Written by: Learning Objectives
This module explains how to identify and fix SEO index bloat, keyword cannibalisation, and over-built websites. It covers auditing page inventories, mapping keywords to pages, removing low-value pages, consolidating duplicate content, improving location/service pages, and building scalable SEO structures for trades and local service businesses.
By the end of this module, you’ll be able to:
- Recognise the warning signs of index bloat — “Crawled, not indexed” and “Discovered, not indexed” — before they cripple a new site’s traffic
- Explain keyword cannibalisation in plain terms and spot it happening on your own website
- Audit a site’s page count against genuine search demand, not gut feeling
- Apply a cull-consolidate-rewrite framework to clean up an over-built website
- Build new pages with a “one keyword, one page” rule from day one
Key Concepts
- Index bloat — When a website has far more pages than Google is willing to give shelf space to. Think of Google’s index like a library with limited shelving. If you print a thousand near-identical books on “how to unblock a drain,” the librarian starts refusing your later editions — not because they’re bad, but because there’s no room and no need.
- Crawled, not indexed / Discovered, not indexed — Google Search Console statuses meaning Google has seen the page but decided it isn’t worth storing in search results. This is different to a 404 (page doesn’t exist) — the page is live, Google just won’t show it.
- Keyword cannibalisation — When two or more pages on the same site compete for the same search term, so Google can’t decide which one to rank and often ranks none of them consistently. It’s like sending five staff to quote the same job — the client gets confused, not impressed.
- TF-IDF (term frequency–inverse document frequency) — A way of measuring how often top-ranking pages use certain terms relative to the rest of the web, so you can naturally include the language Google expects on a topic without keyword-stuffing.
- Exact-match domain (EMD) cluster — A network of niche-specific domains (e.g., a “gas fitting” site, a “blocked drains” site) built to generate directory-style mentions of a client’s brand, boosting citations picked up by both search engines and AI answer engines.
Real-World Case Studies
Case Study: Southern Cross Plumbing Co. — Trades (Plumbing) — Parramatta, Sydney
The Challenge: A newly built website — under 12 months old — had ballooned to over 2,000 pages. Google Search Console showed the majority flagged as “Crawled, not indexed” or “Discovered, not indexed.” More than 30 separate pages (service pages, suburb pages, and blogs) were all targeting the same core term, “emergency plumber,” and several suburbs had four or five near-identical pages (north, south, east, west, and central variants).
The Strategy: We ran a full page export and audit, cross-referencing every URL against Google Search Console impressions and click data. Pages with zero search demand were culled outright — around 500 in the first pass. Suburb variants were consolidated down to a single strongest-performing page per suburb, with the others 301-redirected. Blog posts competing with commercial service pages were either rewritten as clearly informational content (with the commercial keyword removed from title, H1, and permalink) or folded in as FAQ sections on the service page itself.
The Results: Within weeks of deleting the redundant suburb variants, the primary suburb page — which hadn’t been indexed at all — was indexed without any manual “Request Indexing” action in Search Console. Ranking volatility on the core “emergency plumber” term dropped as the number of competing internal pages fell from 30-plus to a handful of clearly differentiated ones.
The Lesson: Google’s index isn’t rewarding page count — it’s rewarding clarity. A site with 50 unique, well-targeted pages will consistently outperform one with 500 overlapping ones.
Industry Application: Any trade or local service business tempted to build a page for every suburb, every service variation, and every long-tail phrase should map keywords to pages first, then build — not the other way around.
Case Study: Brightline Commercial Cleaning — Retail & Hospitality Services — Regional Victoria
The Challenge: A commercial cleaning operator had built hundreds of “[service] + suburb” pages, including a run of “church cleaning [suburb]” pages across towns where there wasn’t a single church client, or in some cases, barely a church. The pages had no meaningful content difference beyond the suburb name swapped in a template.
The Strategy: We cross-checked each suburb page against actual local search volume and existing client locations. Pages targeting suburbs with genuinely no demand were removed. The remaining suburb pages were rewritten with specific, verifiable local detail — recent job types, real building names where permission allowed, and photos on location — rather than templated filler.
The Results: Indexation of the retained pages improved, and the business began appearing in ChatGPT responses for a couple of the niche cleaning terms it had previously been invisible for, generating enquiries traced back to AI-referred traffic in the contact form notes.
