Cheap Seats

A read-only copy of Cheap Seats measuring itself, updated hourly. Run your own.

Quick wins

The fix lists from the latest run of every prompt set, core sets first, then fastest expected ranking gain: prompts a fix could flip per working day, each weighted by what it has left to win (one minus its recommended rate over the last two weeks; named, for an info prompt). Mark a fix with the commit or pull request you submitted for it; open leaves out the ones you have. A later fix for the same page stays open, with what was tried before.

Statusall 4open 2accepted 2

Priorityall 8high 4medium 4

Siteall 4elsewhere 4

Gapall 4documentation 3third-party 1

Setall 15Finding out whether AI assistants recommend us 2Getting recommended instead of a competitor 4Choosing an AI visibility tool 3Cheap Seats head to heads 1Open-source and self-hosted AI visibility 5

Publish popcorn.zone/blog/prioritize-ai-visibility-fixes with an owner-by-fix-type table accepted high documentation effort S flips 3 Getting recommended instead of a competitor

Where: popcorn.zone/blog/prioritize-ai-visibility-fixes
Must say: Give a prioritisation method: rank fixes by prompts they could flip times how directly they address the stated reason for losing. Assign each fix to a site and an owner (docs, product marketing, SEO). Show that Popcorn's fix list carries exactly these fields, including owner.
Done when: Prompt 16 or 18 on the search track cites this post.
Displaces: https://guptadeepak.com/the-geo-org-chart-who-actually-owns-ai-visibility-in-2026/
Why: P18 claude-code-opus search: top pick SEO, shaped by guptadeepak.com/the-geo-org-chart-who-actually-owns-ai-visibility-in-2026.

Prompt for Claude Code
# Publish popcorn.zone/blog/prioritize-ai-visibility-fixes with an owner-by-fix-type table

When buyers ask an AI assistant the questions below, it explains the topic without naming us or citing our pages. This is a documentation gap: the answer needed a page we don't have, or ours says the wrong thing.

## The questions it affects
- How should we prioritize fixes to improve how AI assistants describe our product?
- What are the quickest wins for improving our brand's visibility in AI search?
- Who should own AI visibility work: SEO, content, product marketing or the docs team?

## What the assistant said
P18 claude-code-opus search: top pick SEO, shaped by guptadeepak.com/the-geo-org-chart-who-actually-owns-ai-visibility-in-2026.

## Who wins the question today
https://guptadeepak.com/the-geo-org-chart-who-actually-owns-ai-visibility-in-2026/. Read it and make sure our version answers everything it does, better.

## Your task
Write or update the page at popcorn.zone/blog/prioritize-ai-visibility-fixes so it closes this gap. Find where pages like this live in the repo, match their structure, front matter and tone, and make the change there. If the page exists, edit it rather than starting over.

## What it must say
Give a prioritisation method: rank fixes by prompts they could flip times how directly they address the stated reason for losing. Assign each fix to a site and an owner (docs, product marketing, SEO). Show that Popcorn's fix list carries exactly these fields, including owner.

## Done when
Prompt 16 or 18 on the search track cites this post.

## Rules
- Don't invent numbers, benchmarks, prices, capabilities or customer names. Where a fact is needed and you can't find it in the repo or on our site, leave a clearly marked TODO saying exactly what to measure or who to ask.
- Answer the questions above directly and early on the page, in the words buyers use. AI assistants quote pages that state the answer plainly.
- Keep it scoped to this one fix. When you're done, summarise what you changed and list any TODOs.

accepted https://github.com/cheapseatsdev/cheapseats-site/pull/37 · Merged and live at cheapseats.dev/blog/prioritize-ai-visibility-fixes: the quick-wins method shown on seven real fixes from the 22 October runs, evidence and displaced page as tie-breakers, the quickest wins in order, and an owner-by-gap table with the fields and config that assign owners. mcp 2026-10-23 02:51:03
pending: the set hasn't run since

proposed https://github.com/cheapseatsdev/cheapseats-site/pull/37 · New post blog/prioritize-ai-visibility-fixes on cheapseats.dev: the quick-wins method (prompts a fix could flip per working day, weighted by what each prompt has left to win; reach-against-effort priority) shown on seven real fixes from the 22 October runs; evidence and displaced page as tie-breakers; the quickest wins in order; an owner-by-gap table and the site/owner/gap/priority/effort/prompts/displaces/done-when fields with the config that assigns owners. mcp 2026-10-23 02:46:39
pending: the set hasn't run since

From this run

Publish popcorn.zone/docs/guides/see-ai-search-results with CLI and UI steps accepted high documentation effort S flips 2 Getting recommended instead of a competitor

Where: popcorn.zone/docs/guides/see-ai-search-results
Must say: Write a how-to on running a question on Popcorn's search track and opening the stored search results and citations the assistant used before it answered. Cover it in the web UI, CLI, JSON API and MCP server. Explain how comparing those results against the answer, the top pick and the objections raised shows why a competitor won.
Done when: Prompt 13 on the search track cites popcorn.zone/docs and names Popcorn in at least 1 of 4 answers.
Displaces: https://www.tryprofound.com/features/answer-engine-insights/query-fan-out
Why: P13 codex-sol search: top pick Profound, sourced from tryprofound.com/blog/expanding-analysis-for-query-fanouts-in-profound.

