How 11x Processes 11 Million Requests to Automate Prospect Research for AI Outbound

The Problem: AI Outbound Needs Real Company Context

When you set out to automate sales outreach with AI, the first wall you hit is usually this: how do you make the AI actually “understand” a company? Not just the homepage, but what it sells, the problems it solves, its customer proof, and even the tools in its tech stack. Without that depth, any personalized message is just a template with a company name swapped in.

11x, an AI growth platform used by revenue teams at companies like Checkr, Rho, and Xerox, has built a system that solves this problem at scale. The platform automates GTM workflows across the full outbound cycle—from identifying and researching prospects to personalizing outreach and booking meetings. At the core of that research layer is Firecrawl, a web scraping and search API that turns raw web pages into structured markdown and data. To date, 11x has processed more than 11 million requests across 175,000 unique domains with Firecrawl.

This article breaks down how 11x uses Firecrawl to build company research briefs, why that research is the foundation of effective AI outbound, and what product builders can learn from the architecture. We’ll also look at the trade-offs and limitations of relying on public web data.

How Firecrawl Powers 11x’s Research Briefs

11x’s outbound messages need to feel relevant to each prospect. That means understanding what each company sells, the problems it solves, its customer proof, and its brand identity. Firecrawl gathers and structures all of that information into research briefs that 11x uses to tailor its outreach.

Most of 11x’s Firecrawl usage runs through two endpoints:

• /crawl: Goes deep on an identified company’s website by following its sitemap and internal links. This builds a complete profile of the site, capturing pages that describe products, solutions, pricing, and more.
• /search: Complements the site-level research by finding sources elsewhere on the web. This includes customer case studies, partnership announcements, and company facts such as headcount or headquarters location.

Together, /crawl provides depth and /search provides breadth. Firecrawl handles the heavy lifting: it finds relevant sources, follows sitemaps, renders JavaScript, and extracts page content. The results return as clean markdown and structured data, giving 11x one research workflow that works for any site on the web—no custom parsers needed.

As Jeson Patel, CTO at 11x, puts it: “Firecrawl gathers and structures information from across the web to give us a detailed picture of any company, from its position in the market down to the tools in its tech stack. That depth of research is what makes our outbound so context-rich.”

Why Research Depth Determines Outbound Quality

The quality of AI-generated outreach is directly tied to the richness of the input context. If you feed the AI only a homepage URL, you get a shallow message. But if you give it a structured brief that includes the company’s market position, customer proof, and tech stack, the AI can craft messages that feel genuinely tailored.

11x’s approach shows that research isn’t just a nice-to-have—it’s the foundation. The platform uses Firecrawl to build these briefs automatically, so every prospect gets the same level of deep research without manual effort. This is what makes outbound “context-rich,” as Patel notes.

For product builders, the lesson is clear: when designing AI workflows that generate personalized content, invest in the data pipeline first. A sophisticated language model can’t compensate for thin input. Structured research briefs, like the ones 11x builds, directly improve the relevance of the output.

Scaling Research: 11 Million Requests and Counting

The sheer volume—11 million requests across 175,000 unique domains—highlights how labor-intensive prospect research is in sales. If you tried to do that manually, even a few thousand companies would overwhelm a team. Automating repetitive GTM work like prospect research isn’t about replacing salespeople; it’s about giving them more time for the human side of sales: building trust and closing deals.

For product builders, this scale also demonstrates the importance of infrastructure reliability. Processing 11 million requests means Firecrawl’s uptime, speed, and cost control are all critical. When choosing a tool for your own AI workflow, you need to consider not just features but whether it can handle your growth.

Practical Implementation: Building a Research Brief Pipeline

If you’re building an AI outbound tool or a similar product, you can start by replicating 11x’s pattern. Here’s a high-level workflow:

1. Identify a target company from your CRM or lead list.
2. Run /crawl on the company’s website to build a site profile. Firecrawl will follow sitemaps and internal links, render JavaScript, and return clean markdown.
3. Run /search for supplementary facts: case studies, partnerships, headcount, headquarters, and tech stack mentions.
4. Combine the results into a structured research brief (e.g., JSON with fields like products, customers, tech_stack).
5. Feed the brief to your AI model along with a prompt to generate personalized outreach messages.

Here’s a simplified example of how you might call the Firecrawl API (using a hypothetical SDK):

import FirecrawlApp from '@firecrawl/sdk';

const app = new FirecrawlApp({ apiKey: 'your-api-key' });

// Crawl the company website
const crawlResult = await app.crawlUrl('https://example.com', {
  limit: 50,
  scrapeOptions: { formats: ['markdown'] }
});

// Search for supplementary facts
const searchResult = await app.search('Example Corp case study', {
  limit: 5
});

// Combine into a research brief
const researchBrief = {
  website: crawlResult.data.map(page => page.markdown).join('\n'),
  externalSources: searchResult.data.map(item => item.markdown)
};

This is a simplified illustration, but it captures the core idea: automate the research, structure the output, and let the AI do the personalization.

Limitations and Trade-offs

While Firecrawl provides a powerful way to gather public web data, there are important limitations to keep in mind.

First, the data is limited to what’s publicly available on the web. Firecrawl cannot access non-public information such as internal decisions, unpublished product plans, or private customer data. If your outbound strategy depends on such insights, you’ll need to supplement with other data sources.

Second, relying on a single vendor like Firecrawl introduces dependency risk. If the service experiences downtime or changes its pricing or policies, your workflow could be affected. It’s wise to have a fallback or abstraction layer.

Third, web data can be noisy or outdated. Not every page is accurate, and websites change. You may need to add validation steps or periodic re-crawls to keep your research briefs fresh.

Finally, there’s the question of cost and rate limits. Processing millions of requests isn’t free, and you’ll need to monitor usage to keep costs under control. Firecrawl offers various plans, but you should evaluate whether the pricing scales with your expected volume.

Key Takeaways for Product Builders

If you’re building AI-powered outbound or any product that relies on personalized content, here are the three most important lessons from 11x’s approach:

1. Research quality determines output quality. Invest in a robust data pipeline that gathers rich, structured context about each prospect. The better the input, the better the AI’s output.
2. Balance depth and breadth. Crawling the company website gives you depth, but you’ll miss customer case studies or partnership news. Searching the web gives you breadth but may lack site-specific details. Use both to build a complete picture.
3. Plan for scale from day one. If your product succeeds, you’ll face millions of requests. Choose infrastructure that’s reliable, fast, and cost-effective, and design your workflow to handle growth.

Start small: pick a handful of prospects, build a research brief pipeline using Firecrawl or a similar tool, and measure how much it improves your AI’s output. The results might surprise you.

As 11x has shown, automating repetitive research tasks isn’t just about efficiency—it’s about enabling a level of personalization that would be impossible manually. By giving your AI the right context, you can make every outbound message feel like it was written for that specific company.

Sources

• Original source

AI-assisted summary compiled from the sources above, reviewed by a human before publishing.