Presented by Omnia
Search is changing. If you’re part of a small team that’s expected to do more with less, Omnia is for you.
Omnia helps you track your brand’s visibility across the leading AI search platforms: ChatGPT, Perplexity, Google AI Overviews, and Google AI Mode. See how you compare with competitors, which sources these platforms cite, and where you have opportunities to improve your presence.
Omnia also helps you make concrete improvements, with an autonomous agent and built-in project management. Turn data into action, prioritize optimizations, and keep your AEO strategy on track without overwhelming your team.
Try Omnia free for 14 days and get 20% off your first quarter when you decide to upgrade.
TL;DR
- Yes, Programmatic SEO still works, but the margin for getting it wrong has disappeared.
- Google tightened its policies against scaled content abuse. It no longer matters whether the content is produced by a person, a template, or an AI. What matters is whether it exists primarily to manipulate rankings.
- Projects that keep growing share three traits: proprietary or hard-to-replicate data, pages that solve a specific task, and a brand with topical authority.
- The projects hit hardest follow the same pattern: near-identical templates, interchangeable variables (city, keyword, product) and generated text with no new information.
- AI Overviews and conversational assistants are cutting clicks on simple informational queries, precisely the territory where low-quality pSEO used to thrive.
- The strategic question has changed. It used to be “how many pages can we publish?” Now it’s “how many pages deserve to exist?”
What Programmatic SEO is (and isn’t)
Programmatic SEO (pSEO) is the creation of pages at scale from a template + a database, aimed at covering long-tail searches with a repeatable pattern.
Some classic examples:
| Search pattern | Type of site | Conceptual example |
|---|---|---|
| “[app A] + [app B] integration” | SaaS | Zapier integration pages |
| “convert [currency] to [currency]” | Fintech | Wise or XE converters |
| “things to do in [city]” | Travel | Tripadvisor |
| “houses for sale in [neighborhood]” | Real estate | Zillow, local listing sites |
| “[document type] template” | Design tools | Canva template galleries |
| “alternatives to [software]” | B2B reviews | G2, Capterra |
What pSEO is not: duplicating an article and swapping out the city name, generating thousands of texts with AI with no data behind them, or publishing pages “just in case” one of them ranks.
The difference between the two doesn’t lie in the technique. It lies in whether the page has a reason to exist beyond capturing traffic.
What changed: the timeline that redefined pSEO
2022–2023: the helpful content system
Google introduced the Helpful Content System, a site-wide signal that penalizes domains with a high proportion of content created “for search engines rather than people.” pSEO sites with thousands of thin pages were among the first to feel it, because the problem affected the perception of the entire domain, not just isolated pages.
March 2024: core update and new spam policies
This was the turning point. Google:
- Folded the helpful content system into its core algorithm.
- Launched the policy against scaled content abuse, targeting the mass production of low-value pages regardless of how they’re created.
- Launched the policy against site reputation abuse, aimed at third parties publishing content on authoritative domains to exploit their ranking.
- Cracked down on the use of expired domains repurposed to manipulate results.
By the end of the rollout, Google said it had reduced low-quality, unoriginal content in search results by about 45%. During those weeks, many sites built on mass AI-generated content were de-indexed entirely through manual actions.
2024–2025: AI in the SERP
The expansion of AI Overviews changed the economics of the click. Queries like “how much is 100 dollars in euros” or “what is X” can now be answered directly on the results page. This hits low-value informational pSEO hard, while leaving transactional, comparison and tool-based formats relatively better off.
The generative AI paradox
AI drove the cost of producing text down to nearly zero. The result is that text stopped being a competitive advantage. If anyone can generate 10,000 pages in an afternoon, Google needs signals other than volume to decide what to rank, and it finds them in data, experience, brand, and user behavior.
What still works
Proprietary or hard-to-get data
This is the single most decisive factor. The pSEO pages that hold up rely on information users can’t find anywhere else:
- Real prices, inventory, or up-to-date availability.
- Aggregated usage data from the platform itself.
- Reviews, ratings, and user-generated content.
- Public data processed in a useful way, for instance cross-referenced, cleaned, or visualized.
Quick test: if a competitor can replicate your database in a weekend, your advantage is fragile.
Pages that work as a tool
Converters, calculators, comparison engines, filtered search tools, and generators. Users come to do something, not just read. This format holds up better against AI Overviews because the interaction itself is the value.
Transactional or commercial intent
“Billing software for restaurants,” “labor lawyers in Córdoba,” or “flights from Madrid to Lima in March.” These are searches where the user needs to see concrete options, compare, and act, something an AI summary doesn’t fully replace.