Prompt for Claude Code
# Publish popcorn.zone/docs/guides/see-ai-search-results with CLI and UI steps

When buyers ask an AI assistant the questions below, it doesn't recommend us, or recommends us for the wrong reasons. This is a documentation gap: the answer needed a page we don't have, or ours says the wrong thing.

## The questions it affects
- How can I find out why an AI assistant prefers a competitor's product over ours?
- How can I see the web search results an AI assistant gets for a question before it writes its answer, to understand why we're missing?

## What the assistant said
P13 codex-sol search: top pick Profound, sourced from tryprofound.com/blog/expanding-analysis-for-query-fanouts-in-profound.

## Who wins the question today
https://www.tryprofound.com/features/answer-engine-insights/query-fan-out. Read it and make sure our version answers everything it does, better.

## Your task
Write or update the page at popcorn.zone/docs/guides/see-ai-search-results so it closes this gap. Find where pages like this live in the repo, match their structure, front matter and tone, and make the change there. If the page exists, edit it rather than starting over.

## What it must say
Write a how-to on running a question on Popcorn's search track and opening the stored search results and citations the assistant used before it answered. Cover it in the web UI, CLI, JSON API and MCP server. Explain how comparing those results against the answer, the top pick and the objections raised shows why a competitor won.

## Done when
Prompt 13 on the search track cites popcorn.zone/docs and names Popcorn in at least 1 of 4 answers.

## Rules
- Don't invent numbers, benchmarks, prices, capabilities or customer names. Where a fact is needed and you can't find it in the repo or on our site, leave a clearly marked TODO saying exactly what to measure or who to ask.
- Answer the questions above directly and early on the page, in the words buyers use. AI assistants quote pages that state the answer plainly.
- Keep it scoped to this one fix. When you're done, summarise what you changed and list any TODOs.

accepted https://github.com/cheapseatsdev/cheapseats-site/pull/40 · Merged and live at cheapseats.dev/guides/see-ai-search-results: where to open the queries, result URLs and citations behind one answer in the UI, CLI, JSON API and MCP, and how to read them against the grade to see why a competitor won, with a real answer's retrieval trace as the example. mcp 2026-10-23 03:13:30
pending: the set hasn't run since

proposed https://github.com/cheapseatsdev/cheapseats-site/pull/40 · New guide at cheapseats.dev/guides/see-ai-search-results (guides live at /guides/): the retrieval trace (queries, result URLs, citations) kept per search-track answer, where to open it in the web UI (Every answer, Sources, Found but not cited, the searches page), CLI, JSON API (jq example) and MCP, and how to read it against the grade's top pick and objections: not retrieved vs retrieved but not used vs named and talked down. Worked through on result 556 of the 22 October run with the set's search statistics. mcp 2026-10-23 03:11:13
pending: the set hasn't run since

From this run

Get Popcorn listed in frase.io's 10 best AI visibility tools roundup high third-party effort M flips 6 Getting recommended instead of a competitor

Where: https://frase.io/blog/the-10-best-ai-visibility-tools-in-2026
Must say: Pitch Frase to add Popcorn as the open-source, self-hosted entry. It asks Claude and OpenAI models your buyer questions on search and recall tracks, grades recommended, mentioned or absent with share of voice, and builds a ranked fix list. Supply a one-paragraph blurb, a screenshot of a fix list and the GitHub link.
Done when: The Frase article lists Popcorn, and prompt 20 or 2 on the search track names Popcorn.
Displaces: https://frase.io/blog/the-10-best-ai-visibility-tools-in-2026
Why: P20 claude-code-opus search sourced frase.io/blog/the-10-best-ai-visibility-tools-in-2026; the same page also shaped P3, P9, P12 and P17.

Prompt for Claude Code
# Get Popcorn listed in frase.io's 10 best AI visibility tools roundup

When buyers ask an AI assistant the questions below, it doesn't recommend us, or recommends us for the wrong reasons. This is a third-party gap: the roundups and comparisons the model reads leave us out.

