Evergreen vs News: How to Balance Content Mix for AI Search Freshness Signals

A 2025 analysis by Ahrefs found that pages receiving substantive content updates showed a 23% improvement in organic click-through rates within 60 days — but when the same team examined AI citation rates specifically, the freshness effect was three times larger for certain query categories and nearly zero for others. That asymmetry is the core problem most content teams are not solving.

Content strategy in 2026 is a freshness arbitrage problem. AI search systems — ChatGPT Browsing, Perplexity, Google AI Overviews, and Claude.ai — apply query-type-specific freshness weighting that renders a uniform content calendar strategy obsolete. The content teams winning AI citation share are running deliberate splits between time-anchored news content and durable evergreen foundation pieces, with each type serving a distinct function in the citation economy. The teams losing ground are publishing by instinct, volume, or whatever the editorial team enjoys writing.

This piece breaks down the mechanism, the allocation framework, the update cadence rules, and the measurement system for operating both content types at the same time without burning out your team or diluting your domain authority.

How AI Freshness Signals Actually Work

To understand why content mix matters, you need to understand how AI assistants evaluate recency — and it is more nuanced than most practitioners think.

The three-layer freshness model

AI search systems apply freshness evaluation at three distinct layers, each affecting citation probability differently.

Layer 1: Training data recency. Base models like GPT-4o and Claude 3.5 have training cutoffs. Content published after those cutoffs is not in the base model's knowledge and will never be cited from base model memory, regardless of quality. This is why live-retrieval systems — ChatGPT with browsing, Perplexity, Bing Copilot, Google AI Overviews — have become the primary citation targets for time-sensitive content. The base model cutoff is a hard ceiling on evergreen content's citation reach for queries with high recency sensitivity.

Layer 2: Retrieval recency weighting. When AI systems perform live web retrieval to augment their responses, they apply a recency weight to the retrieved content before selecting citation candidates. The weight is query-type conditional: queries containing temporal intent signals ("current," "latest," "2026," "now," "today," "recently") receive strong recency discounting on older content. Queries without temporal signals are evaluated primarily on authority and relevance, with recency as a weak secondary factor. Perplexity's engineering team noted in a 2025 blog post that their ranking system uses query temporal classification as a first-pass filter before applying relevance scoring.

Layer 3: Domain freshness signal. At the domain level, AI crawlers track the publication and update cadence of content across a site. A domain that consistently publishes substantive content — not trivial updates, but material additions of new information — receives a higher baseline freshness score that benefits all pages on the domain, including older evergreen content. This is the mechanism that makes the 60/40 content mix strategically important: the 40% news content is not just capturing news query citations, it is also lifting the freshness score of the 60% evergreen base.

Query classification: what the AI sees

AI retrieval systems classify queries into freshness-sensitivity tiers before applying weighting. Understanding these tiers helps you know which content type to produce for which topics.

The implication is that publishing a news piece about a definition-level concept wastes editorial resources and produces a piece with a short citation shelf life. Publishing an evergreen how-to guide in response to a breaking regulatory story produces a piece that is immediately discounted by AI systems as a poor freshness match. Format alignment to query type is the first principle of intelligent content mix strategy.

The 60/40 Allocation Framework

The benchmark allocation for teams optimizing both citation durability and freshness signal is 60% evergreen, 40% news-anchored. This ratio is not arbitrary — it reflects the structural proportions of the citation economy.

Why 60% evergreen

Evergreen content is the compounding asset in your AEO strategy. A well-built cornerstone piece — a comprehensive explainer, a durable how-to guide, a category comparison page — accumulates citation authority over 18 to 36 months as it gets linked to, quoted, and referenced. The research on AEO citation patterns consistently shows that the highest-citation-rate content on most domains is 12 to 24 months old, not newly published. That age advantage takes time to build and it cannot be rushed.

The 60% floor ensures that the domain is building durable citation assets rather than running the citation equivalent of a content treadmill — publishing constantly, getting short bursts of citation activity, and then watching each piece fall off AI retrieval shortlists as the news cycle moves on.