The Lesson: A templated suburb page with the town name swapped out isn’t a real page in Google’s eyes — it’s a duplicate with a find-and-replace. Genuine local detail is what earns the page a place in the index.
Industry Application: Retail and hospitality operators expanding into multiple locations should resist the urge to auto-generate location pages. Every page needs its own reason to exist.
Case Study: Coastal Removals & Storage — Trades & Logistics — Newcastle to Sydney corridor
The Challenge: A removalist wanted visibility across roughly 30 towns along a single transport corridor but had no existing authority to support that many pages at once.
The Strategy: Rather than publishing all 30 town pages simultaneously, we built them progressively, tied to actual booking demand, and gave each one a genuinely unique visual and written identity — the business owner’s van photographed in front of a locally recognisable landmark (Sydney Harbour Bridge for the Sydney page, Newcastle’s foreshore for the Newcastle page), paired with route-specific detail rather than generic “we move your furniture” copy. Alongside this, we built a small cluster of separate exact-match domains around adjacent niche terms (e.g., a domain focused on interstate removals) that referenced the client business within directory-style listicles, generating additional brand mentions.
The Results: Search visibility and enquiry volume increased in stages as each town page earned its own indexation and authority, rather than diluting a single domain’s link equity across 30 pages published in one hit.
The Lesson: Multi-location content should be sequenced to match the site’s actual authority and indexation velocity, not published as a single bulk exercise.
Industry Application: Any business covering a service corridor or multiple branch locations — removalists, mobile trades, franchise-style operators — should stagger location page rollouts and give each one distinct, verifiable content.
Implementation Guide
Step 1: Audit every page on the site
What it looks like: Export a full URL list (WordPress export or a Screaming Frog crawl), then connect it to Google Search Console and Google Analytics to see impressions, clicks, and indexation status for each URL.
Pro tip: Start with the pages already flagged “Crawled, not indexed” or “Discovered, not indexed” in Search Console — these are your highest-priority list and don’t need re-crawling to identify.
Step 2: Map every keyword to exactly one page
What it looks like: Build a simple spreadsheet — keyword, intended page, current pages competing for it. Any keyword with more than one page attached is a cannibalisation risk.
Pro tip: Keep blog content informational and service pages commercial. If a blog title contains your money keyword (e.g., “emergency plumber Sydney”), that blog is very likely fighting your service page for the same ranking spot.
Step 3: Cull pages with no genuine search demand
What it looks like: Delete or de-index suburb, service, or blog pages that don’t correspond to real, measurable search volume or local demand — not every suburb needs its own page.
Pro tip: When multiple suburb variants exist for one area, keep the strongest performer and 301-redirect the rest. Often the “parent” page will get indexed on its own once the duplicates are gone.
Step 4: Consolidate and rewrite what’s left
What it looks like: Where content genuinely overlaps, merge it into one stronger page. Add FAQ sections, embedded video, and specific, personalised detail — the kind of insight a generic AI answer can’t produce, such as “we’ve inspected 500 hot water systems in this suburb; here’s what we consistently find.”
Pro tip: Use TF-IDF analysis on the current top 10 ranking pages for your target term to identify commonly used related terms, then work them in naturally rather than stuffing keywords.
Step 5: Fix the technical foundation before republishing
What it looks like: Check page load speed, hosting, and plugin compatibility. A 10-minute page load time (yes, this happens) will undermine every other fix you make.
Pro tip: Before choosing a page builder, hosting provider, or plugin stack for a new build, confirm it can handle the number of pages you’re planning — cheap hosting and heavy plugin stacks are a common hidden cause of crawl and indexation problems.
Step 6: Judge success by rankings and enquiries, not page count
What it looks like: Track keyword rankings and lead volume in Search Console and your CRM or call-tracking tool, not just how many pages are live.
Pro tip: A site with fewer, cleaner pages that convert well will always beat a bloated one on return per dollar spent building it.
