Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minuteThere is no demonstrated AI-visibility benefit so far. A February 2026 randomized experiment found no statistically significant increase in AI crawler visits when pages were served as Markdown rather than HTML. That result is not proof that the feature can never help: the experiment was too limited to rule out moderate effects, and it did not measure whether AI systems cited pages in their answers.
What the experiment found
Brandon Punturo’s February 2026 experiment assigned pages on six websites to an HTML control or a Markdown treatment. During the 20-day test, from January 19 to February 8, 2026, the treatment served clean Markdown to detected LLM bots; human visitors and non-LLM crawlers received standard HTML. The detailed sample reports 380 analyzed pages—189 control and 191 treatment—after one page with unusually high Meta/Facebook bot traffic was excluded. The abstract initially describes 381 assigned pages, so 380 is the analyzed count, not the original assignment total. Punturo’s experiment
As an Amazon Associate I earn from qualifying purchases.
| Measure | HTML control | Markdown treatment |
|---|---|---|
| Median AI bot visits per page during the test | 6.0 | 7.0 |
| CUPED-adjusted mean lift | +13.9% for Markdown; 95% confidence interval: −11.2% to +46.8% | |
The estimate favored Markdown, but its confidence interval includes zero, so the overall result was not statistically significant. Punturo’s conclusion was: “We found no statistically significant evidence that serving Markdown to LLM crawlers increases bot visit volume.”
The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →What “AI visibility” means here
The experiment counted bot requests to pages, deduplicated at one-minute granularity, using Profound Agent Analytics. It included activity from bots such as ChatGPT-User, OAI-SearchBot and GPTBot, ClaudeBot, PerplexityBot, Meta/Facebook bots, and DuckDuckGo DuckAssistBot. Googlebot and Bingbot were excluded because AI-related activity cannot be reliably separated from their ordinary search indexing.
#1 Best Overall
That metric is only one part of visibility. The study measured crawler visits—not whether content appeared as a citation in a generated answer, how prominently it appeared, or whether a user clicked through to the site. As Punturo put it, “This experiment measured crawl volume (bot visits to pages), not citations (inclusion in AI-generated answers).” Those outcomes should not be treated as interchangeable.
How certain is the result?
The evidence supports a narrow conclusion: this experiment did not detect a statistically significant overall increase in crawler visits under its setup. It does not establish that Markdown has exactly zero effect in every site, topic, time period, or AI system. The positive point estimate and wide confidence interval leave room for effects in either direction.
Rank #2
- The test had limited power. Its minimum detectable effect was about 41%; the authors say it could not confidently detect moderate effects, including effects around 10–15%.
- The sample and duration were limited. The experiment covered six websites and about three weeks, so it cannot settle what happens across all content types or over longer periods.
- Crawl activity is not downstream visibility. The authors note that crawling and citations are not perfectly correlated and call for future studies that directly measure citation rates.
The earlier January 2–12 period was used as a CUPED covariate in the analysis. That adjustment does not change the central limitation: the measured outcome was bot traffic, not citations or referral behavior. Read the experiment and its limitations
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
What Cloudflare Markdown for Agents does
Cloudflare describes Markdown for Agents as an on-the-fly conversion of HTML to Markdown for enabled zones when a requesting client negotiates the text/markdown format. In other words, it makes a page available in a different delivery format to clients that request it; it is not, by itself, a documented ranking or citation signal. Cloudflare’s Markdown for Agents documentation
Markdown can omit styling and other presentation aimed at human readers, which may make the returned content easier for some clients to consume. But the feature’s documented mechanism does not show that bots visit a page more often—or that an AI system cites it more often—because it is served in Markdown.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What to check before interpreting bot traffic
Markdown delivery is only one factor in whether a crawler can fetch a page. Cloudflare’s controls distinguish Search, Agent, and Training behavior, and access policies can change which bots are able to visit a site. Cloudflare’s current descriptions and defaults are product details that can change, so check its documentation before changing settings. Cloudflare’s overview of its crawler controls and Cloudflare documentation on blocking AI bots
- Confirm that the relevant crawlers are allowed to fetch the pages you want them to access; blocking or rate-limiting them can affect visit counts.
- Keep crawler visits separate from citations, answer prominence, and referrals when evaluating results.
- If testing Markdown, compare equivalent content and structure, use randomized treatment and control pages, and track uncertainty and outliers as well as average changes.
Cloudflare reported in a September 30, 2026 blog that daily AI-agent requests on its network had grown by more than 1,700% over the preceding year. That is a company-reported network statistic about request growth, not evidence that Markdown improves a site’s visibility. Cloudflare’s September 2026 post
Quick Recap
Best Value
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