The types of content that belong in the evergreen 60%:

• Category and concept explainers. Definitional pieces that answer "what is X" and "how does X work" queries. These are the most consistently cited content type across ChatGPT, Perplexity, and Claude for informational queries.
• Playbooks and how-to guides. Procedural content answering "how do I do X." These get cited in AI answers to action-oriented queries and maintain relevance as long as the underlying procedure does not change.
• Comparison and alternatives pages. The comparison-query category is one of the highest-volume in B2B AI search. Comparison pages dominate AI recommendations and retain citation value for as long as the products being compared remain relevant — often 18 to 36 months.
• Statistical benchmark roundups. Annual data studies and benchmark reports with clear year-specific labeling. These are updated once per year, maintain high citation value in their update year, and provide temporal anchor data for other content.
• Glossary and definition pages. One of the most underestimated evergreen assets; definition content is cited across the widest range of query types and accumulates authority without significant maintenance overhead.

Why 40% news

The news 40% serves three functions that pure evergreen publishing cannot fulfill.

Function 1: Freshness signal for the whole domain. AI crawler freshness scoring happens at both the page and the domain level. A domain that publishes no news content appears to AI crawlers as a static resource — authoritative, perhaps, but not actively maintained. Regular news-anchored publication keeps the domain's freshness score elevated, which improves the retrieval probability of evergreen pieces on the same domain. This is the most counterintuitive benefit of news publishing for SEO-trained teams, who typically think of news as a separate traffic channel rather than a domain-wide signal.

Function 2: Burst citation capture. Industry events — regulatory changes, major product launches, research publication, market dislocations — create temporary high-demand query windows that pure evergreen content cannot serve. Publishing a well-structured news analysis within 48 to 72 hours of a major event captures citation share during the window when retrieval systems are actively surfacing new content on the topic. That citation activity also sends engagement and link signals back to the domain that compound over time.

Function 3: Data pipeline for evergreen updates. News pieces are the natural source of the dated statistics, regulatory updates, and product changes that evergreen content needs to stay temporally current. A news piece published in Q1 2026 about a new AI model pricing change becomes the cited source for the updated pricing data point in the evergreen "AI tool comparison" page. Without a news publishing operation, evergreen content teams have no internal pipeline for the temporal anchors their cornerstone pieces need.

When to shift the ratio

The 60/40 allocation is a steady-state baseline, not a fixed rule. Two conditions justify a temporary shift toward news:

Major industry events. Product category launches, significant regulatory changes, and major research publications warrant a surge to 70-80% news for two to four weeks. Teams that publish five to eight timely news analyses during a major industry event capture citation share that compounds over the following six months as those pieces become the canonical references for what happened.

New domain or new category entry. A new domain or a domain entering a new content category has no evergreen authority to draw on. The first 60 to 90 days of a new category effort should weight heavily toward news to establish the domain's freshness signal and crawl frequency before the evergreen foundation build begins.

Evergreen Content Update Cadence

Understanding what qualifies as a "substantive update" for AI freshness purposes versus a trivial edit is one of the most practical operational questions in AEO. AI models do not respond to adding a comma or fixing a typo — they respond to material changes in the information content of the page.

The three-tier update framework

Tier 1 — Fast-moving topics (update every 3-4 months): Topics where the underlying facts change significantly at least quarterly. This includes content about AI tools and models, software pricing and features, regulatory compliance requirements, and market share data. AI assistants trained or refreshed on recent data will have updated information about these topics, and content that lags behind creates the accuracy mismatch that damages citation trust. For Tier 1 content, plan for quarterly review cycles and budget author time accordingly.

Tier 2 — Moderately stable topics (update every 6 months): Topics where changes are meaningful but not constant — marketing strategy frameworks, management methodologies, technical best practices for established technologies. Twice-yearly reviews catch the meaningful changes without creating excessive maintenance overhead.

Tier 3 — Foundational content (update annually): Concept explainers, historical context pieces, established methodology descriptions, and glossary definitions that change only when the underlying discipline changes. Annual reviews are sufficient, with the primary task being a pass to confirm that the temporal anchors in the piece are still current and that no material inaccuracies have emerged.

What counts as a substantive update

The threshold for triggering an AI model's freshness re-evaluation is a meaningful change to the factual content of the page — specifically, changes that would make a new reader's understanding of the topic materially different from what the original version communicated.

Substantive updates include: adding a new section covering a development that postdates the original publication, updating a statistic to a newer year's data, revising a recommendation based on a product or regulatory change, and adding a new comparison row to a feature table. These changes are worth updating the `lastModified` date for and are likely to improve AI retrieval scores within 30 to 60 days of recrawl.

