Content freshness
Definition
Content freshness is how recently a page was published or substantively updated, as signaled to AI assistants and search engines through `dateModified`, visible publish dates, and changed body content. Retrieval-grounded AI engines — Perplexity, Google AI Overviews, Bing Chat, Gemini — weight freshness heavily when choosing citation sources for time-sensitive queries.
How it works
AI engines and search crawlers detect freshness through three stacked signals:
-
dateModifiedin JSON-LD: the canonical machine-readable freshness signal. Set on every Article, BlogPosting, and NewsArticle. -
Visible publish/update date in the rendered page, typically in the byline or page footer. Reinforces the JSON-LD and is what users see.
-
Substantive content change: AI crawlers compare successive crawls. Updating only the date without touching the body is increasingly detected and discounted.
For time-sensitive queries — pricing, benchmarks, industry data, "best X in 2026", trending news — retrieval-grounded engines preferentially cite the freshest credible source. For evergreen queries (definitions, how-to guides), freshness still matters but the penalty for older content is gentler.
Freshness vs evergreen content
Not every page needs to be fresh. Two strategies coexist:
Fresh content (industry reports, pricing pages, "best X this year" comparisons, news commentary) requires aggressive refresh — quarterly at minimum, monthly when the underlying data shifts. AI engines materially prefer recent dates here.
Evergreen content (definitions, how-to guides, technical reference) can sit for 6–12 months between updates as long as the substance remains accurate. AI engines accept older dateModified here when the content matches the prompt precisely.
The mistake brands make is treating all content as evergreen. A pricing page or competitor-comparison page that hasn't been touched in two years gets bypassed for fresher alternatives — even when the underlying facts haven't changed.
Quarterly
Recommended refresh cadence for time-sensitive AI-cited pages (pricing, benchmarks, comparisons)
Indexly best practice
6–12 months
Acceptable refresh interval for evergreen content (definitions, foundational how-tos)
Indexly best practice
Detected
Zero-change `dateModified` updates are increasingly identified and discounted by AI crawlers
Indexly research, 2026
Why it matters
Freshness is the lowest-effort, highest-leverage AI ranking signal. A 30-minute pass to update dateModified, refresh a few stat lines, and re-publish often lifts citation rate noticeably for time-sensitive pages.
Conversely, stale content is the most common silent killer of AI visibility. A page that earned citations consistently can drop off the source list within weeks of falling behind the freshness curve set by competitors. Monitoring dateModified against citation rate exposes this drift early.
How to implement it
Five tactics for credible freshness signaling:
-
Set
dateModifiedin Article JSON-LD on every page. Update it whenever the content materially changes — not when you just touch a typo. -
Show a visible "Last updated" date in the page byline or footer. Users trust it; AI engines read it as a reinforcing signal.
-
Make substantive updates, not cosmetic ones. Refresh a stat, add a new example, replace an outdated screenshot. Crawlers detect zero-change updates and discount them.
-
Set a refresh cadence per content type. Pricing monthly, benchmarks quarterly, definitions semi-annually, foundational how-tos annually.
-
Audit stale citations regularly. Monitor citation rate against
dateModified. Pages that earned citations 6+ months ago and now sit untouched are the priority refresh queue.
Frequently asked questions
Does just changing dateModified boost AI citations?
No — and increasingly the opposite. AI crawlers compare successive snapshots and detect zero-change updates. Bumping the date without substantive edits is read as a freshness hack and discounted. Real updates — new stats, refined examples, restructured sections — are what move the needle.
Where should the dateModified live in my markup?
Inside Article (or BlogPosting / NewsArticle) JSON-LD schema, with the visible "Last updated" date rendered in the byline. Both signals reinforce each other — JSON-LD is machine-readable, visible date is user-trust.
How fresh do AI engines expect content to be?
Depends on the query. "Best CRM in 2026" expects content updated within the last few months; "What is a CRM" tolerates a 12-month-old definition page. Match refresh cadence to query time-sensitivity.
Does freshness apply to all AI engines equally?
No. Retrieval-grounded engines (Perplexity, Google AI Overviews, Bing Chat, retrieval-augmented ChatGPT and Claude) weight freshness heavily. Pure training-grounded mentions are frozen at the model's training cutoff and don't respond to freshness updates between training runs.
Should I add publish AND update dates?
Yes. AI engines and Google both read both. Original publish date establishes when the topic was first covered (a relevance signal); update date establishes current freshness. Pages with only publish dates and no update markers age out of citations faster.
AI citation optimization
AI citation optimization is the practice of structuring web content so AI assistants — ChatGPT, Claude, Perplexity, Gemini, Bing Chat, and Google AI Overviews — choose to cite it as a source in their generated answers. It is the citation-layer counterpart to traditional SEO link building and a core discipline within Generative Engine Optimization (GEO).
AI content strategy
AI content strategy is the deliberate plan for producing, structuring, and maintaining content so it earns visibility inside AI assistants — ChatGPT, Claude, Perplexity, Gemini, Grok, and Google AI Overviews. It rebuilds traditional editorial planning around the way LLMs choose, cite, and synthesize sources rather than the way Google ranks links.
Schema markup
Schema markup is structured data added to web pages using the schema.org vocabulary that tells search engines and AI systems exactly what the content represents — a product, an article, a recipe, an FAQ, a person. It powers rich results in Google, drives entity understanding in knowledge graphs, and increasingly determines whether content is cited in AI Overviews and LLM-generated answers.
Generative engine optimization (GEO)
Generative engine optimization (GEO) is the practice of structuring content and brand presence so that AI systems like ChatGPT, Claude, Perplexity, and Google AI Overviews cite, quote, or recommend it when generating answers. Unlike traditional SEO, which competes for ranked positions in a list of links, GEO competes for inclusion inside the answer itself.
AI content ranking
AI content ranking is the relative position your content holds in AI-generated answers — first-cited, mid-list, or never surfaced — across ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews. Where traditional SEO ranking is a numbered position in a SERP, AI content ranking is order-of-mention and citation-prominence inside a synthesized answer.
Retrieval-augmented generation (RAG)
Retrieval-augmented generation (RAG) is an AI architecture that gives a large language model real-time access to external documents at query time — retrieving relevant passages from a vector database or search index and inserting them into the model's context before it generates a response. RAG is the foundation of modern AI search and the most effective technique for reducing hallucination.