The agent trust page says 49% found an agent reaching systems they didn't know it could, 71% name governance as the top barrier and 30% couldn't stop a harmful agent within minutes, but no source sits next to the numbers. Monte Carlo is creating a category ("agent trust") and its SDRs are building the inbound playbook from scratch: a CDO needs a way to see where they stand.
First action
Put the source and date beside the 49%, 71% and 30% figures today.
How to measure it
Meetings from completed assessments
- Now
- –
- Target
- 20% of completions
- By
- 60 days
Steps
Source the numbers
Add the survey, sample size and date next to each statistic, or replace any that can't be sourced.
Product Marketing 1 day
Write 15 questions
Three to four per layer, each answerable by a data or AI leader in a minute.
Product Marketing 2 days
Score and compare
A score per layer and the share of companies at each level.
Data Science 2 days
Build the form
HubSpot form with a results page and a PDF by email.
Marketing Operations 2 days
Route to SDRs
Each result to the inbound queue with the answers and a priority rule.
Strategic Sales Development Representative 2 hours
Use in outbound
AEs send the assessment to target accounts as a pre-call.
Strategic Account Executive, AMER 1 hour
Feed the BVA
Bring the answers into the Business Value Assessment template.
Sr. Sales Engineer (FDE, Pre-Sales) 2 hours
Publish the totals
After 200 responses, a short write-up of what companies reported.
Product Marketing 1 day
Why it works
Add the source and date to the three statistics now. Then publish an "Agent trust assessment": 15 questions across context, performance, behavior and output (can you list every agent in production? who can switch one off, and how fast? is the data each one retrieves monitored?). The result is a score per layer against everyone who took it, a short PDF, and an offer to review it with a sales engineer. Route every result to SDR triage with the answers attached.
An assessment turns a new category into a question the buyer answers about their own company, which is the start of the Business Value Assessment the Strategic AEs must co-build. It gives inbound SDRs a reason to call and a ranked queue ("cannot stop an agent within minutes" first).
Ready to use
ChecklistAgent trust self-assessment (sample)
Context [ ] We can list every agent running in production [ ] The data each agent retrieves is monitored for freshness and completeness Performance [ ] We track cost and latency per agent Behavior [ ] We would notice if an agent reached a system it shouldn't [ ] We can disable or roll back any agent within minutes Output [ ] We check agent answers against their source data [ ] A person reviews high-risk outputs before they reach customers
Risks and how to handle them
Feels like a lead trap
Show the score on the page before asking for an email for the PDF.
Self-reported answers
Label it a self-assessment and use it to start the conversation, not to prove anything.
Evidence
- Your website"What is agent trust?" page: four layers (context, performance, behavior, output) and three statistics, "49% have discovered an agent accessing systems they didn't know it could reach", "71% cite security and governance as the top barrier" and "30% could not disable or roll back a harmful agent within minutes", with no source shown next to them; it mentions an MCP server to "add gates into your own AI coding environment". Its structured data names the author "Ashley Rothrock" but carries Michael Segner's bio ("Head of Product Marketing ... co-authored the O'Reilly report ... with Barr Moses").Source
- HiringStrategic Sales Development Representative (Americas): "triage inbound leads against our best-fit customer profile, outbound to our most strategic prospects" and "help build the inbound playbook from scratch, working hand in hand with marketing"; uses AI tools daily to research accounts and personalize outreach.Source
- HiringStrategic Account Executive, AMER (California and New York/Boston): the Strategic team "owns our Global 2000 business: landing new logos and expanding within accounts like Pepsi, Fox, General Mills, and Amgen"; a 50-60-account territory built "through outbound, ABM, partner co-sell, and AI tools, holding 3-4x quota coverage"; reach CTOs, CDOs and VPs of Data/AI, "co-build Business Value Assessments with Sales Engineering", close "six- and seven-figure, multi-year deals". "This team is set to quadruple."Source
- Job postsCompany boilerplate: "Monte Carlo is the agent trust platform that unifies data and agent observability"; "Founded in 2019"; "We created the data observability category, and we're doing it again with agent trust". 10 open roles, all remote, 7 in sales plus a forward-deployed sales engineer.Source