Building SaaS Competitive Battlecards From G2 Review Data
B2B buyers leave a surprisingly structured trail on G2. Every review carries a star score, a set of sub-ratings, written pros and cons, and firmographics about who left it. Pull that at scale and you have the raw material for competitive battlecards, win/loss analysis, and switching signals. This post covers what is actually in a G2 review, why it is hard to collect directly, and how to turn it into a clean table.
What a G2 review actually contains
The value is in the structure, not just the star rating:
The overall score plus sub-ratings like ease of use, quality of support, and ease of setup.
Written pros and cons, and often what the reviewer was doing before, which is where switching signals live.
Reviewer firmographics: role, company size, and industry, so you can segment by ICP.
Sentiment and recommendation signals you can roll up per product or per competitor.
Aggregate those across a category and you can see exactly where a competitor is weak and where you win.
Why pulling it directly is hard
G2 sits behind a serious anti-bot layer, so a plain HTTP client gets blocked fast, and the review fields are spread across a rendered page that changes shape. It is solvable, but it is ongoing engineering, and it breaks whenever the site shifts. There is also no public G2 API for open collection of arbitrary product reviews, so scraping is the route.
The shortcut: a maintained scraper
Instead of building and babysitting that, you can run the G2 Reviews Scraper on Apify. Point it at any product or category and it returns the reviews as clean JSON, CSV, or Excel, including the sub-ratings, pros and cons, reviewer firmographics, and auto-ranked competitor battlecards built from the data. No login and no anti-bot tuning on your side.
To try it, create a free Apify account, open the actor, and click Try for free. The input is pre-filled with an example you can edit.
What you can do with it
Battlecards: roll up strengths and weaknesses per competitor from real customer language, not marketing copy.
Win/loss: mine the cons and switching mentions to see why buyers leave a rival.
ICP targeting: filter reviews by company size and role to find where a product actually lands.
RAG and AI: every row ships a self-contained markdown block, ready to drop into an LLM pipeline.
Pairing it with hiring signals
Competitive intelligence gets sharper when you add hiring data. The same team maintains an Indeed Jobs Scraper that pulls listings, salary ranges, and company data across 60+ countries, so you can line up what a competitor says on G2 against who they are hiring.
Start with a sample
Before running anything, you can open a free sample of real G2 review data to see the exact shape, and the field reference and code snippets are on GitHub. For a full walkthrough, see the complete G2 scraping guide.