google connects concepts, not just words
How to Adapt
Research conversational queries:
- "How should a plumbing business in Auckland approach SEO?"
- "What's the best marketing strategy for a small NZ business with a $2K monthly budget?"
These natural-language queries are growing rapidly. Your content should answer them directly.
Build entity relationships:
- Create content that clearly defines and connects entities (your brand, your services, your industry, your location)
- Use schema markup to help AI systems understand your content structure
- Build a knowledge graph around your business through consistent content
Optimise for citations:
- Provide clear, factual, well-sourced answers
- Structure content with headers that match question patterns
- Include data and statistics that AI systems can extract and cite
Common Keyword Research Mistakes
- Chasing volume over intent — 50,000 searches means nothing if none of them are your customers
- Ignoring long-tail keywords — less volume, but much higher conversion rates and easier to rank for
- Targeting keywords you can't rank for — a 6-month-old site won't rank for "digital marketing" against established domains
- No content mapping — creating content without knowing which keyword each piece targets
- Keyword stuffing — repeating a keyword 47 times doesn't help anymore. It hurts.
- Researching once and never again — search behaviour evolves. Refresh research quarterly.
- Ignoring Search Console data — the best keyword insights come from queries you're already appearing for
- Separate pages for synonyms — Google understands that "SEO agency" and "SEO company" are the same intent. One page can rank for both.
A Practical Starting Point
- List 5 seed topics your business should own
- Use a keyword tool to find 20-30 keywords per topic
- Tag each keyword with intent
- Group keywords into 3-5 topic clusters
- Audit your existing content against these clusters
- Identify gaps (topics you should cover but don't)
- Create a content plan that fills the gaps, starting with transactional keywords
- Build internal links between related content
- Review Search Console monthly for new keyword opportunities
- Refresh your research quarterly
The goal isn't to rank for the most keywords. It's to rank for the right keywords — the ones that bring people who are ready to become customers, at the moment they're looking for exactly what you offer.