Skip to content
Academy
Marketing Academy · Field Work●AI in Marketing
MiniTeardown· 25 minutes

SERP Overlap vs Semantic Similarity: Teardown of Three Keyword Clusters

Zomato

Objective: Given three candidate keyword clusters generated by different AI methodologies, apply the lesson's SERP-overlap validation framework to identify keyword cannibalization, mismatched search intents, and improper pillar-to-supporting page mappings.

You're reviewing automated keyword clustering outputs for Zomato's restaurant dining and Gold membership landing pages before handing content briefs to the writing team.

Examine three clustering specimens. Detect which clusters group keywords by surface meaning rather than true SERP intent overlap, and flag any instances where informational queries are mixed with transactional landing page targets.

Before you start

What you'll need

Free path (everything below is enough to finish)

FreeAudit keyword lists against SERP overlap percentages and intent classification

Free, collaborative matrix analysis

FreemiumRun intent classification prompts and detect cannibalization risks

Free tier handles semantic analysis of keyword groupings

Paid upgrades (optional, faster/deeper)

Manual SERP checking via Google Search and Google Sheets is 100% free; Surfer SEO or Keyword Insights automates overlap calculations for thousands of queries at once.

Surfer SEO(optional)
PaidAutomated SERP overlap auditing and content brief generation

Extracts live top-10 SERP competitor entities and keyword clusters

The process

Specimens to review

Evaluate this keyword cluster against SERP overlap and search intent principles. Name every defect you find.

Sample output
Cluster Name: 'Bangalore Dining & Restaurants'
Target Page: Single new landing page (/dining/bangalore-guide)
Keywords in Cluster:
1. best restaurants in bangalore (vol: 49,500)
2. top 10 places to eat in indiranagar (vol: 14,800)
3. how to register a restaurant on zomato bangalore (vol: 6,600)
4. fine dining bangalore reviews (vol: 8,100)
5. restaurant license requirements karnataka (vol: 2,400)

Specimen: synthetic, realistic

Evaluate this keyword cluster against SERP overlap and search intent principles. Name every defect you find, or state clearly that there are none.

Sample output
Cluster Name: 'Buffet Deals & Dining Out Offers'
Target Page: Transactional category page (/dining/buffet-deals)
Keywords in Cluster (80%+ SERP Overlap):
1. buffet offers near me (vol: 22,000, intent: Transactional)
2. best buffet dinner deals (vol: 12,100, intent: Transactional)
3. 5 star hotel buffet discounts (vol: 8,900, intent: Transactional)
4. luxury buffet booking offers (vol: 4,400, intent: Transactional)
SERP Analysis: Top 5 ranking URLs across all 4 queries are identical discount booking pages.

Specimen: synthetic, realistic

Evaluate this clustering and content plan. Name every defect you find.

Sample output
Proposed Content Plan:
Article A: 'Zomato Gold Membership Discounts & Promo Codes' (/blog/zomato-gold-discounts)
Article B: 'How to Get Zomato Gold Offers and Coupons' (/blog/zomato-gold-coupons)
Existing Page: Main Zomato Gold landing page (/gold)
Strategy: Create both blog articles to target the two keyword variations separately.

Specimen: synthetic, realistic

Final deliverable

A keyword cluster diagnosis report identifying cannibalization and intent mismatches across the specimens, with a corrected topic cluster mapping.

See a reference example
Sample output
Swiggy Instamart Keyword Cluster Diagnosis Report

Cluster A Specimen: 'Quick Grocery Delivery'
Keywords: 'instant grocery delivery', 'grocery delivery app', 'history of grocery stores in India'
Defect Found:
- Severity: Critical
- Issue: 'history of grocery stores in India' is purely informational/academic intent, while 'instant grocery delivery' is transactional.
- Action: Remove historical query to an educational blog post; keep transactional terms on the /instamart category page.

Cluster B Specimen: 'Late Night Snack Delivery'
Keywords: 'late night food delivery 2am', 'midnight snacks delivery', 'order food late night'
Analysis: Clean cluster (85% SERP co-ranking overlap, shared late-night transactional intent). Map to single landing page.

Success criteria

You're done when you can:

  • Correctly identifies Specimen 2 as the clean SERP-overlap cluster
  • Flags B2B vs. consumer intent mixing in Specimen 1 as a critical defect
  • Identifies keyword cannibalization in Specimen 3 where duplicate pages target the same intent