<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:media="http://search.yahoo.com/mrss/">
  <channel>
    <title>Content Strategy · Jared Lynskey</title>
    <link>https://jared.lynskey.co.nz/en/tags/content-strategy/</link>
    <description>Short stories of value on software engineering and engineering leadership — performance reviews, hiring and scaling teams, CI/CD, DevOps monitoring and agentic AI. Written from Seoul.</description>
    <generator>Hugo</generator>
    <language>en</language>
    <copyright>© 2026 Jared Lynskey</copyright>
    <lastBuildDate>Sat, 07 Nov 2026 00:00:00 +1300</lastBuildDate>
    <atom:link href="https://jared.lynskey.co.nz/en/tags/content-strategy/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>How to Get Cited by ChatGPT: What My Blog&#39;s Logs Actually Show</title>
      <link>https://jared.lynskey.co.nz/en/posts/2026/2026-09-16-blog-as-a-source-for-ai-assistants/</link>
      <guid isPermaLink="true">https://jared.lynskey.co.nz/en/posts/2026/2026-09-16-blog-as-a-source-for-ai-assistants/</guid>
      <pubDate>Wed, 16 Sep 2026 00:00:00 +1200</pubDate>
      <dc:creator>Jared Lynskey</dc:creator>
      <category>SEO</category>
      <category>Crawlers</category>
      <category>LLM</category>
      <category>Hugo</category>
      <category>Cloudflare</category>
      <category>Content Strategy</category>
      <category>Writing</category>
      <description>How to get cited by ChatGPT and other AI assistants (generative engine optimization, GEO): which bots to let in, what to serve them, and why llms.txt didn&#39;t help.</description>
      <content:encoded><![CDATA[<p>To get cited by ChatGPT and other AI assistants, first let their search bots fetch your pages. They are separate from the training crawlers, and blocking one does nothing to the other. Then serve HTML that reads without JavaScript, with structured data naming the author. llms.txt, as far as my logs show, does nothing, because nothing requests it; what matters most is writing so each section states its claim first and a paragraph can be quoted on its own.</p>
<p>Search traffic to this blog has been flat for a couple of years and I&rsquo;ve stopped expecting it to move. So I went looking at where visitors come from instead, and whether the AI assistants are among them. They are, barely. In the twelve months to 16 September 2026, my analytics counted 5,372 sessions here. Twenty-seven of them came from an AI assistant: twelve from chatgpt.com, nine from gemini.google.com, six from Perplexity, and none from claude.ai. Half a percent. The ChatGPT visitors were the good kind, though. They stayed five minutes on average, against 27 seconds for direct traffic and three minutes for Google, because they&rsquo;d already read the answer and wanted the detail.</p>
<p><figure><span class="mc-img"><img
    class="my-0 rounded-md"
    loading="lazy"
    decoding="async"
    fetchpriority="auto"
    alt="Sessions by referring source over twelve months; AI assistants highlighted"
    width="2400"
    height="1164"
    src="https://jared.lynskey.co.nz/en/posts/2026/2026-09-16-blog-as-a-source-for-ai-assistants/chart-referrers_hu_590f408ed875db1f.webp"
    srcset="https://jared.lynskey.co.nz/en/posts/2026/2026-09-16-blog-as-a-source-for-ai-assistants/chart-referrers_hu_590f408ed875db1f.webp 800w, https://jared.lynskey.co.nz/en/posts/2026/2026-09-16-blog-as-a-source-for-ai-assistants/chart-referrers_hu_d8da7a3f1d1c9190.webp 1280w"
    sizes="(min-width: 768px) 50vw, 65vw"
    data-zoom-src="https://jared.lynskey.co.nz/en/posts/2026/2026-09-16-blog-as-a-source-for-ai-assistants/chart-referrers_hu_b3dda7d23abf271d.webp"></span><figcaption>Twelve months of sessions by source. The AI assistants are the blue bars: 27 sessions out of 5,372.</figcaption></figure>
</p>
<p>So I&rsquo;ve spent some time on how a small blog gets picked as the source for those answers. This post is what I&rsquo;ve done on this site and what the logs say about it. Where I have numbers I give them, and where I&rsquo;m guessing I say so. Three sentences in the first draft turned out to be wrong once I looked, and I&rsquo;ve left the corrections in.</p>

<h2 class="relative group">Two kinds of bot, and only one of them cites you
    <div id="two-kinds-of-bot-and-only-one-of-them-cites-you" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#two-kinds-of-bot-and-only-one-of-them-cites-you" aria-label="Anchor">#</a>
    </span>
    
