There is a lot of confident writing about "the Google Discover freshness algorithm" and almost none of it comes from Google. Google has never published a freshness weight for Discover; what it actually says is narrower and more useful than the vendor folklore — recommend content that is timely for current interests, and also note that older content can resurface when it becomes relevant to a reader. Both are true at once, which is exactly what the "freshness is everything / freshness is a myth" camps each get half-right. This guide separates the three layers people conflate: what Google documents, what the February 2026 Discover core update actually changed, and what publishers consistently observe in their own data — the early burst of Discover traffic a timely article gets and the decay that follows. Then it gets practical about the part that is real and controllable: content refresh. Why a genuine update can renew a decayed page and a cosmetic date bump does nothing, how to keep dateModified honest, how fast Google even recrawls to notice, and the cadence question underneath all of it. It ends on the constraint that quietly decides whether any of this works for you — freshness is a supply problem before it is an optimization problem, and most publishers lose the window not because they picked the wrong signal but because they could not produce the timely piece, or sustain the refresh pass, fast enough to matter.
Search "Google Discover freshness algorithm" and you will find a stack of confident posts describing a precise system — a 48-hour window, a 5-to-10x day-one impression boost, a freshness score that decays on a schedule. Almost none of it comes from Google. Google has never published a freshness weight, a window, or a named "freshness algorithm" for Discover. What it does say is narrower, and once you separate it from the folklore it is far more useful.
Here is the honest frame. Google's own Discover guidance recommends content that is "timely for current interests," and it also says, in the same document, that "older content may appear if it's helpful and relevant to individuals based on their interests." Both are true at once. That is why the two loudest camps — "freshness is everything" and "freshness is a myth" — are each half right and both misleading. Timeliness is a genuine input into what Discover surfaces; it is not a dedicated dial you optimize in isolation, and it does not override relevance. This guide separates the three layers people conflate (what Google documents, what the February 2026 Discover core update changed, and what publishers actually observe), then gets concrete about the one part that is real and in your control: the content-refresh workflow. For the wider set of 2026 Discover shifts this sits inside, see the Google Discover changes guide; for producing for the feed in general, Google Discover content strategy.
Start with the primary source, because it is short and most of the argument disappears once you read it. Google's Discover documentation makes a few things explicit. A page is eligible if it is indexed and meets Discover's content policies, and "no special tags or structured data are required" — there is no freshness markup to add. Eligibility is not appearance: "being eligible to appear in Discover is not a guarantee of appearing." Discover uses "many of the same signals and systems used by Search" to judge helpful, people-first content. And on freshness specifically, it recommends content that is "timely for current interests, tells a story well, or provides unique insights," while stating that older content can resurface when relevant.
Two more documented facts matter for expectations. Google says Discover traffic is "less predictable or dependable when compared to keyword-driven search visits," and that sites can see traffic changes "unrelated to the quality or publishing frequency of their content." Read together, that is Google telling you directly: timeliness helps you be relevant, relevance is the actual target, and the traffic is volatile by design. Nothing in the documentation describes a scored freshness window — so when a post quotes an exact decay curve, it is reporting an observation, not a Google rule.
Timeliness did get louder in 2026, and that is where a lot of the "freshness algorithm" talk comes from. In February 2026 Google rolled out a Discover-specific core update, announced on its Search Central blog, aimed at surfacing more in-depth, original, and timely content from sites with demonstrated expertise in a topic, showing more locally relevant content, and reducing sensationalism and clickbait. "Timely" is named as one of several goals there — alongside depth, originality, expertise, and locality — which is the correct way to weight it. The update raised the value of being genuinely current in an area you clearly know; it did not turn Discover into a pure recency feed. The full breakdown of that update and the year's other Discover shifts is in the Google Discover changes guide, so this page does not re-litigate it — the point here is only that it reinforced timeliness as an input, not as a standalone algorithm.
This is where the real-world numbers live, and where to keep your skepticism. Across publisher reports, the consistent pattern is that a timely article gets a concentrated burst of Discover impressions shortly after publishing — often within the first day or two — and then decays as the news interest cools, unless something renews it. The exact window is where sources disagree: estimates run from roughly 24 to 72 hours, and the dramatic "5-to-10x" day-one figures circulate without cited data. The pattern is real and widely seen; the precise multipliers are anecdotal. The practical takeaway survives the imprecision: if you are covering a genuinely timely topic, being early materially helps, because you are catching the interest curve while it is steep.