Non-substantive updates include: fixing typos, improving sentence clarity without changing meaning, reformatting for visual reasons, and adding internal links. These are maintenance activities that do not move the AI freshness signal.

The operational discipline is to review Tier 1 and Tier 2 content on schedule and perform substantive updates when the topic has changed, rather than performing cosmetic updates to game the timestamp.

News Content as an AEO Vehicle

News publishing in an AEO context is structurally different from journalism. The goal is not to break news — it is to be the authoritative synthesis of news that AI systems cite when answering questions about what happened and what it means.

The news-to-AI-citation pathway

Not all news articles are cited equally by AI assistants. The pieces that generate sustained citation activity share five structural properties:

1. Clear temporal anchoring. The headline and first paragraph contain an explicit date reference and a specific event or data point. Vague news pieces that describe general trends without specific anchors ("the AI landscape is shifting") are cited at much lower rates than pieces anchored to specific events with verifiable dates.

2. Named entities and specific figures. AI systems extract named entities — company names, person names, specific numbers — as the primary citation unit from news content. A news analysis that contains "Perplexity's monthly active users reached 85 million in April 2026, up 340% year-over-year" provides AI systems with a quotable, verifiable fact. A piece that says "Perplexity has grown substantially" provides nothing citable.

3. Structured synthesis, not just summary. AI assistants regularly cite news summaries when users ask "what happened with X." They cite analytical syntheses when users ask "what does X mean for operators." The latter citation is more durable and more valuable. A news analysis piece that moves from factual summary to structured implications for the reader is cited across a longer tail of query variants than a straight news summary.

4. Speed within the citation window. For breaking news queries, Perplexity's retrieval data suggests that content published within the first 48 to 72 hours of an event captures a disproportionate share of early citation volume, and that early citation volume correlates with longer-term citation authority on the topic. Publishing a well-structured news analysis within 72 hours of a major event is significantly more valuable than publishing a more polished piece five days later.

5. Clear connection to evergreen context. News pieces that explicitly link to and contextualize related evergreen content on the same domain create a citation graph that benefits both pieces. A news analysis of a new AI regulation update should link to the domain's existing evergreen piece on AI regulation strategy — the link signals topical authority to AI crawlers and creates a path from the fresh news signal to the durable cornerstone piece.

The "anchor event" content calendar structure

The most efficient news publishing operations in 2026 are structured around a calendar of predictable anchor events — scheduled product releases, regulatory deadlines, quarterly earnings, annual industry reports — supplemented by reactive capacity for unpredictable breaking news.

The anchor event structure allows teams to pre-brief evergreen updates that will be needed after each anchor event, assign news coverage in advance, and budget the reactive capacity (typically 20-25% of total content throughput) for unpredictable developments. This structure prevents the common failure mode of news publishing: the team drops all evergreen work to cover a major event, the evergreen pipeline stalls for four to six weeks, and the 60/40 ratio collapses to something closer to 20/80 for a quarter.

Temporal Anchoring: The Technique That Bridges Both Types

Temporal anchoring is the highest-leverage single technique for maintaining evergreen content's AEO performance over time. It is also the most underused.

The principle is simple: every evergreen piece should contain two to four explicitly dated data points that can be updated on an annual cycle without requiring rewriting of the surrounding content. When AI models retrieve and evaluate evergreen content, they use these temporal anchors as recency signals — a 2023 piece with a prominently placed 2026 data point is treated very differently from a 2023 piece with only 2023 citations.

How to implement temporal anchoring

Step 1: Identify anchor candidates. In any evergreen piece, there are typically three to five claims that are supported by statistics, product specifications, regulatory requirements, or market data. These are the anchor candidates — claims that are likely to change on a 12 to 24 month cycle and that, when updated, would meaningfully refresh the piece's temporal signal.

Step 2: Source to datable publications. Each anchor should be cited to a source with an explicit publication date — a named research report, an official company announcement, a regulatory filing. Avoid citing secondary aggregators without dates, as these provide no temporal signal. A citation to "Gartner's 2026 Marketing Technology Survey" is a strong temporal anchor. A citation to "recent research suggests" is noise.

Step 3: Update on schedule. Add anchor-update reviews to the Tier 1 and Tier 2 content calendar with a specific brief: find the updated version of this statistic from a credible source with a 2026 or later date, and update the anchor sentence with the new number. This typically takes 20 to 30 minutes per piece per update cycle and is the most cost-efficient improvement available for evergreen content AEO.