</h2>
<p>The first thing to get straight is that &ldquo;AI bots&rdquo; is two different jobs with different user agents, and blocking one does nothing to the other.</p>
<p>Training crawlers pull pages in bulk to build the next model. OpenAI&rsquo;s is GPTBot, Anthropic&rsquo;s is ClaudeBot, Google&rsquo;s is controlled by the Google-Extended token, Apple&rsquo;s by Applebot-Extended, and Common Crawl&rsquo;s CCBot feeds most of the open datasets. Whatever they take ends up as weights, with no link back to you. There&rsquo;s a case for blocking these and I don&rsquo;t think it&rsquo;s a silly one.</p>
<p>Retrieval bots are the other kind. When someone asks ChatGPT or Claude or Perplexity a question, the assistant runs a search, fetches a handful of pages right then, reads them, and writes an answer with citations. Those fetches arrive as OAI-SearchBot and ChatGPT-User, Claude-SearchBot and Claude-User, PerplexityBot and Perplexity-User. Google&rsquo;s AI Overviews use plain Googlebot, so there&rsquo;s no separate switch for them at all. These are the bots that send a visitor back. If the aim is to be cited, these are the ones you&rsquo;re optimising for, and it&rsquo;s worth knowing that blocking GPTBot does nothing to ChatGPT&rsquo;s search results in either direction.</p>
<p>This site&rsquo;s robots.txt is Hugo&rsquo;s default, which is allow everything:</p>
<div class="highlight-wrapper"><div class="highlight"><pre tabindex="0" class="chroma"><code class="language-text" data-lang="text"><span class="line"><span class="cl">User-agent: *
</span></span><span class="line"><span class="cl">Allow: /
</span></span><span class="line"><span class="cl">Sitemap: https://jared.lynskey.co.nz/sitemap.xml</span></span></code></pre></div></div>
<p>I&rsquo;ve thought about splitting it, letting the retrieval bots in and keeping the training ones out, and decided it wasn&rsquo;t worth doing for a blog whose whole purpose is to be read. If I did want that, it would look like the block below. The reason to show it is the same lesson as in <a href="https://jared.lynskey.co.nz/en/posts/2026/2026-09-09-cloudflare-crawlers-cache/" >Cloudflare Served Googlebot an Empty Page</a>: the list has to name bots, not vendors. There is no line you can write that says &ldquo;OpenAI&rdquo;.</p>
<div class="highlight-wrapper"><div class="highlight"><pre tabindex="0" class="chroma"><code class="language-text" data-lang="text"><span class="line"><span class="cl">User-agent: GPTBot
</span></span><span class="line"><span class="cl">User-agent: ClaudeBot
</span></span><span class="line"><span class="cl">User-agent: Google-Extended
</span></span><span class="line"><span class="cl">User-agent: Applebot-Extended
</span></span><span class="line"><span class="cl">User-agent: CCBot
</span></span><span class="line"><span class="cl">Disallow: /
</span></span><span class="line"><span class="cl">
</span></span><span class="line"><span class="cl">User-agent: *
</span></span><span class="line"><span class="cl">Allow: /</span></span></code></pre></div></div>
<p>One thing to check if the site sits behind Cloudflare. Since the middle of 2025, new zones there block AI crawlers by default, and there&rsquo;s a one-click toggle under the bot settings that does the same to existing ones. It blocks both kinds. This blog moved to Cloudflare Pages last week, so that was on my list. If you turned it on because &ldquo;AI bots&rdquo; sounded like something to keep out, you&rsquo;ve also turned off the citations. Cloudflare has since started writing a <code>Content-Signal</code> line into the robots.txt files it manages, with separate <code>search</code>, <code>ai-input</code> and <code>ai-train</code> flags. Nobody is obliged to honour it, but it&rsquo;s the right vocabulary: the same split as above, expressed as a preference rather than a wall.</p>

<h2 class="relative group">What actually knocks on the door
    <div id="what-actually-knocks-on-the-door" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#what-actually-knocks-on-the-door" aria-label="Anchor">#</a>
    </span>
    
</h2>
<p>The move to Cloudflare had a side effect: for the first time I can see every request by user agent. Here is what came in over the four days from 13 to 16 September 2026, after throwing out the fakes, which I&rsquo;ll get to.</p>
<table>
	<thead>
			<tr>
					<th>Bot</th>
					<th>Requests</th>
					<th>What it did</th>
			</tr>
	</thead>
	<tbody>
			<tr>
					<td>ClaudeBot (training)</td>
					<td>1,777</td>
					<td>1,380 distinct pages in two days, all five languages, plus 37 reads of robots.txt</td>
			</tr>
			<tr>
					<td>meta-externalagent (training)</td>
					<td>1,748</td>
					<td>The same, a day behind</td>
			</tr>
			<tr>
					<td>OAI-SearchBot (ChatGPT&rsquo;s search index)</td>
					<td>59</td>
					<td>Pages in all five languages, and robots.txt sixteen times</td>
			</tr>
			<tr>
					<td>ChatGPT-User (fetched while answering)</td>
					<td>20</td>
					<td>The Japanese post on cold starts, the self-hosted update server post, the homepage</td>
			</tr>
			<tr>
					<td>Claude-User (fetched while answering)</td>
					<td>8</td>
					<td>The update server post, the contact page, the homepage</td>
			</tr>
			<tr>
					<td>PerplexityBot</td>
					<td>2</td>
					<td>The iOS debugging post, in English and Vietnamese</td>
			</tr>
			<tr>
					<td>Claude-SearchBot, Perplexity-User</td>
					<td>0</td>
					<td></td>
			</tr>
			<tr>
					<td>Googlebot / Bingbot / Applebot / CCBot</td>
					<td>168 / 71 / 45 / 28</td>
					<td></td>
			</tr>
	</tbody>
</table>
<p>Two things stand out. The training crawlers read everything, twice, in the time it took the retrieval bots to read a few dozen pages. And the retrieval bots are not fetching at random: ChatGPT-User and Claude-User both went for the update server post, which is the most specific thing on the site, and the Japanese cold-start post got fetched in Japanese.</p>
<p>Now the fakes. Of 20 requests claiming to be GPTBot, 18 came from addresses outside OpenAI&rsquo;s published ranges and asked for <code>/.env</code>, <code>/.git/HEAD</code> and <code>/firebase-adminsdk.json</code>. Fifteen of 36 &ldquo;ChatGPT-User&rdquo; requests were the same kind of thing. Most of the &ldquo;PerplexityBot&rdquo; traffic wanted <code>/id_ed25519</code>. A user agent is a string anyone can send, and vulnerability scanners have noticed that AI bot names get waved through. OpenAI, Anthropic and Perplexity all publish their IP ranges. If you&rsquo;re going to treat a bot specially, in either direction, check the address, not the name. The 20, 59 and 8 above are the requests that came from the right addresses.</p>
<p><figure><span class="mc-img"><img
    class="my-0 rounded-md"
    loading="lazy"
    decoding="async"
    fetchpriority="auto"
    alt="Requests claiming an OpenAI user agent, split by whether they came from OpenAI&rsquo;s published IP ranges"
    width="2400"
    height="722"
    src="https://jared.lynskey.co.nz/en/posts/2026/2026-09-16-blog-as-a-source-for-ai-assistants/chart-spoofed_hu_e7035266dc7ce0b1.webp"
    srcset="https://jared.lynskey.co.nz/en/posts/2026/2026-09-16-blog-as-a-source-for-ai-assistants/chart-spoofed_hu_e7035266dc7ce0b1.webp 800w, https://jared.lynskey.co.nz/en/posts/2026/2026-09-16-blog-as-a-source-for-ai-assistants/chart-spoofed_hu_638e8c6286f5c91f.webp 1280w"
    sizes="(min-width: 768px) 50vw, 65vw"
    data-zoom-src="https://jared.lynskey.co.nz/en/posts/2026/2026-09-16-blog-as-a-source-for-ai-assistants/chart-spoofed_hu_11d44262308d7c6c.webp"></span><figcaption>The same user agent strings, split by source address. Most of the &lsquo;GPTBot&rsquo; traffic never came from OpenAI.</figcaption></figure>
</p>

