Blog Strategy · How to
Programmatic SEO: how to do it without scaled content risk
Part of the guide SEO blog strategy: the whole plan on one page
Programmatic SEO is building many pages from one template and a dataset, so each page targets one variation of a repeatable search, such as "[service] in [city]" or "[tool] integration with [app]". It works when every page carries data a reader cannot get elsewhere. When the pages differ only by a swapped word, Google treats it as scaled content abuse, whatever tool made them.
What programmatic SEO is
Programmatic SEO (pSEO) means generating pages at scale from structured data: a spreadsheet or database supplies the variables, a template turns each row into a page. A currency site with one page per currency pair, a job board with one page per role and city, and a software company with one page per integration are all programmatic.
It fits searches that share a pattern and differ by one or two variables. If a single, well written article could answer every variation, you do not need pSEO; you need one good page. Where it sits in a wider content plan is covered in the SEO blog strategy guide.
The risk: what Google calls scaled content abuse
Google's spam policies define scaled content abuse as "when many pages are generated for the primary purpose of manipulating search rankings and not helping users", and they add "no matter how it's created". The examples include:
- "Using generative AI tools or other similar tools to generate many pages without adding value for users"
- "Scraping feeds, search results, or other content to generate many pages (including through automated transformations like synonymizing, translating, or other obfuscation techniques)"
- "Creating many pages where the content makes little or no sense to a reader but contains search keywords"
Since May 2026, Google's changelog states these spam policies "also apply to generative AI responses in Google Search", so the same pages can be excluded from AI Overviews and AI Mode too.
This is not theoretical. Byword, an AI article tool that sells programmatic campaigns, states in its own robots.txt that its /resources/ section "was retired during Google 'scaled content abuse' manual-action recovery and now returns 410 Gone" (checked 16 November 2026). We mention it because the vendor documented it openly, and because it shows the risk applies to experienced SEO teams, not only beginners.
The test every programmatic page must pass
Before building anything, ask of a sample page: if search engines did not exist, would a real visitor be glad this exact page exists? Then check it against this bar:
| Check | Passes | Fails |
|---|---|---|
| Unique data | Each page shows facts specific to its variable (prices, counts, specs, local details) | Only the city or product name changes |
| Answer value | The page answers the query better than a generic page could | A generic page would serve the reader just as well |
| Real demand | People search this variation, or would land on it from your site | Variations invented to fill a grid |
| Completeness | Every section has data; empty fields hide or the page is not published | "N/A" and placeholder text in visible sections |
| Accuracy | Data is current and sourced | Data copied once and never updated |
If a page fails any row, either fix the template and data or do not publish that page. Fewer, stronger pages beat a large grid with a thin tail.
Programmatic SEO examples: good and bad patterns
The same pattern can be helpful or abusive depending on the data behind it:
| Pattern | Helpful version | Abusive version |
|---|---|---|
| [Service] in [city] | Real local team, service area map, local prices, licences, reviews from that area | Same text with the city name swapped across 500 towns you do not serve |
| [App] integration with [app] | What syncs, setup steps, limits, screenshots of that specific connection | "Connect X with Y easily" with nothing specific to either app |
| [A] vs [B] for many pairs | Real specs and prices for both, a verdict per pair | AI text generated for every pair in a list, unchecked |
| [Term] definitions | Glossary entries written or reviewed by someone who knows the field | Scraped definitions reworded by a synonym tool |
For live examples from real sites and the tools that build them, see programmatic SEO tools and examples.
How to build a programmatic SEO project, step by step
- Find the pattern. List the head term and its modifiers (city, product, use case, integration). Confirm with Search Console, autocomplete and keyword tools that people search the variations, not just the head term.
- Search five variations by hand. If page one is all big directories or official sources, your version needs something they lack, or pick another pattern.
- Build the dataset first. One row per page, one column per fact the page will show. This is where the value lives. Sources include your own product data, public datasets you may reuse, and research you pay for. If you cannot fill the columns for a row, that row does not become a page.