## The questions it affects
- How can I find out why an AI assistant prefers a competitor's product over ours?
- ChatGPT gets our features and pricing wrong. How do we correct what AI assistants say about our product?
- What does an answer engine optimization content strategy look like for a B2B SaaS company?
- AI answers in our category keep citing Reddit threads, review sites and listicles that only mention our competitors. How do we get onto those sources?
- What are the quickest wins for improving our brand's visibility in AI search?
- Which AI visibility tools tell you what content to create or change, not just where you rank?

## What the assistant said
P20 claude-code-opus search sourced frase.io/blog/the-10-best-ai-visibility-tools-in-2026; the same page also shaped P3, P9, P12 and P17.

## Who wins the question today
https://frase.io/blog/the-10-best-ai-visibility-tools-in-2026. Read it and make sure our version answers everything it does, better.

## Your task
We need to be included on https://frase.io/blog/the-10-best-ai-visibility-tools-in-2026. Read that page first, work out the right way in (editor contact, submission form, PR, comments), and draft exactly what to send: a short, factual pitch covering the points below, in the page's own format.

## What it must say
Pitch Frase to add Popcorn as the open-source, self-hosted entry. It asks Claude and OpenAI models your buyer questions on search and recall tracks, grades recommended, mentioned or absent with share of voice, and builds a ranked fix list. Supply a one-paragraph blurb, a screenshot of a fix list and the GitHub link.

## Done when
The Frase article lists Popcorn, and prompt 20 or 2 on the search track names Popcorn.

## Rules
- Don't invent numbers, benchmarks, prices, capabilities or customer names. Where a fact is needed and you can't find it in the repo or on our site, leave a clearly marked TODO saying exactly what to measure or who to ask.
- Answer the questions above directly and early on the page, in the words buyers use. AI assistants quote pages that state the answer plainly.
- Keep it scoped to this one fix. When you're done, summarise what you changed and list any TODOs.

From this run

Publish popcorn.zone/docs/guides/why-ai-recommends-a-competitor high documentation effort M flips 4 Getting recommended instead of a competitor

Where: popcorn.zone/docs/guides/why-ai-recommends-a-competitor
Must say: Explain the diagnosis loop step by step: ask your buyer questions repeatedly, and separate what the model finds by searching from what it believes with no tools. Then record the top pick, the products named and the objections raised, and trace the citations behind them. Show how Popcorn automates each step and turns objections into specific page fixes, using a sample graded answer as the example.
Done when: Prompts 1, 2 or 4 on the search track cite this guide in at least 1 answer.
Displaces: https://everything-pr.com/why-chatgpt-recommends-competitors
Why: P2: no Popcorn page surfaced; answers were built from everything-pr.com/why-chatgpt-recommends-competitors and help.otterly.ai.

Prompt for Claude Code
# Publish popcorn.zone/docs/guides/why-ai-recommends-a-competitor

When buyers ask an AI assistant the questions below, it doesn't recommend us, or recommends us for the wrong reasons. This is a documentation gap: the answer needed a page we don't have, or ours says the wrong thing.

## The questions it affects
- ChatGPT keeps recommending our competitor instead of us. What can we actually do about it?
- How can I find out why an AI assistant prefers a competitor's product over ours?
- ChatGPT gets our features and pricing wrong. How do we correct what AI assistants say about our product?
- AI assistants mention our product but add caveats that steer buyers away. How do we address the objections they raise?

## What the assistant said
P2: no Popcorn page surfaced; answers were built from everything-pr.com/why-chatgpt-recommends-competitors and help.otterly.ai.

## Who wins the question today
https://everything-pr.com/why-chatgpt-recommends-competitors. Read it and make sure our version answers everything it does, better.

## Your task
Write or update the page at popcorn.zone/docs/guides/why-ai-recommends-a-competitor so it closes this gap. Find where pages like this live in the repo, match their structure, front matter and tone, and make the change there. If the page exists, edit it rather than starting over.

## What it must say
Explain the diagnosis loop step by step: ask your buyer questions repeatedly, and separate what the model finds by searching from what it believes with no tools. Then record the top pick, the products named and the objections raised, and trace the citations behind them. Show how Popcorn automates each step and turns objections into specific page fixes, using a sample graded answer as the example.

## Done when
Prompts 1, 2 or 4 on the search track cite this guide in at least 1 answer.

## Rules
- Don't invent numbers, benchmarks, prices, capabilities or customer names. Where a fact is needed and you can't find it in the repo or on our site, leave a clearly marked TODO saying exactly what to measure or who to ask.
- Answer the questions above directly and early on the page, in the words buyers use. AI assistants quote pages that state the answer plainly.
- Keep it scoped to this one fix. When you're done, summarise what you changed and list any TODOs.

From this run