Keyword Research in 2026: Using AI to Find Opportunities Your Competitors Miss

How AI Overviews Changed Keyword Strategy

Not all keywords are created equal in 2026. The rise of Google AI Overviews has bifurcated the keyword landscape:

Informational keywords (how to X, what is Y, explain Z) now frequently trigger AI Overviews that reduce organic CTR significantly. A #1 ranking for "how to fix CLS" might generate 60% fewer clicks than it would have in 2022, because many users get their answer from the AI Overview without clicking through.

Commercial and transactional keywords (best X for Y, X pricing, buy X) are less affected - AI Overviews appear less frequently, and users with purchase intent tend to click through anyway.

This means your keyword research needs to account for intent more carefully than before. Ranking #1 for an informational keyword with an AI Overview is still valuable for brand exposure, but the click value per impression has dropped.

Using Semrush for AI-Assisted Keyword Grouping

Semrush's Keyword Magic Tool generates thousands of related keywords from a seed. The key is grouping them intelligently:

1. Enter your seed keyword (e.g., "project management")
2. Export all keywords with volume > 100
3. Use Semrush's built-in clustering, or export to CSV and use ChatGPT/Claude to group by topic and intent

Prompt for AI grouping: "Here are 200 keywords about project management. Group them by: (1) search intent (informational/commercial/transactional), (2) topic cluster (scheduling, team, budgeting, tools, etc.), and (3) estimated content type needed (guide, comparison, tool page). Return as a structured table."

This reduces a 2-hour grouping task to 5 minutes.

Ahrefs Content Gap Analysis

Content gap analysis finds keywords your competitors rank for that you don't. In Ahrefs:

1. Site Explorer → your domain
2. Content Gap
3. Enter 3-5 competitor domains
4. Filter: competitor ranks in top 10 for at least 2 of them, you rank below 20 or not at all
5. Sort by volume

These are validated keywords - competitors are already getting traffic from them, meaning there's proven search demand. Your content can compete if you create something better.

GSC Hidden Gems: Position 8-20 Queries

Your own GSC data contains some of the best opportunities. You already have some relevance signal for these queries - Google is showing your content, just not prominently.

1. GSC Performance → filter position > 7 and < 21
2. Filter impressions > 200 (meaningful volume)
3. Export to CSV

For each keyword, check what page is ranking and whether it directly targets that query or is incidentally relevant. Often you'll find the page needs: a more specific title tag, a dedicated section for that query, or better internal links from related pages.

People Also Ask for Semantic Expansion

Google's "People Also Ask" boxes show questions that are semantically related to your target query. Use these to expand your content outlines:

1. Search your target keyword
2. Expand every PAA question and note them all
3. Include answers to the top 5-8 PAA questions in your content

This strategy directly targets the question-based queries that often trigger AI Overviews - and appearing in PAA is another visibility signal separate from your main ranking.

AI-Assisted Intent Classification

Use Claude or GPT to classify a list of keywords by search intent at scale:

• Informational - user wants to learn something
• Navigational - user wants to reach a specific site
• Commercial investigation - user is comparing options before buying
• Transactional - user is ready to act (sign up, purchase, download)

This classification informs which pages to create (guides vs comparison pages vs landing pages) and how to measure success (engagement vs conversions vs rankings).

Links: Semrush Keyword Magic Tool | Ahrefs | Google Trends