<h2 class="relative group">The page has to be readable without a browser
    <div id="the-page-has-to-be-readable-without-a-browser" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#the-page-has-to-be-readable-without-a-browser" aria-label="Anchor">#</a>
    </span>
    
</h2>
<p>Retrieval bots fetch at answer time, with a timeout, and most of them don&rsquo;t run JavaScript. A page that is an empty div until a bundle loads is a page they don&rsquo;t read. Hugo writes finished HTML, so this blog gets that for free, and it&rsquo;s the main reason I&rsquo;ve never been tempted to move it to anything cleverer. The marketplace I run is the opposite case, a React Native web app, and getting crawlers a real document there took a dynamic-rendering branch and then a caching bug that quietly undid it. The Cloudflare post above is that story.</p>
<p>Speed counts the same way, and here the logs corrected me. I&rsquo;d assumed a static site behind Cloudflare was served from the nearest edge. It isn&rsquo;t. This blog is on Cloudflare Pages, the HTML goes out uncached, and from Seoul the response comes back from Los Angeles in about half a second. That&rsquo;s still fine for a bot with a timeout of several seconds, but it isn&rsquo;t the tens of milliseconds I&rsquo;d have claimed, and I only know because I measured it. An assistant assembling an answer from six pages will drop the one that takes three seconds, and it doesn&rsquo;t tell you it did.</p>

<h2 class="relative group">Write so a paragraph can be lifted out
    <div id="write-so-a-paragraph-can-be-lifted-out" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#write-so-a-paragraph-can-be-lifted-out" aria-label="Anchor">#</a>
    </span>
    
</h2>
<p>The way these systems use a page is not the way a reader does. They split it into chunks, score each chunk against the question, and quote the ones that score. A post that builds up to its point over eight hundred words scores badly on every chunk. A post where each section states its claim in the first sentence and then supports it scores on the section that matches.</p>
<p>That changed how I write here more than any technical thing did. The description in the front matter is now a summary of what the post says, not a hook to get you to read it, because the description is often the only part an assistant shows. Section headings say what the section concludes. Numbers go in the sentence with their units rather than in a chart. Dates are written out, because &ldquo;last week&rdquo; means nothing to a model reading the page in a year. And a post is about one thing.</p>
<p>The other half is having something worth quoting, and here the data is mixed. In the logs, the pages the answer-time bots fetch are the specific ones: cold starts, the update server, iOS debugging. Naver&rsquo;s crawler is called Yeti and Daum&rsquo;s is Daumoa, a restart on my host is a thirty-second window for 504s, Cloudflare&rsquo;s on-the-fly Brotli measured worse than gzip on my bundle. Those facts aren&rsquo;t written down in many places, so a question that needs one of them lands here. But the people who click through from an assistant mostly land on the 2023 and 2024 posts about management, one or two sessions each. The specific posts get read by the machine. The general ones get the click. I don&rsquo;t have a tidy explanation for that and I&rsquo;d rather report it than invent one. The lever, if there is one, is still to write down the specific thing you found out this week that you couldn&rsquo;t find when you searched for it, because that&rsquo;s what the machines are demonstrably reading.</p>

<h2 class="relative group">Say who wrote it, and keep saying the same thing
    <div id="say-who-wrote-it-and-keep-saying-the-same-thing" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#say-who-wrote-it-and-keep-saying-the-same-thing" aria-label="Anchor">#</a>
    </span>
    