Step 4: Update the lastModified date. When anchor updates are made, update the page's lastModified metadata to signal the recency of the update to AI crawlers. The combination of updated factual content and updated metadata is the signal package that moves AI retrieval scores.

The temporal anchoring technique is particularly important for the 60% evergreen base because it allows a team of three to five content professionals to maintain a library of 200 to 400 evergreen pieces in good temporal health without a full rewrite schedule that would otherwise be unaffordable. This approach aligns directly with how AI crawlers assess entity currency — systems that read structured metadata alongside content freshness indicators together.

The "Last Reviewed" Signal

Related to temporal anchoring but distinct from it, the "last reviewed" meta-signal is a structured indicator that a human expert has evaluated the accuracy of a piece within a defined recent window, even if no factual changes were made.

Google has explicitly used "last reviewed" dates in its content quality evaluation for health and medical content. AI assistants have adopted a similar heuristic for YMYL-adjacent categories (finance, health, legal, technical guidance) and are extending it to B2B content as their quality evaluation matures.

The operational implementation is: for Tier 1 and Tier 2 content, add a "last reviewed" timestamp in a structured position on the page (typically in the byline area or at the bottom of the article) and update it when the scheduled review is performed, even when no substantive changes were made. The timestamp communicates active editorial stewardship to both human readers and AI crawlers.

For AEO purposes, content that visibly carries "last reviewed: April 2026" on a publication date of 2023 is treated as more current than equivalent content with only a 2023 publication date. This is particularly important for evergreen how-to and explainer content in fast-moving categories where the practices have not changed but AI models are uncertain whether the author has verified currency.

Date Manipulation Penalties and What to Avoid

As AI citation systems have become more sophisticated about freshness signals, a parallel problem has emerged: teams gaming freshness signals through artificial date manipulation, and AI systems developing countermeasures.

The behaviors that trigger freshness penalties in AI retrieval systems are well-documented by 2026:

Bulk timestamp updates without content changes. Changing the publication or modification date on hundreds of pages simultaneously without updating the underlying content is a pattern that Bing Webmaster Tools and Google Search Console both track and penalize in their respective AI search integrations. AI crawlers can compare the content hash of a page over time — updating the date without updating the content creates a discrepancy that marks the page as a freshness signal manipulator.

Reposting old content with new dates. Taking a 2023 piece, changing only the date to 2026, and republishing is the most common and most penalized freshness manipulation. AI retrieval systems cross-reference content similarity across index versions and apply a trust discount to pages that have low content-to-date-change ratios.

Swapping current year references without updating supporting data. Changing every mention of "2025" to "2026" in a piece without updating the underlying statistics, tools, or recommendations creates a specific mismatch pattern that AI systems are increasingly good at detecting. The year in the headline claims one thing; the cited sources (still pointing to 2024 reports) say another.

The safe operating principle: update dates only when substantive content changes have been made. The temporal anchoring and "last reviewed" approach described above provides all the freshness signal needed without triggering manipulation detection. Measurement of these citation patterns is covered in the AEO citation tracking playbook, which includes specific metrics for tracking freshness penalty signals.

Hybrid Content Formats

Between pure evergreen and pure news lies a category of hybrid formats that serve both freshness and durability functions simultaneously. These formats are underused and represent one of the highest-efficiency content investments available.

The "State of X" annual report

An annual report on a defined topic — "State of B2B AI Search 2026," "State of Content Marketing 2026" — is explicitly date-anchored by design, captures burst citation activity on its publication date, and provides year-specific data points that can serve as temporal anchors for evergreen content throughout the year. The same research investment serves both content types.

The structural requirements for high-citation "State of X" reports: primary data (survey, proprietary dataset, or original analysis), specific year-labeled findings, methodology description sufficient for AI models to assess credibility, and an executive summary formatted as a standalone quotable block. Reports that meet these criteria are cited at 4x to 8x the rate of reports that compile third-party statistics without original research.

The "Running update" evergreen piece

A running update format maintains a single URL but appends new developments below a clear "update" header, creating a growing document with explicit temporal layers. A piece titled "AI Search Citation Rate Benchmarks: Updated Quarterly" can begin as a 2,000-word evergreen foundation and grow to a 5,000-word document with four quarterly updates appended over a year. Each update adds a freshness signal to the URL while the original evergreen content retains its citation authority.

