How Tally.so Gets 10% of New Signups from ChatGPT: A Complete GEO Audit

Tally.so, the free online form builder, recently announced that AI search has become their biggest acquisition channel, driving the majority of their new signups. When a SaaS company claims that 10% of their referral traffic comes directly from ChatGPT, we need to pay attention. It signals a fundamental change in how users discover products and how brands should approach visibility. In this comprehensive analysis, I'll break down exactly how Tally.so achieved this through Generative Engine Optimization (GEO), and what actionable insights you can apply to your own marketing strategy.

What We'll Cover

1. Traffic Analysis: How much traffic Tally.so gets from AI search engines and what this means
2. Keyword & Prompt Breakdown: What people are actually asking ChatGPT that leads them to Tally
3. Source Analysis: How Tally appears in LLM answers with a surprisingly simple strategy that you can copy
4. GEO Improvement Opportunities: Where Tally could further optimize their AI search presence

Let's dive in and decode this new frontier of digital marketing.

Part 1: Traffic Analysis - The Numbers Behind Tally's AI Search Success

According to data from SimilarWeb, Tally.so receives approximately 9 million visits worldwide. Of these visits, about 19% (roughly 2 million) come from referrals. And here's the kicker: 💡 ChatGPT alone accounts for 10% of all referral traffic to Tally.so This translates to ChatGPT sending over 3,000 leads to Tally every week. Not just visitors - qualified leads that convert. The implications are profound: if your brand isn't optimizing for AI search engines, you're potentially missing out on your most valuable traffic source.

Part 2: Keyword & Prompt Analysis - What People Ask ChatGPT About Form Builders

Understanding what prompts users type into ChatGPT is crucial for any GEO strategy. Unlike traditional SEO where we have tools like SEMrush or Ahrefs to analyze search volumes, the AI search landscape is still developing its analytics infrastructure. At Qwairy (the tool used for this analysis), we approach this by first examining traditional search queries and then transposing them into likely AI prompts. We did a full analysis on how Tally generated 1$M in revenue with ChatGPT:

💡 For a step-by-step guide on identifying prompts that users actually ask AI search engines, check out [How to Identify Prompts on AI Search Engines](https://www.qwairy.co/blog/how-to-identify-prompts-ai-search-engines).

Traditional Keywords People Search:

• Online web form builder

• Form creator online

• Free form maker online

We added those keywords on Qwairy to transform them into prompts:

How These Transform Into ChatGPT Prompts:

• "What is the best free form builder?"

• "Which no-code form builders allow for unlimited forms and submissions without hidden fees?"

• "What are the most user-friendly free form creators for non-technical users?"

• "Which free form creator is most recommended for startups and why?"

• "Which free form creator offers the best integration with popular productivity tools?"

When analyzing Tally's performance across these prompts, we found they have exceptional visibility for certain query categories:

This means when someone asks ChatGPT about free form creators, Tally.so appears in the answer 70% of the time - an impressive achievement in the AI search landscape.

Part 3: Source Analysis - How Tally Appears in LLM Answers

There are two primary ways to appear in LLM responses:

1. As a brand mention - When the AI explicitly names your brand in its answer
2. As a source - When the AI uses content from your website to formulate its response

Tally.so has masterfully optimized for both, but their source strategy is particularly clever.

Tally's Content Strategy for LLM Visibility

The analysis revealed that Tally.so created specific content pieces designed to be picked up by LLMs:

1. Comparison content: "Best Free Online Form Builders in 2025" → Article Here
2. Alternative content: "Jotform Alternatives" (targeting competitor keywords) → Article here

These pages are among the most frequently cited sources when ChatGPT answers questions about form builders. (data from Qwairy) By creating comprehensive, well-structured content that directly addresses common user queries, Tally has positioned themselves as an authoritative source that LLMs naturally reference. 💡 Key Insight: LLMs prefer to cite content that itself appears to be objective and comprehensive, even when that content comes from a brand discussing its own category.

💡 🔍 Want to know which exact sources ChatGPT uses in its answers? This [deep-dive on how to know ChatGPT's sources](https://www.qwairy.co/blog/how-to-know-chatgpt-sources) breaks it all down.

Part 4: GEO Improvement Opportunities for Tally

Despite their success, the analysis identified several opportunities for Tally to further improve their AI search visibility:

1. Google Forms Alternative Gap

While Tally performs exceptionally well for "free form creator" queries, they have lower visibility for "Google Forms alternatives" prompts. Their competitor Jotform currently dominates this category.

2. Missing from Key Sources

The audit identified specific high-authority sources that ChatGPT frequently references but where Tally is either:

• Not mentioned at all

• Mentioned but positioned last in the list

For example:

• On Zapier's form builder comparison, Tally appears last

• On G2's form builder guide, Tally isn't mentioned

• On Captera's "Google Form Alternatives" page, Tally isn't positioned as a direct alternative

This is what you discover on the backlink opportunity tab on Qwairy:

3. Opportunity for Source Diversification

By securing mentions on these key third-party sites, Tally could significantly increase their already impressive AI search visibility.

How to Apply These Insights to Your Brand (Actionable GEO Playbook)

Tally's success with AI search is no accident. It's the result of a deliberate content strategy focused on Generative Engine Optimization. Here's how to replicate it with a proven playbook that works across LLMs like ChatGPT, Perplexity, Gemini, and beyond.