</h2>
<p>An assistant deciding whether to cite a page is doing something like what a careful reader does: who wrote this, when, and does it agree with everything else I&rsquo;ve found? The mechanics of that on this site are mostly the theme&rsquo;s doing. Every post carries Article structured data with an author, a published date and a modified date, and the author is a Person with links to the same public profiles. That&rsquo;s a few lines in a template, and it&rsquo;s the difference between &ldquo;a page&rdquo; and &ldquo;a page by a named person with a history&rdquo;.</p>
<p>The part that isn&rsquo;t the theme&rsquo;s doing is consistency. The about page, the llms.txt, my public profiles and the line at the top of every post all describe the same person with the same job history. When something changes, all of them change, or the one I forgot becomes the one that&rsquo;s wrong. That sounds fussy, but one page saying I&rsquo;m an engineering manager and another saying I&rsquo;m freelance is exactly the kind of contradiction that makes a model hedge, or pick someone else.</p>
<p>The modified date has to be honest too. It changes when the content changes and I don&rsquo;t touch it otherwise. Bumping it to look fresh is the kind of thing that works once and then gets your whole domain discounted.</p>

<h2 class="relative group">What llms.txt is, and the fact that nothing reads it
    <div id="what-llmstxt-is-and-the-fact-that-nothing-reads-it" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#what-llmstxt-is-and-the-fact-that-nothing-reads-it" aria-label="Anchor">#</a>
    </span>
    
</h2>
<p>llms.txt is a proposal from September 2024 for a Markdown file at the site root that gives a language model a summary of the site and a curated list of links. I added one in February this year. It&rsquo;s a paragraph about who I am, then sections for management, infrastructure, mobile and community, with one line per post: the title, the URL, and a sentence saying what the post concludes. It reads like a table of contents written for someone in a hurry.</p>
<p>The caveat is that nothing reads it. In the four days of logs I have, the only requests for <code>/llms.txt</code> were my own. Common Crawl has captured 76 pages of this site in its August 2026 crawl and never that file. No major assistant has said it reads it. So it is not why anything here gets cited, and I&rsquo;d stop short of recommending it to anyone on those grounds. I keep it for two reasons. It cost an hour, and writing one sentence per post that says what the post actually concludes was worth more than the file, because it showed me which posts didn&rsquo;t conclude anything. It&rsquo;s also the one page on the site that tells a machine what the site is for, and I&rsquo;d rather have that and not need it. It has drifted since, because I write it by hand and forget, which is the argument for generating it from the front matter as a Hugo output format instead. That&rsquo;s on the list.</p>

<h2 class="relative group">Feeds, sitemaps and five languages
    <div id="feeds-sitemaps-and-five-languages" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#feeds-sitemaps-and-five-languages" aria-label="Anchor">#</a>
    </span>
    
</h2>
<p>The boring plumbing still counts. The sitemap is how a crawler finds a new post the day it goes up. RSS is how it learns a post changed. Both are Hugo defaults and I&rsquo;ve never had to think about them. ClaudeBot read the sitemap 36 times in two days, which is how it found 1,380 pages so quickly.</p>
<p>The languages are the part I didn&rsquo;t expect to matter. Everything here is in English, Japanese, Korean, Simplified Chinese and Vietnamese, with alternate links so a crawler knows they&rsquo;re the same post. The crawlers treat them as five sites: ClaudeBot&rsquo;s crawl split almost evenly across the five, and of the 73 post pages Common Crawl took in August, only four were the English ones. Assistants answer in the language they&rsquo;re asked in and lean towards sources in that language, so a Korean question about cold starts can be answered from the Korean page rather than from a translation of the English one done on the fly, and the Japanese page is the one ChatGPT actually fetched. The condition is that the translation reads as if it were written in that language. The pass I did on that earlier this year, with native speakers checking, was for human readers, but it&rsquo;s also what makes the Vietnamese page a source rather than a curiosity.</p>

<h2 class="relative group">What I can measure
    <div id="what-i-can-measure" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#what-i-can-measure" aria-label="Anchor">#</a>
    </span>
    
