Back to GrowthDex Blog

GrowthDex Blog

Fix the answer before you chase AI traffic

A plain essay on Glasp's AEO treatment: canonical URLs, 404 demand mining, question titles, short lead answers, and a guardrail for pages that already earn Google clicks.

Published 2026-06-10 AI Search SEO content systems SaaS AI products creator tools video products developer tools marketplaces
Ian Goh Updated 2026-06-10T03:13:40.000Z 5 linked tactics 3 sources
arXiv: Glasp AEO natural experiment
SEO path 5 linked tactics 3 sources

arXiv: Glasp AEO natural experiment + 2 more

Most AI-search advice starts too late. It starts with prompts, rankings, dashboards, and screenshots of a chatbot answer.

Glasp's study is more useful because it starts with the pages. The team changed a large YouTube Q&A corpus, measured ChatGPT referral traffic with first-party data, and compared treated pages against untreated pages on the same domain. The result was not a miracle claim. It was a bounded lift with caveats, which makes it easier to trust.

Give each answer one home

Glasp one URL per video before AEO rewrite is the plumbing move. If the same answer lives at a slug URL and a video-ID URL, the corpus is asking crawlers to choose between copies before they even judge the answer.

That sounds boring. It is also the kind of boring that compounds. A stable URL makes analytics cleaner, links cleaner, and later rewrites easier to audit.

Use broken bot paths as demand research

Glasp AI-bot 404 logs as page demand map turns a server log into a research tool. If bots keep asking for missing pages in the same pattern, the market may be showing you a page shape you have not built yet.

The caveat matters. A bot hit is not a customer. Do not create thin pages for every strange crawl. Group the repeated patterns, check that you can answer them well, then build the pages that deserve to exist.

Make the top of the page answer the question

Glasp question title rewrite from bot demand and Glasp standalone TLDR answer before long page belong together. The title frames the question. The first two or three sentences answer it.

This is good for people too. A reader should not need to scroll through setup language to learn whether the page is useful. In markets with noisy feeds and short attention, the first answer has to carry its weight.

Do not break pages that already work

Glasp SEO Guard before AI-search rewrite queue is the restraint every AI-search project needs. Lock pages with meaningful Google clicks. Remove pages with no organic or AI interest. Rewrite the middle.

That is the part I would copy first for a startup. AI search is still messy to measure. If a page already earns trust from Google, do not casually rewrite it because a new dashboard made the team impatient.

What I would test this week

Pick one useful page family: docs, templates, videos, comparisons, city pages, marketplace listings, or support articles. Fix duplicate URLs. Pull the top missing URL patterns from AI-bot 404s. Rewrite only the pages with clear demand and weak openings. Protect pages with recent Google clicks. Then compare the changed group against a similar unchanged group.

That last comparison is the honest part. Glasp's own paper warns that platform tailwind can make every AEO number look bigger than it is. A founder does not need a perfect academic experiment. But they do need a control group before they start congratulating the tactic for growth the whole platform delivered.

Start with these related tactics

If this essay matches what you are working on, these are the first tactic pages to open.

Sources, tactic links, and more reading

All tactic links

Sources

Essay chronology

If this piece was useful, move one step newer or older instead of bouncing back to the full archive.

Keep reading

Continue through the blog

Why this is worth your time

GrowthDex starts with tactics that founders, marketers, and product teams have actually tried. Each essay turns the evidence into a practical move you can test without pretending one case study is a guarantee.

Ian Goh has helped grow consumer platforms across Southeast Asia, India, and MENA. His work includes scaling Tiki to 100M+ users, doubling BIGO's MENA revenue in 7 months, and increasing OYO's direct booking share across 6 Southeast Asian markets.

Editing notes

Want a growth system instead of loose tactics?

Ian works with founders on growth, market entry, creator economy loops, and operator-led distribution.

Work with Ian on growth advisory