ZoomInfo Launches GTM Benchmark to Measure AI Performance in Go-To-Market Work — ZoomInfo Technologies Inc.

Event summary

ZoomInfo released GTM Bench, a benchmark evaluating AI models on go-to-market tasks like building target lists and enriching records.
Version 1 covers 20 jobs, 4 systems, and 3 models, with methodology and grading rubrics published for scrutiny.
ZoomInfo's GTM.AI led the benchmark with a GTM Bench Index of 77, outperforming competitors like Apollo (47) and Exa (36).
GTM.AI completed 98% of the operator's work product and returned 478 verifiable records per 1,000 contacts.
The benchmark will be versioned and re-run on major model releases, with version 2 adding international coverage and multi-step workflows.

The big picture

ZoomInfo's GTM Benchmark addresses a critical gap in evaluating AI models for real-world go-to-market tasks, where data accuracy and grounding are paramount. This move underscores the growing importance of specialized benchmarks in the AI industry, particularly as companies seek to automate and optimize their revenue operations. The benchmark's versioned approach suggests a long-term strategy to maintain relevance as AI models evolve.

What we're watching

Benchmark Adoption: Whether other data providers and AI developers will submit their systems for evaluation, validating the benchmark's industry-wide relevance.
Performance Improvement: The pace at which ZoomInfo and competitors enhance their AI models to close performance gaps identified in GTM Bench.
Market Differentiation: How ZoomInfo leverages GTM Bench results to differentiate its offerings in a competitive B2B data and AI market.