</h2>
<p>More than I thought, once I stopped guessing. Cloudflare&rsquo;s free plan gives request logs by user agent and path, one day per query, which is where the bot table came from; referrers need the paid plan. The analytics tool has the referrers, and pulling them through its reporting API rather than squinting at a dashboard is what produced the numbers at the top. Together they show the shape of it. In four days the retrieval bots fetched about ninety pages from OpenAI and Anthropic addresses. In the last month, two people arrived from ChatGPT. Read a lot, cited sometimes, clicked rarely. There&rsquo;s still no way I know of to see the answers that used a page and didn&rsquo;t get clicked. The most useful check is still the manual one: ask the assistant the question your post answers and see whose page it quotes.</p>
<p><figure><span class="mc-img"><img
    class="my-0 rounded-md"
    loading="lazy"
    decoding="async"
    fetchpriority="auto"
    alt="Sessions referred by AI assistants per month over thirteen months, showing no trend"
    width="2400"
    height="968"
    src="https://jared.lynskey.co.nz/en/posts/2026/2026-09-16-blog-as-a-source-for-ai-assistants/chart-monthly_hu_cd5dc99be8b0cfc3.webp"
    srcset="https://jared.lynskey.co.nz/en/posts/2026/2026-09-16-blog-as-a-source-for-ai-assistants/chart-monthly_hu_cd5dc99be8b0cfc3.webp 800w, https://jared.lynskey.co.nz/en/posts/2026/2026-09-16-blog-as-a-source-for-ai-assistants/chart-monthly_hu_7d401a84f177509c.webp 1280w"
    sizes="(min-width: 768px) 50vw, 65vw"
    data-zoom-src="https://jared.lynskey.co.nz/en/posts/2026/2026-09-16-blog-as-a-source-for-ai-assistants/chart-monthly_hu_5c308d89ad110ef7.webp"></span><figcaption>Referrals from AI assistants by month. There is no trend to speak of.</figcaption></figure>
</p>
<p>If I had to keep one thing from all of this, it&rsquo;s the writing rule. The robots.txt, the structured data and the llms.txt are an afternoon&rsquo;s work and then they&rsquo;re done. Writing so each section says its claim first, and writing about the specific thing rather than the general one, is a habit, and it&rsquo;s the habit that decides whether a page is worth quoting at all. The second thing is to measure before you claim. Being trusted by a machine turns out to look a lot like being trusted by a person: say who you are, say what you know, and say it the same way every time.</p>
]]></content:encoded>
      <media:content url="https://jared.lynskey.co.nz/en/posts/2026/2026-09-16-blog-as-a-source-for-ai-assistants/featured.jpg" medium="image" type="image/jpeg" />
    </item>
    <item>
      <title>Content Ranking Algorithms: Signals Beyond Views and Likes</title>
      <link>https://jared.lynskey.co.nz/en/posts/2023/2023-10-06-reinenforcement-learning/</link>
      <guid isPermaLink="true">https://jared.lynskey.co.nz/en/posts/2023/2023-10-06-reinenforcement-learning/</guid>
      <pubDate>Fri, 06 Oct 2023 00:30:03 +0000</pubDate>
      <dc:creator>Jared Lynskey</dc:creator>
      <category>Content Ranking</category>
      <category>Metrics</category>
      <category>SEO</category>
      <category>Machine Learning</category>
      <category>User Engagement</category>
      <category>Content Strategy</category>
      <description>What a content ranking algorithm does, why ranking on views and likes alone makes the popular more popular, and the quieter signals worth blending into one score.</description>
      <content:encoded><![CDATA[<p>A content ranking algorithm decides the order a feed, a search page or a &ldquo;top posts&rdquo; list shows things in. It gives every item a score built from several signals (how people engage with it, what the content itself is like, how fresh and relevant it is) and sorts by that score. Rank on views and likes alone, though, and you get a rich-get-richer loop: whatever sits at the top gets seen, so it collects more views and likes, so it stays at the top.</p>
<p>That&rsquo;s the obvious flaw in views and upvotes: they reward whatever&rsquo;s already popular. Fresh content starves at the bottom of the feed, and the platform slowly turns into a rerun channel. Ranking a feed well is one of the quietly hard problems where I work, so I&rsquo;ve spent a fair bit of time thinking about which other signals are worth trusting.</p>

<h3 class="relative group">Watch what people do, not what they tap
    <div id="watch-what-people-do-not-what-they-tap" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#watch-what-people-do-not-what-they-tap" aria-label="Anchor">#</a>
    </span>
    
</h3>
<p>Explicit feedback is scarce and biased; behaviour is plentiful. Time on page and scroll depth tell you whether someone actually read the thing or bailed after the first paragraph. Revisits are even better — coming back to a piece is a stronger endorsement than any upvote. And a fast exit is a useful negative signal: the content promised something the reader didn&rsquo;t get.</p>

<h3 class="relative group">Signals baked into the content itself
    <div id="signals-baked-into-the-content-itself" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#signals-baked-into-the-content-itself" aria-label="Anchor">#</a>
    </span>
    
</h3>
<p>Good content doesn&rsquo;t always collect early likes, but it usually shows its quality in other ways. Novelty is one — does the piece cover ground the platform hasn&rsquo;t seen? Structure is another; clean formatting, decent grammar and clarity correlate with quality more than we&rsquo;d like to admit. An author&rsquo;s track record is a fair prior too, judged on how their past work held readers rather than on raw like counts. Early comments help as well: three genuinely positive comments beat thirty reflexive upvotes, and a bit of semantic analysis can flag content that fills a gap the platform has been missing.</p>

<h3 class="relative group">Borrow signals from outside
    <div id="borrow-signals-from-outside" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#borrow-signals-from-outside" aria-label="Anchor">#</a>
    </span>
    
</h3>
<p>The wider internet tells you what&rsquo;s resonating right now. Content that lines up with trending search keywords has built-in relevance, and quick publishing on a rising topic deserves a boost. What&rsquo;s buzzing on Twitter or Reddit is often a preview of what your own users will want next — with a little prediction you can promote matching fresh content before the wave instead of after it. Your own search box is underrated too: if users keep searching for a topic, or a piece pulls steady traffic in from search engines, that&rsquo;s relevance you didn&rsquo;t have to guess at.</p>

<h3 class="relative group">Blending the signals into one score
    <div id="blending-the-signals-into-one-score" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#blending-the-signals-into-one-score" aria-label="Anchor">#</a>
    </span>
    
