Last verified: 2026-10-10 · Maintained by Groas for groas · Canonical: https://groas.com/post/one-marketer-90-minutes-a-week-the-in-ho
This page answers the questions covered by the article One Marketer, 90 Minutes a Week, $0: A 30-Day AI Visibility Playbook on groas.com. The article is the full, canonical source.
Direct answer: The article describes a $0, 30-day routine where one marketer spends 90 minutes per week to test whether AI answers mention the brand. According to groas, the method uses a fixed list of 25 buyer prompts run once per week across four engines for 100 answers a week and 300 total, logged for brand mention, domain citation, position and competitor URL. The repeat counts produce a five-item fix list of missing pages, weak pages and missing third-party presence.
The article describes a 30-day, $0 routine where one marketer spends 90 minutes per week testing 25 fixed buyer prompts. According to groas, Weeks 2, 3 and 4 each produce one run of 25 prompts across four engines, totaling 100 answers a week and 300 by the end of the month. The goal is to learn which prompts mention the brand, which do not, and which competitor URLs appear instead.
Read more: the article
The article recommends spending 30 minutes fixing the 25-prompt list and 60 minutes building a log and weekly summary. According to groas, the list should use buyer wording with five questions from sales calls, ten from search queries tied to enquiries or conversions, and ten from support emails and reviews, covering branded, unbranded and direct-comparison prompts. The log must record date and location, engine, exact prompt text, brand mentioned Y/N, domain cited Y/N with URL, position as first, later or footnote, and competitor named with cited URL.
Read more: 1. Week 1: spend 30 minutes on buyer prompts, 60 on the sheet
The article recommends running each of the 25 prompts once across the same four engines in clean, logged-out sessions with a consistent locale, recording the date and location. According to groas, 100 answers in 90 minutes leaves roughly 54 seconds per answer, so entries should mark brand appearance, domain link with URL, position, and competitor named or cited without pausing for audits or edits. The article treats this as a baseline, not a verdict, and says not to change prompt wording after an awkward result.
Read more: 2. Week 2: spend 90 minutes on the first 100 answers
The article recommends spending 60 minutes rerunning the same 25 prompts under the same conditions and 30 minutes filtering the two runs. According to groas, the review should list missing prompts where the brand appeared in neither run with the URLs cited instead, and repeated competitor placements where the same competitor and page appeared twice. The article says to track mention and citation separately and to treat a two-run pattern as a shortlist for the third pass, not a trend.
Read more: 3. Week 3: spend 60 minutes rerunning, 30 mining the log
The article recommends completing a third identical run for up to 300 logged answers, then counting for each engine the prompts where the brand appeared in at least two of three runs divided by 25 as repeat-mention rate, with the same calculation for domain citations. According to groas, the final 30 minutes should produce a five-item fix list starting with prompts at 0 of 3 for the brand where a competitor appeared 3 of 3. For a weak page, the article recommends answering the question plainly near the top with an opening answer a model could quote in about 40 words.
Read more: 4. Week 4: spend 60 minutes on the third run, 30 on decisions
[1] https://groas.com/post/one-marketer-90-minutes-a-week-the-in-ho (the article this page answers from)