The running update format is particularly effective for benchmark data, regulatory compliance tracking, and tool comparison content — categories where the underlying framework is stable but the specific data changes frequently.

The news analysis with evergreen scaffolding

Standard news analysis: a piece published within 72 hours of an event, covering what happened and immediate implications, with a 12-to-18-month citation shelf life before the news context fades.

News analysis with evergreen scaffolding: the same piece, structured so that the first 40% of the content is evergreen context (the history, the mechanism, the framework for understanding events like this one) and the last 60% is the specific news analysis. The evergreen scaffolding portion is cited indefinitely. The news-specific portion drives the burst citation activity. The combined piece captures both citation types on a single URL with a single production investment.

Building the Calendar for Both Types

Operationalizing a 60/40 content mix requires a calendar architecture that prevents the most common failure modes: news events collapsing the evergreen pipeline, evergreen build-out crowding out reactive capacity, and anchor updates getting perpetually deprioritized.

The 12-week rolling calendar structure

The calendar structure used by the highest-performing content operations runs on a 12-week rolling window with the following allocation:

Evergreen pipeline (60% of capacity): Pre-planned pieces covering category explainers, comparison pages, how-to guides, and glossary build-out. These are briefed six to eight weeks in advance, produced on a two-to-three-week cycle, and published on a fixed weekly cadence. The pipeline structure allows the team to maintain output even when news events temporarily divert attention.

News capacity (25% of capacity): Reserved bandwidth for reactive news coverage. This is explicitly not assigned to scheduled work. Teams that fill this capacity with overflow evergreen work consistently fail to publish timely news coverage, because there is always overflow evergreen work. The discipline of holding 25% capacity as reserved reactive bandwidth is the single largest operational difference between teams that consistently publish timely news and teams that perpetually miss the citation window.

Anchor updates (15% of capacity): A rolling queue of Tier 1 and Tier 2 content pieces due for their scheduled review and update. This queue is maintained in a spreadsheet or content management system with review dates, and pieces are pulled into the sprint when their review date arrives. The 15% allocation is sufficient to maintain a library of 150 to 200 actively monitored evergreen pieces on a Tier 1 and Tier 2 schedule.

The monthly content mix audit

Once per month, the content lead reviews the prior month's publication output against the 60/40 target. The audit answers four questions:

1. What was the actual evergreen/news split by piece count and by word count? Word count is the more meaningful metric — a team publishing twelve short news items and two long evergreen pieces is running at a ratio closer to 80/20 by word count, even if it appears 83/17 by piece count.

2. How many Tier 1 and Tier 2 pieces received their scheduled update? A target of 90%+ of scheduled updates completed on time keeps the update pipeline credible.

3. What was the average time-to-publish for reactive news pieces? The target is under 72 hours from the triggering event. A team averaging five days on reactive coverage is missing the primary citation window for news content.

4. What is the citation rate trend for the evergreen foundation pieces published 6+ months ago? This is the lagging indicator that validates the whole strategy. If citation rates on mature evergreen content are growing, the compounding is working. If they are flat or declining, the quality or temporal anchor maintenance of the evergreen base needs attention.

The Measurement System

Content mix strategy without measurement is guesswork. The metrics system for tracking both evergreen and news AEO performance runs across three time horizons.

Weekly: Time-to-publish on reactive news pieces, citation capture rate on major industry events (did the team publish within the window, and did those pieces get cited within 30 days?).

Monthly: Evergreen-to-news split by word count, Tier 1 and Tier 2 update completion rate, new evergreen pieces added to the monitored library.

Quarterly: Citation rate trend for evergreen pieces at 6-month and 12-month age cohorts, domain freshness score trend (tracked via tools like Ahrefs or Semrush as a proxy), share of AI citation for category-defining query sets.

The quarterly cohort analysis is the most strategically important metric because it directly measures whether the evergreen compounding is happening. A healthy program shows evergreen pieces gaining citation share between their 6-month and 12-month cohorts. An unhealthy program shows flat or declining rates, which typically indicates either declining content quality, inadequate temporal anchor maintenance, or category competition that has not been matched by new comparison-page investment.

For a complete view of how to instrument this measurement stack, including the query-set design and tooling options, the share of model measurement framework covers the architecture in detail. The evergreen/news citation split maps directly onto the total citation share metrics that framework produces.

The Content Team Implications

The 60/40 framework has direct implications for how content teams should be staffed, structured, and incentivized — implications that most marketing organizations have not yet absorbed.

