G2

Abdallah S.

Digital Marketing Analyst

Small-Business (50 or fewer emp.)

11/9/2026

"The Missing Web Layer for My AI Workflows"

4.5/5

What do you like best about Firecrawl?

What I like most is how reliably Firecrawl turns messy web pages into clean, structured markdown that AI tools can actually use. I run a web development and design business alongside several side projects, and I use Firecrawl mainly through its MCP connector inside Claude. That integration is the standout: I can ask an AI agent to search the web, scrape a client's or competitor's site, or pull product and pricing details, and it just works without me writing scraper code or fighting with JavaScript-heavy pages.

The search plus scrape combination saves me hours on research for client site audits, market research for my online store and books, and gathering content before redesigns. JSON extraction with a schema is great when I need specific fields instead of a whole page, and the output is consistent enough to drop straight into automations. Performance is fast, setup took minutes, and the documentation is clear. For anyone building AI agents or automation pipelines that depend on web data, it removes a huge amount of friction. Review collected by and hosted on G2.com.

What do you dislike about Firecrawl?

My main gripe is cost predictability. Credits can go faster than expected on bigger jobs, especially crawls of large sites or when you combine search, scraping and structured extraction, and it is not always obvious up front how many credits a task will use. A clearer cost estimate before running a job would help a lot. Occasionally a heavily protected site returns incomplete content and needs a retry or a different setting. There are also a lot of tools and options now, so new users may need some time to figure out which one fits their task. None of these are deal breakers, and the value I get easily outweighs them. Review collected by and hosted on G2.com.

What problems is Firecrawl solving and how is that benefiting you?

Before Firecrawl, gathering web data meant copying and pasting from pages by hand or writing one-off scrapers that broke whenever a site changed. With a small team juggling client websites, an online store, published books and a few SaaS products, that research time added up fast.

Now Firecrawl is the web layer for my AI workflows. Through the MCP connector, Claude can search, scrape and extract structured data on demand, so tasks like auditing a client's existing site before a redesign, comparing competitor pricing and positioning, or collecting reference content for a project take minutes instead of hours. Because the output is clean markdown or JSON, I can feed it straight into automations and AI prompts without cleanup. The result is faster turnaround for clients, better-informed decisions on my own projects, and no scraper code to maintain. I estimate it saves me several hours every week. Review collected by and hosted on G2.com.