The decay half is the more actionable observation. A Discover-driven page does not usually fade because Google "expired" it on a timer; it fades because the reader interest that made it relevant cooled. That distinction is the whole basis of the refresh workflow — you cannot reset a timer that does not exist, but you can make a page relevant again by giving it new reasons to be, which is a content problem, not a date problem.
Refreshing older content is the single most repeated Discover tactic, and it is sound as long as you understand what actually does the work. A refresh helps when you change the content substantively: new information, a rewritten or expanded main body, a current angle on an evolving story, updated figures. It does nothing when you change only the visible date or the dateModified field on otherwise-stale text. Google's position on this has been consistent for years — fake or cosmetic date changes are "just noise," they do not help, and misleading dates can erode the trust signals that help you in the first place. The mechanism to keep in mind is that Google's systems look at the extent of the change, and edits to the main body text carry more weight than changes to navigation, comments, or boilerplate.
So the honest refresh pass is: identify a page whose topic has either evolved or whose interest is cycling back, genuinely update the substance, and let the dateModified move because the content actually moved — not the other way around. Keep the visible "last updated" date, the structured dateModified, and the real change in sync so the date is a truthful signal rather than a gamed one. Done this way, a refresh gives a decayed page a fresh reason to be relevant; done as a date-only bump, it is wasted effort at best and a credibility risk at worst.
A refresh only matters once Google recrawls and re-evaluates the page, and that is a separate, physical constraint most freshness advice skips. Recrawl frequency is not uniform: Google crawls pages it considers important or fast-changing more often, and a rarely-updated page on a small site can wait longer than you expect. The levers that speed discovery of a change are the plumbing ones — a current sitemap with accurate lastmod values, internal links from frequently-crawled pages, and for news-shaped content, feed-based signals. That whole layer, including which freshness signals actually influence recrawl speed and what the plumbing can and cannot do, is covered in AI crawler access to sitemaps and RSS feeds; the relevant point here is that "I refreshed it" and "Google has re-evaluated it" are two different events, and the gap between them is real.
Publishers love a prescribed cadence — "two to three posts a week" is a common one — but Google does not document a required frequency, and it explicitly warns that traffic swings can be unrelated to how often you publish. The useful version of the cadence idea is not a quota; it is coverage. A steady stream of genuinely helpful, timely-or-insightful content in areas where you have demonstrated expertise gives you more chances to catch a timely window and keeps a topic area active enough that your pages are the ones Google reaches for when interest spikes. Frequency is a means to that, not the goal. Publishing three thin pieces to hit a number is the opposite of what the February 2026 update rewards.
The last misconception worth killing is that Discover is a fresh-only surface. It is not. Google says older content can resurface when it becomes relevant to a reader, and it recommends content that "tells a story well or provides unique insights" right alongside content that is "timely for current interests." Breaking and seasonal topics lean fresh; a strong explainer or a distinctive take can resurface months after publishing when someone's interest cycles back to it. The durable strategy is a portfolio: timely pieces to catch the steep part of interest curves, and evergreen depth that Discover can re-surface on its own schedule. The refresh workflow is what links them — it is how an evergreen page re-earns relevance, and how a timely page gets a second life. For the video-specific version of this calculus, see Google Discover video optimization and the optimize videos for Google Discover walkthrough.
Here is the honest boundary before the bridge, because it is what makes the rest credible: Kompozy is not a Discover-optimization tool or an SEO analytics product. It does not decide whether Google puts a page in Discover, it does not read your Search Console Discover report, and it will not — by design — game a dateModified for you. Discover inclusion is Google's call on signals you influence but do not control. Reading and gaming the feed is not its job, and anyone selling a tool that "guarantees Discover" is selling the thing Google explicitly says no one can guarantee.
