G2 continues to expand structured content, AI evaluation, and data integrations across the platform. These updates focus on improving how buyers discover products, increasing trust in review data, and providing more actionable insights for evaluating vendors.
Q1 2026
New Features
Structured FAQs
Structured FAQ sections are now available on category pages, formatted using a Q&A schema. This improves how search engines and large language models interpret page content, increasing eligibility for rich results and AI-generated citations.
Guided discussion prompts
AI-generated prompts guide buyers in creating more detailed and structured discussions. This improves content depth and strengthens signals used by search engines and AI-generated answers.
Focused feature list
Feature lists are now identified from what reviewers write, replacing manual selection from predefined lists. This reduces effort during review submission and captures feature data using the reviewer’s own language.
Data and integrations
SaaS spend and pricing data
Expanded data coverage now provides deeper visibility into SaaS spend, pricing benchmarks, and discount trends. This helps customers analyze purchasing activity alongside G2 buyer behavior data.
Verified reviews with LinkedIn
LinkedIn identity verification is now supported during review submission. Verified reviews strengthen trust signals, improve moderation quality, and increase the reliability of data used in rankings and AI-generated outputs.
AI and platform updates
Competitive pulse
Competitive pulse provides a sales intelligence dashboard that combines CRM data with G2 buyer activity. It highlights deal prioritization signals, competitive presence, and indicators of churn risk.
G2 MCP integration for Claude
G2 buyer intent and review data is now available in Claude through the Model Context Protocol (MCP). This enables AI responses to reference verified first-party data for more accurate and actionable outputs.
Benchmark performance for AI agents
This update introduces standardized benchmarking for AI agents using simulated user queries and predefined scenarios.