Petra Labs

Petra Labs is a New York-based provider of answer engine optimization (AEO) software and services for enterprise brands. Its platform builds client-specific "prompt maps" from real user searches — drawn from opt-in AI search panel data, clickstream purchases, custom scraping of forums such as Reddit and LinkedIn processed with natural language processing, and traditional Google/SEO keyword data — then tracks how a brand performs against those prompts across major AI assistants including ChatGPT, Claude and Gemini.

Tracking modules cover visibility (share of queries in which a brand appears versus competitors), average rank, final citations and interim chain-of-thought citations surfaced during model reasoning, sentiment, share of voice, and performance against customer Key Purchase Criteria. Because large language models are probabilistic, Petra runs prompts repeatedly at scale until results stabilize before generating insights. A third layer is attribution: the company builds custom multi-touch attribution and marketing mix models that plug into a client's existing data, revenue and analytics stack, on the premise that most AI-referred traffic arrives without UTM tags and appears as direct, organic or dark social traffic and is therefore undercounted.

Alongside the software, Petra embeds a team of AEO specialists who handle platform onboarding, query selection and weighting, prompt tagging, custom reporting, AI search strategy, and execution across owned media (copywriting, technical site audits, programmatic AEO), earned media placements and social presence.

Founding story

Founded in 2025 by CEO Sami Akkawi, CTO Rishi Penmetcha and COO Ramie Khoury. Akkawi spent five years at Strategy& advising private equity firms on more than 50 acquisitions in marketing data, technology and services. Penmetcha built large-scale data systems and most recently worked on AI agents at Traba, where Khoury built and led the growth team focused on enterprise sales and operations. The founders' stated premise was that AEO could only be delivered credibly by understanding how AI search investments translate into revenue [4][5][7].

Business model

Combined software-and-services model: a proprietary measurement and attribution platform sold together with an embedded team of AEO experts that performs strategy and execution work on the client's behalf, positioned as a long-term partnership rather than a self-serve dashboard [0][2][3].

Revenue derives from engagements with enterprise brands covering platform access plus delivered AEO services, including custom-built software, attribution modeling and execution work [2][3].

Traction

Published a customer case study on an end-to-end AEO engagement with Zapier, listed at approximately 30 employees, and raised $5.2 million in seed funding in July 2026 [0][5][6].

Latest developments

Announced a $5.2 million seed round led by Work-Bench in July 2026, with Afore and Pathlight participating alongside strategic angels; funds are directed at expanding the measurement platform, embedded services and attribution capabilities. The company has also been publishing research on AI citation behavior, including reports on prompt wording effects on citations and on distinguishing real change from sampling noise in LLM search outputs [0][4][5][7].