</h3>
<p>Here&rsquo;s the idea in a few lines of Python. It&rsquo;s an illustration, not a production system. Each signal is normalised to a 0 to 1 range so that one with big raw numbers can&rsquo;t drown the rest, a weighted sum turns them into a single score, and a time decay gives a new post a fair shot against last month&rsquo;s hit.</p>
<div class="highlight-wrapper"><div class="highlight"><pre tabindex="0" class="chroma"><code class="language-python" data-lang="python"><span class="line"><span class="cl"><span class="n">WEIGHTS</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&#34;read_depth&#34;</span><span class="p">:</span> <span class="mf">0.3</span><span class="p">,</span> <span class="s2">&#34;revisits&#34;</span><span class="p">:</span> <span class="mf">0.25</span><span class="p">,</span> <span class="s2">&#34;quality&#34;</span><span class="p">:</span> <span class="mf">0.25</span><span class="p">,</span> <span class="s2">&#34;early_comments&#34;</span><span class="p">:</span> <span class="mf">0.2</span><span class="p">}</span>
</span></span><span class="line"><span class="cl"><span class="n">HALF_LIFE_HOURS</span> <span class="o">=</span> <span class="mi">48</span>  <span class="c1"># a post loses half its score every two days</span>
</span></span><span class="line"><span class="cl">
</span></span><span class="line"><span class="cl"><span class="k">def</span> <span class="nf">normalise</span><span class="p">(</span><span class="n">values</span><span class="p">):</span>
</span></span><span class="line"><span class="cl">    <span class="n">lo</span><span class="p">,</span> <span class="n">hi</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">values</span><span class="p">),</span> <span class="nb">max</span><span class="p">(</span><span class="n">values</span><span class="p">)</span>
</span></span><span class="line"><span class="cl">    <span class="k">return</span> <span class="p">[(</span><span class="n">v</span> <span class="o">-</span> <span class="n">lo</span><span class="p">)</span> <span class="o">/</span> <span class="p">(</span><span class="n">hi</span> <span class="o">-</span> <span class="n">lo</span><span class="p">)</span> <span class="k">if</span> <span class="n">hi</span> <span class="o">&gt;</span> <span class="n">lo</span> <span class="k">else</span> <span class="mf">0.0</span> <span class="k">for</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">values</span><span class="p">]</span>
</span></span><span class="line"><span class="cl">
</span></span><span class="line"><span class="cl"><span class="k">def</span> <span class="nf">rank</span><span class="p">(</span><span class="n">items</span><span class="p">):</span>
</span></span><span class="line"><span class="cl">    <span class="s2">&#34;&#34;&#34;items: dicts holding raw signal values plus &#34;age_hours&#34;.&#34;&#34;&#34;</span>
</span></span><span class="line"><span class="cl">    <span class="n">norm</span> <span class="o">=</span> <span class="p">{</span><span class="n">name</span><span class="p">:</span> <span class="n">normalise</span><span class="p">([</span><span class="n">item</span><span class="p">[</span><span class="n">name</span><span class="p">]</span> <span class="k">for</span> <span class="n">item</span> <span class="ow">in</span> <span class="n">items</span><span class="p">])</span> <span class="k">for</span> <span class="n">name</span> <span class="ow">in</span> <span class="n">WEIGHTS</span><span class="p">}</span>
</span></span><span class="line"><span class="cl">    <span class="n">scores</span> <span class="o">=</span> <span class="p">[]</span>
</span></span><span class="line"><span class="cl">    <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">item</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">items</span><span class="p">):</span>
</span></span><span class="line"><span class="cl">        <span class="n">base</span> <span class="o">=</span> <span class="nb">sum</span><span class="p">(</span><span class="n">weight</span> <span class="o">*</span> <span class="n">norm</span><span class="p">[</span><span class="n">name</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="k">for</span> <span class="n">name</span><span class="p">,</span> <span class="n">weight</span> <span class="ow">in</span> <span class="n">WEIGHTS</span><span class="o">.</span><span class="n">items</span><span class="p">())</span>
</span></span><span class="line"><span class="cl">        <span class="n">scores</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">base</span> <span class="o">*</span> <span class="mf">0.5</span> <span class="o">**</span> <span class="p">(</span><span class="n">item</span><span class="p">[</span><span class="s2">&#34;age_hours&#34;</span><span class="p">]</span> <span class="o">/</span> <span class="n">HALF_LIFE_HOURS</span><span class="p">))</span>
</span></span><span class="line"><span class="cl">    <span class="n">order</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">items</span><span class="p">)),</span> <span class="n">key</span><span class="o">=</span><span class="k">lambda</span> <span class="n">i</span><span class="p">:</span> <span class="n">scores</span><span class="p">[</span><span class="n">i</span><span class="p">],</span> <span class="n">reverse</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
</span></span><span class="line"><span class="cl">    <span class="k">return</span> <span class="p">[</span><span class="n">items</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">order</span><span class="p">]</span></span></span></code></pre></div></div>
<p>The weights are guesses until the tests below say otherwise. Views and likes can still go in; they just become one input among several instead of the whole score. The half-life decides how quickly old posts sink, which is a topic of its own: I went into it in <a href="https://jared.lynskey.co.nz/en/posts/2023/2023-09-27-simulating-views/" >how feed ranking algorithms decide when to bury old posts</a>. Real systems usually learn the weights with a model rather than picking them by hand, but the shape is the same: many signals in, one number out, sort.</p>

<h3 class="relative group">Then test it honestly
    <div id="then-test-it-honestly" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#then-test-it-honestly" aria-label="Anchor">#</a>
    </span>
    
