--- title: Infinite Article Feed & Content Recirculation for Publishers description: The next article is served in-scroll before the reader leaves. Content recirculation measured at +46% multi-page sessions while comparison titles fell 16%. url: https://preview.artificialpoets.com/use-cases/infinite-article-feed/ site: Artificial Poets type: page date: 2026-08-09T19:46:31+00:00 modified: 2026-08-18T02:10:14+00:00 image: https://preview.artificialpoets.com/wp-content/uploads/a13s-cards/1499-social-5e093950.png --- # Serve the Next Article Before They Leave **Measured on live deployments** "What happens after a reader finishes an article — can the next one just be there?" It can. The Platform serves the next article into the page as the reader approaches the end of the current one, chosen for that reader in that session with no click required: the measured flagship, at +46% multi-page sessions while comparison titles fell 16%. [See it on my numbers](/request-a-demo/) Horse & Rider · Ride TV · Equine Network · Equus Magazine · The Score · Equine Network Lockup ## The problem, in your numbers The end of an article is where sessions die. The feed moves the decision from a link the reader must find and click to a page that is already there. - **3–4 → 1–2** Pageviews per arriving visit under AI search (Nitropay) - **6.2%** Sessions that reached a second article before deployment (our panel) - **−16%** Multi-page sessions over nine months on titles that change nothing (our panel) - **8% → 22%** Week-one return rate with one extra pageview (Chartbeat) ## The moment, simulated The measured continuation curve: once a reader accepts the first serve, each next one is more likely to be read, not less. Serving is positional rather than timed, and about 19 seconds of reading per served article is how we know it is reading, not scrolling. - **32%** continue past the 1st served article - **50%** past the 2nd - **70–82%** past the 4th and beyond ## The evidence, at its honest tier - **Multi-page sessions** — +46% on enabled titles (10.3% → 15.1% over nine months) vs −16% on comparison titles (12.0% → 10.1%) — same network, same weeks - **Largest single title** — +19.1% pages per session in the first quarter after enablement, +4.0% at steady state — both published; plan on the second - **The continuation curve** — 32% of readers continue past the first served article, then 50%, 59%, 70%, 78%, 82% at each step deeper. Engaged sessions add ~50 seconds of median reading time, roughly 30 seconds per served article - **A four-week pause in serving on the second title** — the lift went +12.2% → −2.7% → +6.9% as serving stopped and resumed. The metric follows the mechanism - **Traffic did not grow.** — Sessions and users were flat — this is a depth result, and we say so The deployment behind that number: [two titles grew multi-page sessions 46% while the comparison group fell 16%](/customers/equine-network-rollout/). **Introducing** ## Artificial Poets Platform One engine behind every solution on this site. It learns your archive and your readers, then acts inside your CMS, your templates and your ad stack. - It learns your archive: **Every story you have published, current again** — The engine understands each piece by what it is about, not when it ran or where it was filed. A feature from 2019 competes for the next slot on merit with one from this morning. - It reads the visit: **What a reader wants, without asking** — Interest builds from what someone actually does in the session. No login, no third-party cookies, nothing leaving your domain. Useful on the second pageview, not the tenth visit. - It chooses: **The right next read, not the popular one** — Someone comparing products and someone following a running story want different things. A most-read list gives both the same five links and serves neither. - It serves: **There before the reader leaves** — The feed, the recommendations, the search answer and the signup ask all run on the same engine, in your templates and your ad stack. Any slot that arrives with them is yours to sell. - It proves: **A lift you can defend, or we say so** — Every deployment runs beside titles that did not get it, plus a serving pause. That is how a result becomes a number you can take to a board instead of a vendor claim. ## FAQ ### A recommendation widget pays us a guaranteed check. What replaces that? Your own inventory. Chumbox widgets earn their check by selling exits — the click ends the session. The feed keeps the reader on pages you monetise. We publish no revenue figures, so the math stays yours: served pages × your session RPM, against the check. The depth side is measured — +46% vs −16% — and Chartbeat's return-rate figure (8% → 22% with one extra page) prices the retention side. ### Is this just an infinite-scroll plugin? No. Infinite scroll loads the next item in a list, the same order for everyone; it is a queue. This chooses the next article for this reader from your whole library, matched on meaning and in-session interest. The difference shows in behaviour: continuation past the fourth served article runs 70 to 82%. ### Where does it decide not to serve? Wherever you say so. Breaking news, service pages, live coverage and anything an editor has excluded stay untouched, and serving is positional, triggered by the reader's place on the page rather than a timer. The decision not to serve is part of the configuration, not an afterthought. ### What does another script do to Core Web Vitals? Serving happens after