UserGems

UserGems sells software that positions itself as an "AI command center" for outbound sales and account-based marketing. The platform ingests first-, second- and third-party data and offers 30+ native signals — including job changes, contact-level intent, website visits, hiring, M&A and 10-K filings — alongside customer-supplied data such as CRM records, call transcripts, product usage, form submissions, event attendance and NPS scores. A research agent can be configured to find additional custom signals from public sources and to analyze call transcripts and CRM data for closed-lost insights.

On top of this data layer, the company's AI agent product, Gem-E, evaluates 600+ ICP criteria and signals and learns from a customer's closed-won history to produce account and contact scores that the company describes as transparent and editable. UserGems also derives ICP definitions, personas and value propositions from a customer's opportunity and customer records, maps accounts into buying stages with customizable weights and automatic decay, and identifies and enriches buying-group contacts, creating new records in the CRM.

Execution is handled by orchestration features that generate personalized outbound email, surface insights to reps through a Chrome extension inside their existing tools, sync high-intent contacts into LinkedIn, Google and Meta ad audiences, and report on TAM progression and pipeline generated. The product is designed to operate within existing stacks rather than as a standalone interface, integrating with Salesforce, HubSpot and Microsoft Dynamics CRMs, HubSpot and Marketo marketing automation, and Outreach, Salesloft and Gong Engage sales engagement tools. A model context protocol (MCP) integration lets users build and run GTM campaigns from Claude or ChatGPT.

Founding story

Christian Kletzl, an Austrian who was living in Chicago after an MBA at Northwestern, recruited his twin brother Stephan, then working for PwC in Europe, to move to the U.S. and start a company. Their first idea, ShelfFlip, an e-commerce marketplace for pre-owned goods, got them into Y Combinator's Winter 2015 batch but was scrapped during the program. They then built SmartHires, a referral network for startups within the same investment portfolio. Part of the SmartHires product tracked when a key contact moved from one company to another, and batchmates showed more interest in that capability than in SmartHires itself, which led to the pivot into UserGems.

Business model

B2B SaaS sold to revenue teams, layered on top of a customer's existing CRM, marketing automation, sales engagement and advertising tools. Deployments include dedicated Customer Success, RevOps and AI specialists for implementation, enablement and campaign recommendations. The company advertises a money-back guarantee: if a customer does not generate at least 1x revenue ROI (framed as $100K revenue on $100K spend), it offers a refund.

Software subscription tied to customer spend, with a stated refund guarantee if revenue generated does not reach at least 1x the amount spent.

Traction

Roughly 60 mid-market and enterprise customers reported in July 2021 and about 90 by October 2021. The website cites company-reported outcomes including median 47x pipeline ROI and 11x revenue ROI, Gem-E sequence reply rates of 6-20%, and customer results such as CaptivateIQ building $1.3M in pipeline in 10 weeks with 15 opportunities, Quickbase reporting 5x leads and 3x pipeline, and Docebo generating 54 enterprise opportunities in a quarter. A 2026 event drew 150+ B2B GTM leaders. These figures are vendor-published and not independently verified.

Latest developments

Recent company announcements include the UserGems MCP, which exposes the platform inside Claude and ChatGPT, unveiled at the Humans of GTM event in San Francisco (Jun 2026); a "Build vs. Buy" framework guide and related video featuring co-founder Christian Kletzl (Jul 2026); the Brains of GTM webinar series launched Mar 2026 with episodes featuring Snowflake and Cordial; and a Feb 2026 report that Gem-E achieves 11-20% reply rates by stacking signals across email sequences.