</h3>
<p>However clever the signals, A/B test them — show new content to a slice of users and compare its engagement against the incumbents. And watch how things spread: shares, mentions and embeds, on and off the platform, say more about value than a view counter ever will.</p>
<p>No single metric survives contact with reality. The rankings that actually work blend behaviour, intrinsic quality, timeliness and honest experiments, and no one signal gets the final word.</p>
]]></content:encoded>
      <media:content url="https://jared.lynskey.co.nz/en/posts/2023/2023-10-06-reinenforcement-learning/featured.jpeg" medium="image" type="image/jpeg" />
    </item>
    <item>
      <title>SEO Trends: Follow the Best Practices or Set Your Own?</title>
      <link>https://jared.lynskey.co.nz/en/posts/2023/2023-10-04-following-seo-trends/</link>
      <guid isPermaLink="true">https://jared.lynskey.co.nz/en/posts/2023/2023-10-04-following-seo-trends/</guid>
      <pubDate>Wed, 04 Oct 2023 00:30:03 +0000</pubDate>
      <dc:creator>Jared Lynskey</dc:creator>
      <category>Search Engine Optimization</category>
      <category>SEO</category>
      <category>Algorithms</category>
      <category>Digital Marketing</category>
      <category>Content Strategy</category>
      <category>Online Visibility</category>
      <description>Should you follow the latest SEO trends and best practices, or try your own? When copying the crowd pays, when an experiment is worth it, and the one rule that outlasts every algorithm update.</description>
      <content:encoded><![CDATA[<p>Every SEO guide I&rsquo;ve read this year says roughly the same things in roughly the same order. That&rsquo;s either reassuring or a warning sign, depending on how you look at it. I&rsquo;ve been chewing on this for a while — partly for this blog, partly at work, where organic search matters more than I&rsquo;d like to admit.</p>
<p><em>Update, October 2026: since I wrote this, a lot of searches get an AI-written answer before any links, from Google&rsquo;s AI Overviews, ChatGPT search and the like. Those answers cite and link the pages they drew on, and the easiest page to quote is a useful one that says the answer plainly. That makes the dull rule at the end of this post matter more, not less. I&rsquo;ve written up what I changed on this blog because of it in <a href="https://jared.lynskey.co.nz/en/posts/2026/2026-09-16-blog-as-a-source-for-ai-assistants/" >Optimising a Blog to Be the Source AI Assistants Cite</a>.</em></p>

<h3 class="relative group">Why following the crowd mostly works
    <div id="why-following-the-crowd-mostly-works" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#why-following-the-crowd-mostly-works" aria-label="Anchor">#</a>
    </span>
    
</h3>
<p>Best practices are best practices for a reason. They&rsquo;re distilled from data across thousands of sites, and if you&rsquo;re new to SEO they&rsquo;ll get you most of the way there without drama. Staying on the beaten path also means fewer nasty surprises — you&rsquo;re unlikely to get punished by an algorithm update for doing exactly what Google&rsquo;s own documentation recommends. And since the algorithms keep shifting, just keeping up with the current guidance is honest work in itself.</p>
<p>The catch is that everyone else is reading the same guides. When every site in a niche has the same structure, the same keyword-heavy headings and the same FAQ block bolted to the bottom, none of them stand out. You also end up permanently reactive: waiting for the next update, then scrambling to comply.</p>

<h3 class="relative group">The case for trying your own thing
    <div id="the-case-for-trying-your-own-thing" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#the-case-for-trying-your-own-thing" aria-label="Anchor">#</a>
    </span>
    
</h3>
<p>Experiments are how you find the tactics that fit your site specifically. What works for a cosmetics community isn&rsquo;t what works for a recipe blog, and no generic checklist will tell you the difference. If an experiment lands, you get a stretch of time where you&rsquo;re the only one doing it. And if it lands publicly, people start treating you as the one to copy, which has its own compounding value.</p>
<p>The honest downside is that most experiments fail quietly. You can sink weeks into one and have nothing to show for it. If search traffic is what keeps the lights on, that&rsquo;s a hard sell to whoever approves your time.</p>

<h3 class="relative group">Where I&rsquo;ve landed
    <div id="where-ive-landed" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#where-ive-landed" aria-label="Anchor">#</a>
    </span>
    
</h3>
<p>Do the boring stuff first, and do it properly. Then carve out a small, contained slice of effort for experiments, and actually measure them instead of judging by vibes. A small site should lean heavily on the established stuff and only experiment at the margins; a big organisation with resources to burn can afford to place real bets.</p>
<p>The one rule that&rsquo;s survived every algorithm update is also the least exciting one: make something genuinely useful for the person searching. Everything else is fashion.</p>
]]></content:encoded>
      <media:content url="https://jared.lynskey.co.nz/en/posts/2023/2023-10-04-following-seo-trends/featured.jpeg" medium="image" type="image/jpeg" />
    </item>
    <item>
      <title>What Is Emotional Valence? Valence, Arousal and Why Content Goes Viral</title>
      <link>https://jared.lynskey.co.nz/en/posts/2023/2023-10-05-viral-success/</link>
      <guid isPermaLink="true">https://jared.lynskey.co.nz/en/posts/2023/2023-10-05-viral-success/</guid>
      <pubDate>Wed, 04 Oct 2023 00:30:03 +0000</pubDate>
      <dc:creator>Jared Lynskey</dc:creator>
      <category>Viral Content</category>
      <category>Marketing</category>
      <category>Emotional Appeal</category>
      <category>Content Strategy</category>
      <category>Social Media</category>
      <category>Engagement</category>
      <description>Emotional valence is how pleasant or unpleasant a feeling is; arousal is how fired up it makes you. What research says about which one makes content go viral, with examples.</description>
      <content:encoded><![CDATA[<p>Emotional valence is how positive or negative an emotion feels: pleasant at one end, unpleasant at the other. Joy and awe have positive valence; anger, fear and sadness have negative valence. In marketing and communication the same word describes a message, so positive-valence content leaves people feeling good and negative-valence content leaves them feeling bad.</p>
<p>Valence usually comes paired with arousal, which is how activating an emotion is, from calm and sluggish to fired up. Psychologists often place emotions on both axes at once, an idea that goes back to James Russell&rsquo;s circumplex model of affect (1980): excitement is pleasant and activated, contentment is pleasant and calm, anger is unpleasant and activated, sadness is unpleasant and calm. If you want content shared, the second axis turns out to matter at least as much as the first.</p>
<p>Working on a content platform gives you a front-row seat to a strange lottery: two posts of similar quality go out, one dies quietly, the other takes off. The closest thing I&rsquo;ve found to an explanation is the emotion a piece of content carries: which way it pulls (its valence, positive, negative or neutral) and how hard it pulls (its arousal).</p>