💡 Here is a complete [guide on Generative Engine Optimisation](https://www.qwairy.co/guides/complete-guide-to-generative-engine-optimization) that you should read 😉

🎯 Step 1: Build the Two Most Powerful Content Types for GEO

If you're only going to create two types of content for AI search visibility, make it these:

1. "Best [Category] in [Year]" Content

This is your cornerstone for appearing as a source in LLM responses.

Why it works:

• LLMs are trained to respond to queries like "What is the best X?"

• They prefer structured, comparison-style content that offers objective recommendations

• This content naturally fits AI-generated listicles and is frequently cited when users ask for product suggestions

How to structure it:

• Title: Use a clear and time-sensitive title like "Best Free CRM Tools in 2025"

• Introduction: Set the context. Who is this list for? What problems does it solve?

• Main List:

  • Use bullet points like:

• [Tool 1] – Best for small teams

• [Tool 2] – Best value for money

• [Tool 3] – Best integration with Notion

    ```
  

  • Keep each bullet under 3 sentences
  • Use phrases like:
    • "It is recommended for…"
    • "A good option if you need…"
    • "Consider this tool if…"

This gives LLMs clear indication about all the products in your niche.

• Add a recommendation section:

  • Help the reader choose based on needs: "If you're a startup → choose X. If you're a freelancer → choose Y."

• Add a table section to summarise everything like Tally at the beginning of your content.
  

🧠 To understand why comparison and alternative content is so effective, read our guide on how to rank on ChatGPT.

2. "Alternative to [Competitor]" Content

This format targets commercial intent queries like "What are alternatives to Google Forms?" - a high-converting category.

Why it works:

• LLMs love showing diverse options, especially when users are evaluating or switching products

• Targeting branded keywords lets you hijack competitor search volume

• It's easier to rank for "alternatives" than for head terms like "best CRM"

How to structure it:

• Title: Use formats like:

  • "Top 5 Alternatives to [Competitor] in 2025"
  • "Best [Competitor] Alternatives You Should Try This Year"

• Intro: Explain why people are looking for alternatives (e.g. pricing, UX issues)

• Comparison list:

  • Feature your product in the top 3
  • Present pros and cons of each option objectively
  • Use a table if possible (LLMs love structured formats)

🔍 Step 2: Identify the Prompts Users Are Asking the LLMs

You can't optimize for LLMs if you don't know what people are asking.

• Use tools like Qwairy to discover prompts people are asking ChatGPT, Perplexity, and Gemini.
  

• Look for:

  • "Best [category]" prompts
  • "Alternatives to [brand]"
  • "What's the easiest tool to…" questions

• Bonus methods:

  • Use regex filters in Google Search Console to extract question-like queries (e.g. ^(what|which|best|how|why)\s)
  • Check Copilot and Perplexity suggestions for additional question variants
  • Add these to your content targeting plan

🌐 Step 3: Get Cited by LLM-Referenced Sources

LLMs frequently rely on third-party sources to generate responses. You need to be mentioned in those sources - not just on your own site.

How to do it:

• Use Qwairy to identify the most cited websites in your space (e.g., G2, Zapier, Capterra, Reddit, Wikipedia)

• Build a backlink outreach strategy targeting:

  • "Top 10" listicles
  • Software review blogs
  • Product directories

• Refresh and optimize your G2, Capterra, and AppSumo profiles

• Run PR campaigns targeting blogs or news outlets often quoted by AI (e.g. Le Monde, TechCrunch, Medium)

🧠 Step 4: Speak Like an LLM

Want LLMs to quote your content? Write like them. Based on analysis of over 32,000 AI-generated answers, here's what we know:

• Keep sentences short

• Use an impersonal tone (e.g., "It is recommended")

• Prefer structured bullet points with "–"

• Use conditional and incentive phrasing: "You might consider…", "Try using…"

💡 You can read our comprehensive study on [32 961 queries on ChatGPT and other LLMs](https://www.qwairy.co/blog/32961-queries-llm-study) to learn more about how LLMs talk.

📈 Step 5: Track, Test, and Iterate

GEO is evolving fast. What works today might need tweaking tomorrow.

• Track your brand visibility weekly across LLMs using Qwairy

• Monitor which sources cite your brand

• A/B test content formats: long vs short, lists vs tables, FAQs vs how-tos

Final Words: Replicate Tally's GEO Flywheel

Here's the simplified version of the strategy:

1. Discover the prompts your audience asks LLMs
2. Create "Best X" and "Alternative to" content tailored to those prompts
3. Secure citations on external sources already used by LLMs
4. Write like an LLM to increase pickup
5. Track your visibility and iterate weekly

Start now - this space is still early, and visibility compounds over time. The brands that move fast will dominate AI search tomorrow. Finally here is a video if you want to use Qwairy to find hidden AI Opportunities:

FAQ: Generative Engine Optimization

What is GEO (Generative Engine Optimization)?
GEO is the practice of optimizing your digital presence to appear in AI-generated responses from tools like ChatGPT, Gemini, and Perplexity.

How is GEO different from SEO?
While SEO focuses on ranking in search engine results pages, GEO focuses on being mentioned directly in AI-generated answers. The strategies overlap but have distinct differences in content structure, authority signals, and measurement.

How can I track my brand's visibility in AI search?
Tools like Qwairy allow you to monitor your brand mentions across multiple LLMs, analyze which sources are being used, and identify optimization opportunities.

What types of content work best for GEO?
Comprehensive comparison content, alternative analyses, and objective guides tend to perform best. Content should be well-structured with clear headings, lists, and concise explanations.

Is it too late to start with GEO?
Not at all. The field is still emerging, and many brands haven't begun optimizing for AI search. This presents a significant opportunity for early adopters to establish dominance in this channel.