Why the dates on your page are an SEO signal
Search engines and AI answer systems care about when content was created and when it was last updated. For many queries — news, prices, software versions, statistics, best-of lists, anything that changes over time — recency is part of relevance. A page that was accurate three years ago may be quietly wrong today, and engines that surface stale information lose trust, so they pay attention to the dates a page declares about itself. This checker fetches a page and pulls every published and modified date it can find from meta tags, time elements, and structured data, then tells you what dates the page is actually broadcasting and whether the content looks fresh or stale.
The catch is that the date a reader sees on the page is not always the date a machine reads. Pages declare dates in several places, those places can disagree, and a visible Last updated line in the body means nothing to a crawler if it is not backed by a machine-readable date. Many sites think they are signaling freshness when they are not, or are accidentally signaling that a recently refreshed article is years old. This tool exists to show you the gap between what your page says to humans and what it says to machines, so you can make the two agree and make the freshest signal the true one.
Where dates live on a page
There are several common homes for a date, and a well-built page uses them consistently. Open Graph article tags carry a published time and a modified time that social platforms and some crawlers read. Structured data — the Article schema in JSON-LD — carries datePublished and dateModified fields that are among the strongest, most explicit signals you can give. The HTML time element can mark up a visible date with a machine-readable datetime attribute. Older or simpler sites sometimes use a meta tag for the article date or rely on the visible text alone. The checker looks across all of these and collects whatever it finds.
Because there are so many places to put a date, the real risk is inconsistency. The Open Graph modified time might say one thing, the schema dateModified another, and the visible body a third. A crawler then has to guess which to believe, and you have lost control of the signal. The tool lays out every date it found and its source so you can see at a glance whether they agree. When they all point to the same recent update, your freshness signal is strong and unambiguous. When they scatter, you have a cleanup job that is usually quick once you know where the discrepancies are.
What the checker reports
Run the tool against a URL and it reports the published date and the modified date it detected, naming the source of each — Open Graph, schema, a time element, or a meta tag. It compares the dates where more than one source exists and flags disagreements. It then judges freshness by comparing the most recent reliable date against today, so you get a plain read on whether the content is recent, aging, or clearly stale. A page last touched many years ago on a topic that changes yearly will be flagged as stale; a page updated last month will read as fresh.
Crucially, the tool distinguishes between a published date and a modified date, because the two play different roles. The published date establishes when the content first appeared, which matters for original reporting and for topics where being first counts. The modified date tells engines the content has been maintained, which is what you want to emphasize on evergreen pages you keep current. The checker shows both so you can confirm that an article you genuinely revised is broadcasting a fresh modified date rather than only its original publication date, which is one of the most common freshness mistakes there is.
How to read the results
A clean result shows a published date, a modified date that reflects your most recent genuine update, agreement across every source that declares a date, and a freshness verdict that matches reality. With that, engines and readers both get a consistent, trustworthy story about how current the page is. If you recently updated the content, the modified date should be recent and visible, and the tool should report it as fresh. That is the state you are aiming for on any page where recency adds value.
Warnings point at specific gaps. No machine-readable date at all means the page relies on visible text that crawlers may not associate with your content; add structured dates. A modified date stuck in the past on a page you actually refreshed means your update is invisible to engines; the freshness work was wasted. Disagreeing sources mean a crawler cannot tell which date to trust; reconcile them. A stale verdict on a topic that moves means the content itself needs a real review, not just a date bump. Read each flag as a precise instruction rather than a grade, and fix the signal and the substance together.
The date-faking trap and other mistakes
The most damaging mistake is faking freshness. It is tempting to bump the modified date automatically on every page each night so everything looks recently updated, but engines have seen this trick for years and discount it. Worse, when the declared date says updated today and the content obviously has not changed in years, you erode trust rather than build it. Freshness signals only help when they are honest; a fresh date on stale content is a promise the page cannot keep. The right move is to actually review and improve the content, then let the date reflect that real change.
Other frequent errors are quieter. Teams update an article in their content system but the template never updates the schema dateModified, so the strongest signal stays frozen. People show a visible date in the body with no machine-readable backing and assume crawlers can read it. Sites set a future date by accident through a timezone or scheduling bug, which looks broken to engines. And many pages declare only a published date and never a modified one, so years of maintenance go unrecognized. The checker catches all of these because it reads exactly what a machine reads, not what the page hopes it conveys.
Freshness, evergreen content, and AI search in 2026
AI answer engines lean heavily on recency cues because nothing damages an AI answer like confidently citing outdated information. When a model assembles an answer from sources, a clear and honest modified date helps it prefer the current version of a fact over a stale one, and helps it decide whether your page is safe to cite for a time-sensitive query. Dates are part of how these systems judge whether a source is maintained and trustworthy. A page that keeps its content and its dates genuinely current is a better candidate for citation than one that looks abandoned.
Not every page needs to be fresh, and chasing recency on truly timeless content is a waste. A definition that has not changed in a decade does not need a new date every month, and constantly re-dating it can look manipulative. The skill is matching the freshness signal to the nature of the topic: aggressively maintain and re-date pages about things that move, and leave genuinely evergreen pages alone while ensuring their original dates are honest and present. The checker gives you the raw facts — what dates exist, whether they agree, how old they are — so you can apply that judgment instead of guessing.
What to do after you run the checker
If the page lacks machine-readable dates, add them — ideally datePublished and dateModified in Article schema, backed by a visible date in a time element so humans and machines see the same thing. If your update did not move the modified date, fix the template or workflow so genuine edits update the schema automatically. Reconcile any sources that disagree until every declared date tells one story. If the freshness verdict is stale and the topic moves, schedule a real content review: verify facts, refresh examples, update figures, and only then update the date.
Build freshness into your maintenance routine rather than treating it as a one-off. Identify the pages where recency matters most, set a cadence to review them, and make sure each genuine update flows into the modified date and the visible date together. Leave evergreen content alone except to confirm its dates are honest. After any fix, re-run the checker to confirm the dates now agree and the freshness verdict reflects the truth. The goal is simple: the freshest honest date you can claim, broadcast consistently everywhere a machine looks, backed by content that actually earns it.
It also helps to think about freshness at the level of your whole site, not just one page at a time. Run the checker across your most important articles and you will usually find clusters of neglect — a category of guides nobody has touched since launch, a series of comparison pages quoting prices that have moved, a set of how-to articles describing an interface that has since been redesigned. Those clusters are where stale dates and stale facts overlap, and they are the fastest place to recover lost trust and lost rankings. Prioritize the pages that combine high traffic, a date-sensitive topic, and a long gap since the last genuine edit, and you will get the most return from the least work. Freshness is not about churning every page constantly; it is about knowing which pages have quietly fallen out of date and fixing those deliberately before a reader, a competitor, or an AI engine notices the gap first.