1PDF: From 12% to 31% AI Visibility in 30 Days
1PDF is a free online PDF tool that improved its AI visibility from 12% to 31% in just 30 days. When people asked ChatGPT, Perplexity, or Google's AI Overviews which PDF tool to use, the brand went from appearing in roughly one of eight answers to nearly one in three. Here is what the team did, week by week.
AI visibility is not the same as search rankings. A page can rank on page one of Google yet never be cited by an AI answer engine, because generative models build answers from a different mix of sources: structured data, community discussions, review platforms, and pages that answer questions directly. For a tool like 1PDF, that difference decides whether a searcher ever reaches you.
Why AI Visibility Mattered for 1PDF
The PDF tool category is crowded. Dozens of free converters, editors, and compressors compete for the same queries, and most have optimized for classic search for years. That is why 1PDF could move the needle in 30 days: almost none of the competitors had worked on the signals AI models actually rely on.
Answer engines prefer sources they can parse reliably: clean markup, direct answers, and third-party validation from platforms like G2 and Reddit. When a model is asked "what is the best free PDF compressor," it synthesizes an answer from whatever it can verify, and tools with none of those signals do not get mentioned. That was 1PDF's starting point.
The Starting Point
Before the sprint, the team measured the brand across a panel of AI models and found:
- AI Visibility: 12%
- No Schema markup
- No comparison pages
- No FAQ pages
- No G2 presence
None of this was unusual, but each gap mapped to a concrete fix that required no redesign and no months of content production.
What Was Done
The sprint ran in four weekly phases, each with a single focus.
Week 1: Technical Foundation
- Added Organization + Product Schema
- Created llms.txt file
- Fixed robots.txt for AI crawlers
Week one made the site easy for AI systems to read. Organization and Product schema gave models unambiguous facts about the company and product, and the llms.txt file provided a machine-readable summary of the key pages, a shortcut crawlers increasingly use before citing a site. Fixing robots.txt mattered because several AI crawlers were being blocked.
Week 2: Content Creation
- Created 3 comparison pages
- Built FAQ page
- Added structured pricing
Week two produced the content that answer engines actually quote. Comparison pages targeting the main competitors gave models a reason to mention 1PDF when users asked about alternatives, since "vs" questions are among the most common in the category. The FAQ page answered buying questions in short, quotable form, and pricing was added in a structured table so models could state it accurately.
Week 3: Authority Building
- Created G2 profile
- Collected 5 initial reviews
- Participated in relevant Reddit threads
Week three attacked the trust problem. The team created a G2 profile, collected five initial reviews from real users, and joined relevant Reddit threads about PDF tool recommendations, answering helpfully and mentioning 1PDF only where it genuinely fit. These third-party signals make an answer engine comfortable recommending a brand it has never met directly.
Week 4: Monitoring & Optimization
- Tracked progress daily
- Made adjustments based on data
- Documented what worked
The final week was about steering. The team checked AI visibility daily, watched which pages and topics gained mentions, and adjusted content based on what the models were quoting. They also documented everything, turning the sprint into a repeatable process.
The Results
| Metric | Before | After | Change |
|---|---|---|---|
| AI Visibility | 12% | 31% | +19pt |
| Mention Rate | 12% | 31% | +19pt |
| Citation Rate | 2% | 15% | +13pt |
| Share of Voice | 5% | 18% | +13pt |
Two patterns stand out. Citation rate grew from 2% to 15%, a far larger relative jump than mention rate, suggesting the work made the site genuinely more citable, not just more discussed. And share of voice nearly tripled, meaning 1PDF took mentions away from competitors.
What Drove the Biggest Gains
The largest contributors were the technical foundation and the comparison pages. Schema and llms.txt made the site legible to crawlers, and the comparison content gave models a natural place to insert the brand into answers. The G2 profile and reviews did more for recommendation quality than raw mention volume: a cited brand with reviews is far more likely to be recommended.
Key Learnings
- Quick wins matter — Schema + FAQ had immediate impact
- Comparison pages work — "vs" queries drive visibility
- Third-party validation helps — G2 reviews built trust
- Consistency compounds — Weekly monitoring caught issues early
You do not need a content machine to move the needle; schema and FAQ landed within days because they fixed a basic readability problem. Comparison content is disproportionately effective in AI answers, where "versus" questions are common. And the weekly cadence mattered: the team caught a stale pricing detail before it cost a week of lost mentions.
Apply to Your Brand
- Start with quick wins — Schema, FAQ, pricing page
- Create comparison content — For top competitors
- Build authority — G2, Reddit, editorial
- Monitor weekly — Track progress and adjust
Run your own baseline first; a free AI visibility check gives you a starting number in minutes. Then follow the same order: fix readability, add answer-ready content, build third-party trust, and measure weekly.
FAQ
How long does it take to see AI visibility improvements? In this case, movement showed up within two weeks and the full gain by day 30. Quick wins like schema and FAQ show within days; authority signals take longer.
Do I need a G2 profile to improve AI visibility? No, but review platforms are among the strongest trust signals. If G2 is not a fit, look for the equivalent: Product Hunt, Capterra, Trustpilot, or community recommendations.
Does traditional SEO help AI visibility? Yes, but not automatically. Pages that rank well and are well-structured tend to be cited more, but AI visibility also requires direct answers and third-party validation.
Can a small team do this in 30 days? Yes, this sprint was run by a small team, and the work was mostly content and configuration. The bottleneck is consistency, not headcount.
30 days is enough to see significant AI visibility improvement.