- Write the template like an article. An answer paragraph that uses the data, H2s for the real sub questions, a table of the row's facts, conditional sections that appear only when data exists, and an FAQ that varies per page.
- Write five pages by hand first. Then make the template reproduce them. If the template cannot match your hand written version, it is not ready.
- Add hub pages. A hub per group (for example, all cities in one state) linking to each page, and each page linking back to its hub and to two or three neighbours. The internal linking strategy guide explains why orphaned pages struggle.
- Publish in batches. Start with your strongest 20 to 50 rows. Watch indexing for a few weeks before adding more.
- Keep the data fresh. Schedule updates, change the visible date only when the data changes, and keep sitemap lastmod honest.
Technical points that trip up pSEO sites
- Sitemaps: Google limits one sitemap file to 50,000 URLs or 50 MB uncompressed; split larger sets and use a sitemap index.
- Canonicals: near duplicate pages (for example "plumber in Springfield" in two states) each need a canonical pointing at themselves, and truly duplicate variations should be merged.
- Indexing: "Crawled, currently not indexed" on a large share of your pSEO pages is the clearest early signal that Google sees them as low value. Google calls the status common, so judge the pattern, not one URL.
- Rendering: the page text must be in the HTML response, not loaded later by JavaScript from your database.
- Pruning: pages that stay unindexed or get no impressions after a few months should be improved, merged or removed. See the guide on content decay and refreshing.
Where AI fits, and where it does not
AI is useful for drafting the template, writing the parts of each page that interpret real data, and checking pages for gaps. It is risky as the only source of the page's substance: Google's first scaled content example is generative AI pages "without adding value for users". Feed the model the row's facts and require it to use them; never ask it to invent local details, prices or reviews.
Postbing has a narrow programmatic feature: service by location pages with a lead form and a hub page, published to your CMS as pages, at 1 credit per page. It suits a local or regional service business with a real service area. It is not a general pSEO platform for arbitrary datasets, and those pages are only as good as the local facts you give it. See pricing and credits.
FAQ
What is programmatic SEO?
It is creating many pages from one template and a dataset, each targeting a variation of a repeatable search like "[service] in [city]". Each page should carry data specific to its variation.
Is programmatic SEO against Google's guidelines?
Not by itself. Google's spam policies target pages made "for the primary purpose of manipulating search rankings and not helping users", no matter how they are created. Pages with unique, useful data per variation are fine.
Can I use AI for programmatic SEO?
Yes, to draft templates and to write around real data you supply. Generating many pages with AI "without adding value for users" is one of Google's named examples of scaled content abuse.
How many programmatic pages should I publish at once?
Start with your strongest 20 to 50 and watch indexing for a few weeks. If most are indexed and getting impressions, add more; if many sit unindexed, fix the template and data first.
What is a good first programmatic SEO project?
Integration or "[product] for [use case]" pages, where you hold real product data for each variation. Location pages only work if you genuinely serve those places and have local details.
Sources
- Google: Spam policies for Google web search: scaled content abuse definition and examples, accessed 16 November 2026
- Google Search Central: documentation updates: spam policies apply to generative AI responses (15 May 2026 entry), accessed 16 November 2026
- Google: Build and submit a sitemap: 50,000 URL and 50 MB limits, sitemap index, accessed 16 November 2026
- Google: Consolidate duplicate URLs: canonical methods, accessed 16 November 2026
- Google Search Console Help: Page indexing report: "Crawled, currently not indexed" status, accessed 16 November 2026
- Byword robots.txt: comment on /resources/ retired during scaled content abuse manual action recovery; /resources/ returns HTTP 410, checked 16 November 2026
Related reading
Last updated: November 16, 2026. Spotted an error? Email support@postbing.com and we will fix it. Editorial policy.
Try Postbing free
Postbing reads your site, plans a month of topics from what your customers search for, writes in your voice and publishes to WordPress, Shopify, Webflow or Ghost. Your first article is free.
Try Postbing freeNo card needed. No link exchanges, ever.