<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
  <title>MurmTools — Writing</title>
  <subtitle>Research and teardowns. Every piece rests on something we measured.</subtitle>
  <link href="https://murmtools.com/feed.xml" rel="self" />
  <link href="https://murmtools.com/writing" />
  <id>https://murmtools.com/writing</id>
  <updated>2026-09-02T00:00:00Z</updated>
  <author><name>MurmTools</name></author>
  <entry>
    <title>What a store's third-party stack tells you</title>
    <link href="https://murmtools.com/third-party-stack" />
    <id>https://murmtools.com/third-party-stack</id>
    <updated>2026-09-02T00:00:00Z</updated>
    <summary>We maintain fingerprints for 317 third-party domains across 243 companies. A third exist to identify visitors. What a competitor's stack reveals.</summary>
  </entry>
  <entry>
    <title>How accurate are Shopify sales estimates?</title>
    <link href="https://murmtools.com/sales-estimates" />
    <id>https://murmtools.com/sales-estimates</id>
    <updated>2026-09-02T00:00:00Z</updated>
    <summary>No storefront publishes its sales. Every revenue figure you have seen from a spy tool is inferred — and the category puts its own accuracy at 70 to 85 percent.</summary>
  </entry>
  <entry>
    <title>The update that disables itself</title>
    <link href="https://murmtools.com/extension-permissions" />
    <id>https://murmtools.com/extension-permissions</id>
    <updated>2026-09-02T00:00:00Z</updated>
    <summary>Adding a permission disables an extension until every user re-approves. One developer lost ~2,000 users in a day. What the warning means, and which words matter.</summary>
  </entry>
  <entry>
    <title>Affiliate disclosure: where it has to be</title>
    <link href="https://murmtools.com/affiliate-disclosure" />
    <id>https://murmtools.com/affiliate-disclosure</id>
    <updated>2026-09-02T00:00:00Z</updated>
    <summary>A disclosure in the footer of a 4,000-word review is not a disclosure. Placement is the requirement — before the link, where a reader will actually see it.</summary>
  </entry>
  <entry>
    <title>Break-even CPA: the number to compute first</title>
    <link href="https://murmtools.com/break-even-cpa" />
    <id>https://murmtools.com/break-even-cpa</id>
    <updated>2026-09-02T00:00:00Z</updated>
    <summary>Margin says whether a product is viable. Break-even CPA says whether you can afford to advertise it — worked from a real landed-cost model, including the 3x rule.</summary>
  </entry>
  <entry>
    <title>Rate beats the point estimate</title>
    <link href="https://murmtools.com/velocity-not-estimates" />
    <id>https://murmtools.com/velocity-not-estimates</id>
    <updated>2026-09-02T00:00:00Z</updated>
    <summary>Every research tool gives you a number. None gives you the derivative. Why the direction of a figure is more useful than the figure, and why nobody sells it.</summary>
  </entry>
  <entry>
    <title>Blocking AI crawlers: what robots.txt can do</title>
    <link href="https://murmtools.com/block-ai-crawlers" />
    <id>https://murmtools.com/block-ai-crawlers</id>
    <updated>2026-09-02T00:00:00Z</updated>
    <summary>robots.txt is a request, not a fence. Which AI crawlers honour it, why silence means permission, and the one rule most sites get backwards.</summary>
  </entry>
  <entry>
    <title>Product schema: declared vs shown</title>
    <link href="https://murmtools.com/product-schema" />
    <id>https://murmtools.com/product-schema</id>
    <updated>2026-09-02T00:00:00Z</updated>
    <summary>Rich results come from structured data, not from what a shopper sees. The two drift apart constantly — and the machine layer is the one Google reads.</summary>
  </entry>
  <entry>
    <title>The de minimis cliff</title>
    <link href="https://murmtools.com/de-minimis-dropshipping" />
    <id>https://murmtools.com/de-minimis-dropshipping</id>
    <updated>2026-08-24T00:00:00Z</updated>
    <summary>Consignments under €150 into the EU clear without customs duty. Cross it and duty hits every unit at once — 40 units landed 9.6% dearer each than 35.</summary>
  </entry>
  <entry>
    <title>The Consent Gap Index 2026</title>
    <link href="https://murmtools.com/consent-gap" />
    <id>https://murmtools.com/consent-gap</id>
    <updated>2026-08-05T00:00:00Z</updated>
    <summary>We measured 5,294 websites. 96% of those with a consent banner contacted third parties before you could answer it. Method, caveats and raw data included.</summary>
  </entry>
  <entry>
    <title>Shopify spy extensions compared</title>
    <link href="https://murmtools.com/compare" />
    <id>https://murmtools.com/compare</id>
    <updated>2026-07-23T00:00:00Z</updated>
    <summary>Shopify research extensions compared: MurmSpy, Koala Inspector, PPSPY and others — permissions, package size, and measured data versus estimates.</summary>
  </entry>
  <entry>
    <title>Meta Ad Library tools compared</title>
    <link href="https://murmtools.com/ad-library-tools" />
    <id>https://murmtools.com/ad-library-tools</id>
    <updated>2026-07-23T00:00:00Z</updated>
    <summary>Meta Ad Library extensions compared: permissions, what each can actually read, and why replaying Facebook's private API risks your own account.</summary>
  </entry>
  <entry>
    <title>We took apart 40 Shopify spy extensions</title>
    <link href="https://murmtools.com/teardown" />
    <id>https://murmtools.com/teardown</id>
    <updated>2026-07-20T00:00:00Z</updated>
    <summary>We unpacked 40 Shopify research extensions and read the code. What that all-sites access is used for, and how to check any extension yourself.</summary>
  </entry>
</feed>