Templates with real variation
The best implementations don’t just swap out a single variable. They change the content structure based on available data:
- Sections that appear or are hidden depending on the information for each entity.
- Contextual recommendations: an integration page between two marketing apps isn’t the same as one between two accounting apps.
- Hand-written editorial blocks for the pages with the highest potential.
Pre-existing topical authority
pSEO works better as an amplifier for a brand that already has relevance in a topic than as a strategy for a brand-new domain. A site with no track record that suddenly publishes 50,000 pages sends exactly the signal that spam policies are designed to catch.
What no longer works
| Tactic | Why it fails today |
|---|---|
| “City + service” pages with identical text | Duplicate content at scale with no differentiating value |
| AI-generated articles with no data or review | Falls squarely within the definition of scaled content |
| Publishing thousands of pages at once on a new domain | A classic spam pattern, with no trust signals |
| Pages with no real demand (“just in case”) | Dilute site quality and eat up crawl budget |
| pSEO content hosted on third-party domains | Site reputation abuse, subject to manual actions |
| Simple informational answers (“what is,” “how much is”) | Absorbed by AI Overviews and featured snippets |
Evaluation framework: is your pSEO project worth it?
Before investing, answer these seven questions. If you answer “no” to three or more, the project is likely high-risk.
- Is there real demand? Is there measurable search volume for the patterns you chose, even if it’s low per page?
- Do you have data others don’t? Or at least public data processed in a clearly superior way?
- Does each page solve something on its own? If a user arrives from Google, do they leave with their problem solved?
- Are the pages genuinely different from one another? Compare them in pairs. If only one word changes, that’s a problem.
- Does the intent hold up against AI? Does the user need to interact, compare, or buy, or is a short answer enough?
- Does your domain have authority on the topic? Or are you starting from scratch?
- Can you maintain it? Outdated data turns assets into liabilities.
Implementation best practices
Before launching
- Validate the pattern with a pilot of 50 to 200 pages before scaling to thousands.
- Set a minimum data threshold: if an entity doesn’t have enough information, the page shouldn’t be generated, or should be marked
noindex. - Design the template with the user in mind, not keyword density.
During rollout
- Publish gradually.
- Build a logical internal linking structure with hubs by category, location, or type, so no pages end up orphaned.
- Use structured data (Schema) where it applies: products, reviews, FAQs, local organizations.
After launch
- Monitor the ratio of indexed versus discovered-but-not-indexed pages in Search Console. It’s the clearest early signal that Google doesn’t see value.
- Don’t be afraid to prune: pages with no impressions after several months usually subtract more than they add.
- Manually enrich the pages that gain traction.
Metrics that matter
| Metric | What it indicates |
|---|---|
| Indexation rate | Whether Google considers the pages worth including in the index |
| Pages with at least one impression | Real coverage of the pattern |
| Clicks per indexed page | Efficiency of the project relative to pages published |
| Engagement (time, interactions, scroll) | Whether the page fulfills the intent |
| Assisted conversions | Business value beyond traffic |
| Performance after core updates | Resilience of the strategy |
A sustained drop in the indexation rate after scaling usually foreshadows domain-level problems.
The role of AI: an ally, not a shortcut
Generative AI is useful within a pSEO project when it:
- Structures and cleans messy data.
- Summarizes verified information, such as reviews, specs, or technical data.
- Generates copy variations based on real data, with human review.
- Classifies and tags entities to improve internal linking.
It becomes a risk when it replaces data instead of presenting it. The rule of thumb is simple: AI can write the copy, but it can’t be the source.
Outlook: where pSEO is headed
- From volume to selectivity. Sites with fewer but better pages, and a high indexation rate, will come out ahead.
- From text to product. pSEO increasingly resembles product development: filters, tools, and interactive experiences.
- Visibility beyond Google. Pages with unique, structured data have a better shot at being cited by answer engines and AI assistants, opening up a new layer of visibility.
- Brand as a filter. Amid the overabundance of content, trust in the source carries more and more weight in what gets ranked and what gets cited.
Conclusion
Programmatic SEO isn’t dead. What died was its lazy version.
The technique of combining templates and data to capture long-tail demand remains one of the most efficient ways to scale organic traffic. But today it only works under one non-negotiable condition: every page has to justify its existence to the user, not to the algorithm.
For brands with their own data, topical authority, and a willingness to treat pSEO as a product, the opportunity remains enormous. For those who see it as a shortcut, the risk is no longer just failing to rank, but jeopardizing the visibility of the entire domain.
The question is no longer whether Programmatic SEO works, but whether your project has something worth scaling.