Common Pitfalls & Solutions
| Pitfall | Fix | Prevention |
|---|---|---|
| Publishing a suburb page for every suburb in a region regardless of demand | Cull pages for suburbs with no measurable search volume; keep one strong page per genuine demand area | Check search volume and local intent before building any location page |
| Multiple pages (blog + service page) targeting the same commercial keyword | Consolidate into one page; convert the other into genuinely informational content or an FAQ | Map keywords to pages before writing a single word |
| Templated pages with only the location name swapped | Rewrite with real, verifiable local detail — photos, specific jobs, local landmarks | Treat every location page as a standalone piece of content, not a template |
| Publishing hundreds of pages at once on a brand-new domain | Stagger the rollout in line with the site’s actual authority and indexation rate | Build a URL and content plan first, then release in phases tied to performance |
| Choosing cheap hosting or a heavy plugin stack that slows load times | Migrate to hosting that can handle the page count; audit and remove unnecessary plugins | Confirm hosting and page-builder capacity before launch, not after |
| Assuming more blogs always means more authority | Focus on content depth and uniqueness over volume | Publish blogs at a rate matched to genuine indexation velocity, not a fixed content calendar |
Practical Exercise
- Quick Win (5 mins): Open Google Search Console → Pages report and check how many URLs are listed under “Crawled, not indexed” and “Discovered, not indexed.” If it’s more than 10% of your total published pages, that’s your starting point.
- Deep Dive (30 mins): Export your full page list and build a simple keyword-to-page map. Flag every keyword with more than one page attached — that’s your cannibalisation hit list for the next content sprint.
- Real Business Application:
- Trades: Check whether you have more than one page per suburb (north/south/east/west variants) and consolidate down to the strongest.
- Professional Services: Check whether educational blog content is unintentionally competing with your core service pages for the same search terms.
- Retail & Hospitality: Audit location pages for genuine uniqueness rather than templated copy with the suburb name swapped.
Keyword Intelligence Table
Note: Ahrefs and Firecrawl data sources were unavailable this session. The figures below are CMO Eugene domain-knowledge estimates based on typical Australian trades search patterns and general industry benchmarks — treat as directional, not precise, and verify with live Ahrefs/Search Console data before committing budget.
| Keyword | Est. Monthly Volume (AU) | Est. Difficulty | Intent | Notes |
|---|---|---|---|---|
| emergency plumber sydney | 1,000–1,500 | High | Commercial | One page only — service page, not a blog |
| plumber near me | 4,000+ | High | Commercial | Highly location-dependent via Map Pack; national volume, hyper-local intent |
| blocked drain melbourne | 300–500 | Medium | Commercial | Suburb-modified variants often outperform if genuine demand exists |
| how to choose an emergency plumber | 50–100 | Low | Informational | Keep as an FAQ or supporting content, not a standalone page competing with your service page |
| commercial cleaning [suburb] | 20–80 (suburb-dependent) | Low–Medium | Commercial | Only build the page if the suburb clears roughly 50+ searches/month |
| interstate removalist sydney to newcastle | 50–150 | Low–Medium | Commercial | Strong candidate for a dedicated corridor page given specific, low-competition intent |
Independent market data supports the commercial value of getting this right: one Sydney-focused SEO guide notes the average plumbing job in that market runs <cite index=”2-1″>between $350 and $850</cite>, and that <cite index=”2-1″>a single page-one ranking for a competitive term like “emergency plumber Sydney” can generate 40 to 60 calls a month</cite> — which underlines why it’s worth having exactly one clearly optimised page competing for that term, not thirty.
Tools & Resources
- Google Search Console — free, essential for identifying “Crawled, not indexed” and “Discovered, not indexed” pages before any other tool
- Screaming Frog — free (up to 500 URLs) / paid for larger crawls; use to export a full site structure and cross-reference with Search Console and Analytics
- Ahrefs — paid; use for keyword volume, difficulty, and TF-IDF-style competitor term analysis when deciding which pages to keep, cull, or build
- 301 redirect mapping template — internal Ranked resource; use when consolidating duplicate suburb or service pages
- Keyword-to-page mapping spreadsheet — internal Ranked resource; the single most useful document for preventing cannibalisation on any new build
Module Summary
- Index bloat isn’t a sign of a site working hard — it’s usually a sign of a site confusing Google, and increasingly, confusing AI answer engines too.
- One keyword should map to exactly one page. If you’re not sure which page owns a term, Google isn’t sure either.
- Culling pages isn’t wasted work — deleting genuinely low-value pages is often the fastest way to get your best pages indexed.
- Judge a website on rankings and enquiries, not on how many pages it has.
CMO Eugene — Ranked Digital Marketing, Australia
