| name | programmatic-seo |
| description | Use when the user wants to build programmatic SEO pages, template-driven content at scale, location pages, comparison pages, or data-driven landing pages. Also use when the user mentions 'programmatic SEO', 'pSEO', 'template pages', 'location pages', 'scalable SEO', 'data-driven pages', or 'SEO at scale'. |
Programmatic SEO
Expert knowledge for building template-driven pages at scale that rank for long-tail keywords — without thin content penalties.
What Programmatic SEO Is
pSEO = a template + a data source = N pages, where N is large and each page targets a specific long-tail keyword cluster.
The constraint: Each page must provide enough unique value to earn a ranking. Google's helpful content system penalizes pages that exist only to capture query volume without serving user intent.
Keyword Patterns That Work
| Pattern | Example | Data Source |
|---|
| [Tool] for [Use Case] | "Notion for project management" | Product categories |
| [Tool] vs [Tool] | "Webflow vs Framer" | Competitor pairs |
| Best [Category] in [Location] | "Best accountants in Montreal" | Business directory |
| [Category] for [Industry] | "CRM for real estate" | Industry list |
| How to [Task] in [Tool] | "How to export CSV in Airtable" | Feature × tool matrix |
| [Job Title] tools | "Tools for growth marketers" | Job title taxonomy |
| [Adjective] [Category] | "Free email marketing tools" | Attribute × category |
Research method: Use Google Autocomplete, People Also Ask, and keyword tools to validate search volume before building. A pattern with < 100 monthly searches per variant is rarely worth the infrastructure.
Data Source Types
| Type | Example | Freshness need |
|---|
| Internal product data | Feature list, pricing, integrations | Real-time or daily sync |
| Public database | Crunchbase, Census, OpenStreetMap | Monthly refresh |
| Scraped / licensed | G2 reviews, Trustpilot ratings | Weekly |
| User-submitted | Directory listings, job board | Real-time |
| Editorial | Curated list with descriptions | Manual, quarterly |
Quality signal: The more proprietary and fresh your data, the stronger the moat. Anyone can build on public data. Proprietary data is defensible.
The 60/40 Rule
60% template: Structure, navigation, calls-to-action, meta tags, schema markup.
40% unique content: At minimum, 40% of each page's content must be unique to that page.
Thin content threshold: pages under ~300 words with > 80% templated text are at risk of manual action or algorithmic demotion.
Unique content strategies:
- Dynamic statistics pulled from your data source
- User reviews or ratings specific to that variant
- Editorially written introductions for high-volume variants
- Auto-generated comparisons or summaries derived from structured data
- "Last updated" timestamps showing freshness
Thin Content Avoidance
Signals that trigger thin content penalties:
- Identical or near-identical meta descriptions across pages
- Body text that only changes the keyword
- No original data, analysis, or editorial perspective
- Pages indexed faster than they can be reviewed (rapid mass publication)
Safe publication cadence: For new pSEO sites, publish 50–100 pages initially, monitor indexation and ranking for 30 days, then scale. Don't publish 10,000 pages on day one.
Canonicalization
- Every page: self-canonical (
<link rel="canonical" href="[this page's URL]"/>)
- Faceted/filtered variants: canonical points to the root template page
- Duplicate content (e.g., location A and location B with identical content): use noindex on the weaker page or consolidate
Internal Linking Strategy
pSEO pages are typically thin on authority. Internal links from high-authority pages pass PageRank.
Linking patterns:
- Hub-and-spoke: Create a category hub page that links to all variants
- Comparison clusters: Each vs. page links to the two individual tool pages
- Blog to pSEO: In-content links from blog posts to relevant pSEO pages
- Breadcrumbs: Structural breadcrumbs on every pSEO page (also add BreadcrumbList schema)
Rule: No pSEO page should exist with zero internal links pointing to it from your site.
Page Quality Scoring
Score each page variant before publishing:
Quality Score = (Unique Content % × 0.4)
+ (Data Freshness Score × 0.3)
+ (User Signals Proxy × 0.2)
+ (Internal Link Count × 0.1)
Pages scoring < 0.5 should be revised or held from indexing.
Phased Indexation
| Phase | Action |
|---|
| 1. Seed (50–100 pages) | Publish, submit to Search Console, monitor for indexation |
| 2. Assess (30 days) | Check: % indexed, ranking positions, click-through rate |
| 3. Optimize | Improve lowest-quality pages before scaling |
| 4. Scale | Publish remaining variants in batches of 500–1,000 |
| 5. Maintain | Monthly freshness update on all pages (update dates, refresh data) |
Common Rationalizations
| Rationalization | Reality |
|---|
| "More pages = more traffic" | Thin pages dilute crawl budget and trigger algorithmic penalties. Quality > quantity. |
| "We'll add unique content later" | Thin content that gets indexed as thin stays thin. Build the 40% unique content into the template from day one. |
| "All our location pages say the same thing with the city name swapped" | That's a thin content penalty waiting to happen. Each location needs genuinely different data. |
| "We published 10,000 pages in a week and nothing is ranking" | You overwhelmed Google's quality assessment. Phased indexation exists for this reason. |
Verification
