evolvsalescopilot_BlogImagery (1920 x 1080 px)

evolv’s Sales Copilot is a blueprint for how AI can increase productivity and empower teams

by Trent Foley

At evolv, we’ve long helped clients transform their business with cutting-edge technology. But as our own Brian Maastricht likes to say, “We drink our own Champagne,” meaning we also actively use our products and services in our daily operations.

It’s an ethos that sparked the creation of evolv’s Sales Copilot, a purpose-built AI assistant designed to revolutionize how our sales team accesses and uses information.

The Problem: Too Much Data, Too Little Insight

evolv’s sales reps were spending valuable time hunting down content — digging through SharePoint folders, pinging colleagues, and navigating HubSpot records — just to prep for meetings or answer client questions. The process was slow, fragmented, and frustrating. Our sellers were swimming in data, but thirsty for insights.

This inefficiency wasn’t just annoying, it was also costly. Deals were delayed or lost, and reps often reinvented the wheel due to lack of trust in available resources. The absence of a system to revisit closed-lost deals meant missed re-engagement opportunities and millions in potential revenue left on the table.

The Solution: AI-Powered Sales Copilot

Interface of evolv’s Sales Copilot.

Enter the Sales Copilot, built during our recent 2025 Craftathon by Team Swipe Right on Data. The goal? Deliver fast, contextual answers and insights on demand, like having your best enablement manager and data analyst on call 24/7.

Sales Copilot integrates data from HubSpot and SharePoint into Snowflake, using Openflow to ingest incremental content changes and Snowflake Cortex to power its intelligence. It combines:

  • Cortex Analyst for semantic SQL querying
  • Cortex Search for indexing unstructured documents and CRM notes
  • Cortex Agents to orchestrate responses using the MEDDPICC sales methodology

This architecture ensures that every answer is grounded in real, internal data. No external API calls, no rabbit holes.

Under the Hood: Architecture and Design

The solution architecture is deceptively simple for users, but deeply sophisticated behind the scenes.

One of the most significant accelerators in building the Sales Copilot was Snowflake’s prebuilt OpenFlow connectors. These connectors drastically reduced the data engineering workload by eliminating the need to manually handle incremental data loads and paginated API responses, tasks that typically require weeks of engineering effort. With OpenFlow, SharePoint documents and HubSpot CRM data were seamlessly ingested into Snowflake, flattened into structured formats, and made immediately available for AI indexing and querying. This out-of-the-box capability allowed the team to focus on designing the Copilot’s semantic view rather than wrestling with backend plumbing, enabling a fully operationalized solution in under a week.

One key lesson from the build? Keep the agent focused. Limit the number of tools made available to the agent and keep the Semantic View simple. This required thoughtful data modeling and engineering but paid off in accuracy and usability.

Real Impact: From Frustration to Empowerment

Sales reps have been transformed by Sales Copilot. No longer reliant on others for content, they can now generate strategies and push deals forward independently. One seller even exclaimed, “This is sick,” after seeing it in action.

To maintain intuitiveness, the team plans to scale by creating persona-specific agents rather than overloading a single agent. This modular approach ensures clarity and reduces error-prone responses.

The roadmap is ambitious and exciting:

  • Phase 1: Finalize testing and launch to the full sales team
  • Phase 2: Expand data inputs, add smart features like re-engagement prompts and coaching
  • Phase 3: Create additional agents for other teams and personas

The ultimate vision is to use AI to make every part of the organization smarter and faster. The approach of identifying a manual process with high potential for efficiency gain, breaking it down into roles, goals, and tasks for AI agents to coordinate on is truly exciting. The possibilities are endless.

Early Wins and Future Potential

Initial projections estimate a 5x productivity gain for meeting prep and next-best-action analysis. KPIs like weekly usage and deal revival rates will guide success metrics moving forward.

The Sales Copilot is a blueprint for transformation. By combining internal data with Snowflake’s AI capabilities, evolv has built a solution that empowers teams, accelerates deals, and sets the stage for client-facing innovation. We’re not just talking about AI — we’re living it.

Reach out to the evolv team.


 

Building on 20 years of experience as a technology leader and consultant, Trent has delivered innovative and high-impact solutions in the finance, healthcare, and defense industries.  He brings a cloud-first approach to designing, building, and deploying enterprise-scale solutions.  He is passionate about helping clients conceptualize and deliver high-quality solutions with modern practices and architectures.  Trent’s expertise includes leading application development efforts, application modernization, micro-service architecture, API development and integration, and data modeling and analytics. He is on a mission to build the dream team of technology consultants, pursuing creative and innovative techniques to attract, develop, and retain the top tier of like-minded professionals that relentlessly pursue excellence.