<h3 class="relative group">Positive, negative, neutral
    <div id="positive-negative-neutral" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#positive-negative-neutral" aria-label="Anchor">#</a>
    </span>
    
</h3>
<p>Positive-valence content makes people feel good: heartwarming stories, dumb memes, uplifting news. It gets shared because passing it on is a small gift, a break from the daily grind. What surprised me more is how well negative valence performs. Anger, anxiety and disgust are powerful sharing engines — people pass along stories about injustice, controversial figures or societal problems to raise awareness, or to rally others to their side. Then there&rsquo;s neutral content, which doesn&rsquo;t stir much feeling but earns shares on pure value: a genuinely fascinating fact, a new discovery, a tutorial that solves a real problem.</p>

<h3 class="relative group">Arousal is what travels
    <div id="arousal-is-what-travels" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#arousal-is-what-travels" aria-label="Anchor">#</a>
    </span>
    
</h3>
<p>The pattern that holds across all of it: the stronger the emotion, the more likely people are to comment, like or share. A mildly pleasant post loses to an infuriating one almost every time. Whether that says anything flattering about us is a separate question.</p>
<p>There&rsquo;s decent research behind that hunch, and it&rsquo;s sharper than &ldquo;stronger is better&rdquo;. Jonah Berger and Katherine Milkman took nearly 7,000 New York Times articles published over three months and looked at which ones made the paper&rsquo;s most-emailed list (&ldquo;What Makes Online Content Viral?&rdquo;, <em>Journal of Marketing Research</em>, 2012). Positive articles were shared more than negative ones, so valence mattered. But arousal mattered more: articles that evoked high-arousal emotions, positive (awe) or negative (anger, anxiety), were more likely to make the list, and articles that evoked sadness, a low-arousal emotion, were less likely to. That held even after they controlled for how surprising, interesting or useful an article was and how prominently the site featured it, and their experiments pointed the same way.</p>
<p>So what travels isn&rsquo;t strength of feeling as such; it&rsquo;s arousal. Sadness can be deeply felt and still not travel, because it&rsquo;s a deactivating emotion: it makes people slow down rather than do something.</p>

<h3 class="relative group">Valence vs arousal, with examples
    <div id="valence-vs-arousal-with-examples" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#valence-vs-arousal-with-examples" aria-label="Anchor">#</a>
    </span>
    
</h3>
<table>
	<thead>
			<tr>
					<th></th>
					<th>High arousal</th>
					<th>Low arousal</th>
			</tr>
	</thead>
	<tbody>
			<tr>
					<td><strong>Positive valence</strong></td>
					<td>Awe, excitement, amusement: a jaw-dropping discovery, a meme that makes you laugh out loud, an underdog win</td>
					<td>Contentment, calm, relief: a cosy routine video, a gentle how-to, a pretty photo</td>
			</tr>
			<tr>
					<td><strong>Negative valence</strong></td>
					<td>Anger, anxiety, outrage: an injustice story, a scam warning, a heated debate</td>
					<td>Sadness, boredom, disappointment: a sombre obituary, a bleak trend piece</td>
			</tr>
	</tbody>
</table>
<p>If Berger and Milkman are right, most shares come from the high-arousal column, and the low-arousal negative corner is where posts quietly go to die.</p>

<h3 class="relative group">What tends to take off
    <div id="what-tends-to-take-off" class="anchor"></div>
    
    <span
        class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none">
        <a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#what-tends-to-take-off" aria-label="Anchor">#</a>
    </span>
    
</h3>
<p>A few kinds of content start with an advantage. Debatable ideas, because people share to agree or to object, and polarising topics gain traction fast. Humour that leans on shared experience — the inside joke a whole community gets at once. Human-interest stories about ordinary people getting through hard things. Anything riding a trending topic with a genuinely fresh angle rather than a reheated take. And interactive formats — quizzes, polls, challenges — because participating is already halfway to sharing.</p>
<p>None of this adds up to a formula, and I&rsquo;m suspicious of anyone selling one. But if I had to bet on a single variable, it&rsquo;s this: content that fires people up, one way or the other, beats content that leaves them calm, week after week.</p>
]]></content:encoded>
      <media:content url="https://jared.lynskey.co.nz/en/posts/2023/2023-10-05-viral-success/featured.jpeg" medium="image" type="image/jpeg" />
    </item>
  </channel>
</rss>