the initial page has rendered, below the first viewport, off the critical path for first paint. On the measured deployments, mobile page experience improved over the same window — not claimed as caused; claimed as not harmed. Vitals are one of the dashboards a pilot runs from day one, so your team can watch the same numbers. ### Is the selection a black box — can editors override it? Every serving decision is logged: which article, on which page, read for how long. Editors can exclude topics and sections, pin or suppress specific pieces, and mark pages where serving never happens — breaking news and legally sensitive coverage are the standard exclusions. The engine also declines to serve on its own when nothing in the archive fits the session. ### Does anything learned from our readers train something another publisher uses? No. Your behavioural data and your content are not pooled into a shared model. What the engine learns from your audience serves your pages only, and it stays yours when the contract ends. ### What does integration take from our engineering team? Authorization, CMS access and infrastructure access. The integration is a JavaScript tag per title, carried out by our team using documented standard implementations, with no migration and no redesign. Your team's real work is agreeing the metric before the four-week baseline starts. ## Related - [Horse & Rider Grew Pages Per Session 19% Without Growing Traffic](https://preview.artificialpoets.com/customers/horse-and-rider/) — In the first quarter after enabling the Platform, one in six Horse & Rider readers went past the first page, up from one in ten. Three comparable titles moved less than a quarter of a point. - [Your Scroll Depth Metric Lies When You Run Infinite Scroll](https://preview.artificialpoets.com/blog/scroll-depth-infinite-scroll/) — Scroll depth measures how far a reader travelled through a document, as a percentage of that document's height. The moment the document grows while the reader is inside it — which is exactly what infinite scroll does — the percentage falls while actual reading rises. If your scroll depth went down after shipping a continuous-scroll experience, you may be looking at your best readers, not your worst. - [Session Depth Assessment](https://preview.artificialpoets.com/resources/session-depth-assessment/) — Two numbers from your analytics place your single-page rate and depth distribution against a continuously measured publisher panel. - [Use Cases](https://preview.artificialpoets.com/use-cases/) — One engine across articles, streaming video and podcasts, for a single title or a whole network. Every use case here keeps the visit going after the first click, and turns that into more impressions and signups. ## See how this works on your titles Thirty minutes on your analytics. We tell you what share of your sessions stop at the first page, and what this would realistically move first for a setup like yours. You leave with the annotated read. No deck. [See it on my numbers](/request-a-demo/) How these numbers are made: [the measurement method](/solutions/measurement/). Figures from a multi-title publisher network measured continuously, August 2025 to May 2026. Last updated August 15, 2026. ## Questions this page answers ### What is an infinite article feed for publishers? It is a reading surface where the next article loads into the page as the reader approaches the end of the current one, with no click required. Each next article is chosen for that reader in that session rather than being a fixed list. On the network we measure, multi-page sessions rose 46% while comparison titles fell 16%. ### How does it decide which article loads next? The engine reads your archive by meaning rather than by tag or date, builds an anonymous interest profile from what the reader does in the session, and picks the next article on merit. Editors can pin, exclude and constrain what may be served, and the trigger is the reader's position on the page, not a timer. ### How much does an infinite article feed cost to run? There is no public price list; pricing follows network size and formats. The operational cost is deliberately small: one JavaScript tag per title, integration carried out by our team, and no migration or redesign. Four weeks of measurement-only baseline come before any serving, so the cost of finding out is a quarter and a tag. ### Who is the infinite article feed for? Publishers whose readers arrive on an article and leave from it: enthusiast networks, multi-title portfolios and news sites carrying mostly search and social traffic. If most of your sessions end on the page they started on, this is the surface that changes that, and it is where our measured results come from. ### What are the alternatives to an infinite article feed? Infinite scroll plugins load the next article by position in a list, which is a queue rather than a choice. Related-article widgets wait for a click most readers never make. Recommendation networks monetise exits to other sites. This serves your own library only, on your own domain, and is measured on voluntary continuation. ### How do I get started with an infinite article feed? Authorization, CMS access and infrastructure access are the requirements. Four weeks of measurement-only baseline follow, collection and indexing with no serving, then enablement on a named day. Eight weeks from authorization to a measurable effect, judged in your own analytics against metrics agreed before anything is switched on.