Evergreen and news require different writing skills. Evergreen cornerstone content rewards depth, structural clarity, and the ability to synthesize a topic comprehensively. News analysis rewards speed, judgment about what matters, and the ability to produce clean prose under time pressure. Few writers excel at both. The teams that produce the best output in each category tend to have writers who specialize: two or three evergreen specialists who build and maintain the foundation, and one or two fast-twitch writers who own the reactive capacity. Generalist teams that ask every writer to do both typically produce mediocre results in both categories.

Update work must be explicitly resourced. The most common failure in content mix programs is that evergreen update work — the 15% of capacity reserved for Tier 1 and Tier 2 reviews — gets squeezed out by new-piece production pressure. The only durable solution is to make update work visible in the content calendar, assign it as named tasks with deadlines, and measure completion rate monthly. Teams that treat update work as "catch up when you have time" never have time.

News publishing requires editorial authority. Reactive news coverage within a 72-hour window requires someone with the editorial authority to approve and publish without a multi-day review cycle. Companies with slow approval chains consistently miss the news citation window. The operational fix is pre-approved format templates, a designated approver for news pieces (distinct from the approver for long-form evergreen content), and a target that a first draft can reach final approval within 24 hours of the trigger event.

For teams building or restructuring in-house AEO operations, the Google AI Overviews mandate for publishers provides useful context on how the distribution economics of content are shifting in ways that directly affect team structure and resource allocation decisions.

Action Playbook: Implementing the 60/40 Framework in 90 Days

1. Audit your current content mix. Pull your last 90 days of published content and classify each piece as evergreen or news-anchored by primary intent. Calculate the split by piece count and word count. Most teams discover they are either much more news-heavy than they realized (because news pieces are faster to produce) or much more evergreen-heavy (because the team has never built a reactive capacity structure).

2. Inventory your evergreen library for temporal health. Tag every evergreen piece published more than 12 months ago with its tier (Tier 1, 2, or 3 based on topic volatility). For Tier 1 pieces, identify the two to four temporal anchors in each piece and flag any that are using data older than 12 months. This inventory becomes your update queue.

3. Establish the content calendar architecture. Set up a 12-week rolling calendar with explicit allocation buckets: evergreen pipeline slots (60%), reserved reactive capacity (25%), and update queue slots (15%). The key discipline is that the reserved reactive capacity cannot be reassigned to overflow evergreen work — hold it empty until a news trigger arrives, then fill it.

4. Implement temporal anchoring on the top 20 evergreen pieces. Identify the 20 evergreen pieces with the highest citation potential (based on traffic, inbound links, or category relevance) and spend two weeks updating their temporal anchors with current 2026 data points. This is the fastest path to moving the domain's overall freshness signal in the near term.

5. Set up the monthly mix audit. Build a simple spreadsheet template tracking the four monthly audit questions. Assign a standing calendar block at the start of each month to complete the audit. This is the feedback loop that keeps the strategy calibrated over time.

6. Train writers on the news-analysis format. The news analysis with evergreen scaffolding format — 40% evergreen context, 60% news analysis — produces the highest return on news content investment. Run one internal training session on the format and apply it to the next three reactive news opportunities.

7. Instrument citation tracking for both types. Set up separate citation tracking query sets for evergreen foundation content (category, definition, and how-to queries) and for news-anchored content (event-specific and time-anchored queries). Review both on a monthly cadence and use the divergence between them to calibrate your mix.

Takeaway: The content teams winning AI search citation share in 2026 are running a deliberate 60/40 split between evergreen foundation content and news-anchored pieces, with each type serving a different function in the citation economy. Evergreen content is the compounding asset — it builds citation authority over 18 to 36 months and answers the high-volume category and definition queries that make up the majority of AI search volume. News content is the freshness signal — it keeps the domain healthy in AI models' recency evaluation, captures burst citation opportunities during major industry events, and provides the dated data points that evergreen pieces need to stay temporally current. Neither type works without the other. The teams publishing 90% evergreen content are building durable assets on a domain that looks dormant to AI crawlers. The teams publishing 90% news content are getting short citation bursts with no compounding foundation. The 60/40 allocation, paired with a disciplined temporal anchoring practice and a 12-week rolling calendar, is the architecture that sustains both. Build it once, instrument it properly, and the compounding becomes observable within two quarters.