What the freshness game actually demands is upstream of all the optimization, and it is the part that quietly decides whether any of this works for you. To catch the early window you have to be able to produce a genuinely timely, genuinely good article fast — faster than the interest curve flattens. To run a real refresh program you have to sustain a cadence of substantive updates across a back catalog. Both are production problems. Most publishers lose the window not because they picked the wrong signal but because they could not get the timely piece written and live in time, or could not spare the hands to refresh last quarter's pages while also shipping this week's. The optimization advice is cheap; the throughput to act on it is the expensive, scarce thing.
That supply side is where Kompozy fits. It is a full AI content generation and multi-platform publishing engine, not a repurposing tool: point it at a source — a long video, a news angle, a URL, a topic — and it generates net-new content across its full range of output formats, including Blog Articles it can publish straight to WordPress, a GHL blog, or a custom webhook, every output governed by one written Persona Brief so a timely piece still reads in your voice under deadline. Then it fans the accompanying social distribution — eight social platforms plus blog and email — on autopilot behind a per-post review gate, which is how a Discover-eligible page gets the surrounding attention and signal rather than sitting alone. Concretely: it collapses the time between "this is timely right now" and "the article plus its promotion are live" from a day to minutes, and it makes a weekly refresh pass across your back catalog something a small team can actually staff. You still own the editorial judgment, the fact-checking, and the decision of what is worth being timely about — Kompozy removes the throughput ceiling that otherwise turns every freshness strategy into a plan you cannot execute.
There is no documented "Google Discover freshness algorithm." Google recommends timely content and also surfaces relevant older content; the February 2026 core update raised the value of being timely, original, and expert without making Discover recency-only; and the observed early-burst-then-decay pattern is real while the exact windows and multipliers are not Google figures. The controllable part is content refresh done honestly — substantive updates with a truthful dateModified, not cosmetic date bumps — plus the plumbing that gets a change recrawled and a portfolio that mixes timely and evergreen. And underneath all of it sits the real constraint: freshness is a supply problem first. Solve the production throughput to publish timely and refresh consistently, and the optimization advice finally has something to work on. Fail to, and the best freshness strategy in the world is a document you never get to act on.
Not as a documented, named system. Google has never published a freshness weight or a "freshness algorithm" specific to Discover. Its own guidance recommends content that is "timely for current interests" but also says older content can resurface when it becomes relevant to a reader. So timeliness is a real factor in what Discover surfaces, but it is one input among many — the same helpful, people-first signals used across Search — not a dedicated scoring dial you can target directly.
There is no Google-published figure. Publishers commonly observe that a timely article gets a concentrated burst of Discover impressions in roughly the first day or two after publishing, then decays, with vendor estimates of the window ranging from about 24 to 72 hours. Treat any precise number as an observation from someone else's data, not a rule — Discover traffic is explicitly less predictable than keyword search, and the curve depends heavily on topic interest and your own audience.
Only if the content genuinely changed. Google's guidance on dates has been consistent for years: changing a visible date or dateModified without substantively updating the content is "just noise" and does not help. A real refresh — rewritten or added main-body content, new information, a current angle — can renew a page's relevance; a cosmetic date bump on stale text does not, and misleading dates can erode trust. Keep dateModified honest and matched to the actual change.
There is no required cadence; Google explicitly says traffic changes can be unrelated to how often you publish. What matters is a steady supply of genuinely helpful, timely-or-insightful content in areas where you have demonstrated expertise, rather than sporadic batches or thin volume for its own sake. Consistency helps because it gives you more chances to catch a timely window and keeps a topic area active, but frequency without quality is not a Discover lever.
It favors relevance, and relevance is sometimes timely and sometimes evergreen. Google says Discover may surface older content when it becomes relevant to an individual's interests, and it recommends content that tells a story well or provides unique insights alongside content that is timely for current interests. Breaking or seasonal topics lean fresh; a strong explainer can resurface months later when a reader's interest spikes. The win is covering both, not betting everything on recency.
Google Discover has no publicly documented "freshness algorithm." Google's own guidance recommends content that is "timely for current interests," but it also says older content can resurface when relevant and that no special tags or dates are required. In practice, publishers observe that a timely article often gets an early burst of Discover traffic that decays over a day or two, and a genuine content update — substantive edits plus an honest dateModified — can renew it. Cosmetic date changes do not work.
Get started → · ← All guides · Compare Kompozy vs other tools