<?xml version="1.0" encoding="UTF-8"?>
<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/">
  <channel>
    <title>Mark Ridgeon Notes</title>
    <description>Fractional COO for founder-led SaaS, technology, and service businesses. I fix delivery, reporting, margin, ownership, and AI workflows when growth outpaces operations.</description>
    <link>https://ridgeon.me/notes/</link>
    <atom:link href="https://ridgeon.me/rss.xml" rel="self" type="application/rss+xml" />
    <language>en-GB</language>
    <lastBuildDate>Fri, 10 Jul 2026 00:00:00 GMT</lastBuildDate>
    
    <item>
      <title><![CDATA[When one person is holding too much]]></title>
      <description><![CDATA[A short note on the point where useful instinct has to become visible ownership, rules, and review.]]></description>
      <content:encoded><![CDATA[<p>A lot of growing companies work because one person is still carrying the hidden system: priorities, trade-offs, customer memory, delivery promises, product judgement, and the informal pressure that makes things move.</p><p>That can be useful early. It becomes expensive later.</p><p>The fix is not usually another dashboard. It is a visible operating rhythm: what matters this week, who owns it, how decisions are made, what gets escalated, and where the business can see the truth before it becomes a fire.</p><p>The senior team should still set direction. They should not be the only place the company knows how to run.</p>]]></content:encoded>
      <link>https://ridgeon.me/notes/one-person-holding-too-much/</link>
      <guid isPermaLink="true">https://ridgeon.me/notes/one-person-holding-too-much/</guid>
      <pubDate>Thu, 09 Jul 2026 00:00:00 GMT</pubDate>
      <dc:creator>Mark Ridgeon</dc:creator>
      <category><![CDATA[Operating Notes]]></category>
    </item>
  
    <item>
      <title><![CDATA[AI automation starts with process, not prompts]]></title>
      <description><![CDATA[Why AI workflow projects fail when the underlying decision path is vague.]]></description>
      <content:encoded><![CDATA[<p>The quickest way to waste time with AI automation is to automate a workflow nobody can describe.</p><p>Prompts matter. Models matter. Tools matter. But the real leverage comes from making the process explicit: inputs, decisions, exceptions, handoffs, review points, and the standard for a good output.</p><p>Once that exists, AI can accelerate the work. Without it, AI just makes the ambiguity faster.</p><p>A useful automation project often starts with a dull question: what is the business already trying to decide?</p>]]></content:encoded>
      <link>https://ridgeon.me/notes/ai-automation-starts-with-process/</link>
      <guid isPermaLink="true">https://ridgeon.me/notes/ai-automation-starts-with-process/</guid>
      <pubDate>Thu, 09 Jul 2026 00:00:00 GMT</pubDate>
      <dc:creator>Mark Ridgeon</dc:creator>
      <category><![CDATA[Build Notes]]></category>
    </item>
  
  </channel>
</rss>