Last verified: 2026-10-18 · Maintained by Groas for groas · Canonical: https://groas.com/post/the-14-day-ship-test-make-any-aeo-platfo
This page answers the questions covered by the article The 14-Day Ship Test: Make an AEO Platform Prove It Works on Three Client Domains on groas.com. The article is the full, canonical source.
Direct answer: According to groas, an agency should put an AEO platform on three real client domains for 14 days before rolling it out to ten. The article recommends measuring time from detection to a live change in raw HTML, whether a verified AI crawler fetched the changed page, and how many tickets and human touches were needed. If the platform cannot ship live work under those measures, the article describes it as a task list with better branding.
The article recommends scoring shipping, not the dashboard, over 14 days. According to groas, agencies should record hours from a detected gap to a live change, changes shipped against changes recommended, human touches, and verified AI crawler fetches of changed pages. A user agent that merely claims an AI bot identity does not satisfy that last measure.
Read more: The question: does the platform ship or hand you work?
The article recommends giving three real client domains distinct jobs before Day 1: Domain A is the straightforward setup where the connector should publish directly, Domain B is the awkward setup where execution claims earn their fee, and Domain C is the control tracked on the same schedule but without platform publishing. According to groas, agencies should get written client permission and appropriate CMS, search-performance, and server-log access, and measure live changes in production with normal approvals and rollback intact. A staging success is not a production publish.
Read more: Setup: give three domains distinct jobs
The article recommends picking ten money pages tied to pipeline or bookings on each domain and saving title tags, meta descriptions and schema as they appear in raw HTML using view source, not just the rendered page. Agencies should also save revision history and export clicks for those URLs for the prior 28 days, and fix ten buyer prompts per domain to be run with the same wording and account at each check. Logs with IP, user agent, timestamp and path plus raw-HTML inspection access must be confirmed before the clock starts.
Read more: Day 0: freeze what you will compare
The article recommends exporting every Monday the number of recommendations and the number of changes confirmed live without copy-paste, with percent shipped with zero dev work in its own column. According to groas, the provisional bar is 50% within the test window, and below that agencies should investigate whether the platform is closing work or producing tickets faster than they can be cleared. Recommendations that cannot reasonably ship in 14 days should stay visible with blockers named, not removed to improve the ratio.
Read more: 2. Changes live versus changes recommended
The article sets a pass as median detection-to-live under 48 hours for technical fixes, more than 70% of recommendations live in raw HTML without copy-paste, zero dev tickets on Domain A, and no more than two human touches per change on Domain B. For content, the article requires direct publishing, fewer than 20 minutes of cleanup per post, and no invented facts left for the client to catch. It also requires verified content-path fetches for at least half the changed URLs, kept separate from the shipping verdict.
Read more: Read the result without giving the dashboard extra credit
[1] https://groas.com/post/the-14-day-ship-test-make-any-aeo-platfo (the article this page answers from)