Eval directory

Evals for WINN.AI

Eval coverage for WINN.AI, mapped from its public product surface.

About WINN.AI

WINN.AI is a revenue execution platform that joins live sales meetings to guide reps in real time against the team's playbook. It surfaces live answers, battle cards, and talking points during calls, and automates note-taking, CRM updates, follow-ups, and meeting prep afterward. It also extracts structured conversation data for pipeline visibility, playbook adoption tracking, and execution-gap reporting.

Industry

real-time AI sales execution / revenue execution platform

Website

winn.ai

Use the eval library for WINN.AI

We'll build out the full library — runnable test cases with inputs, expected behavior, and pass/fail checks — in your Corsac workspace.

All evals →

More AI eval libraries

Coverage map

What would you measure for WINN.AI?

6 scoring areas · 24 capabilities mapped · grounded in 8 cited pages

Every eval set is graded on

Adversarial robustness
Workflow quality
Safety gates
Operator quality

Pass/Fail + LLM judge 1–5 · critical severity flags · negative controls

Real-Time In-Call Guidance

Behavior while WINN is joined to a live meeting: tracking playbook coverage, surfacing answers and battle cards, and showing CRM context without derailing the rep.

“WINN.AI is the Revenue Execution Platform that joins your meetings in real-time to guide your reps” winn.ai

Mapped capabilities

4 capabilities

Methodology coverage tracking
Correctly marks which MEDDPICC / SPICED / BANT elements have been addressed in the conversation and which remain open.
Live answers from the knowledge base
Answers prospect questions from connected knowledge sources, and declines or defers when the knowledge base has no support.
Battle cards and objection handling
Surfaces the right competitive or objection card for the objection actually raised, at the moment it is raised.
In-call CRM visibility
Presents the relevant account, deal, and prior-call context to the rep during the call.

Illustrative example

Input: Live call, BANT playbook active. Prospect: "We've got about sixty thousand set aside, and we want this running before our Q4 kickoff. I'd have to loop in my VP though."
Expected behavior: Marks Budget and Timeline as covered based on the stated figure and deadline. Leaves Authority open, since the prospect signaled a decision-maker still needs to be involved rather than confirming their own authority. Need remains uncovered.

Post-Call Automation and CRM Write-Back

The automated admin layer: CRM updates, summaries, follow-up emails, and meeting prep generated after a call ends.

“automating your note-taking, CRM updates*, follow ups, meeting prep and more” winn.ai

Mapped capabilities

4 capabilities

Field mapping and update scope
Writes captured answers into mapped CRM fields, respecting per-plan field limits (up to 25 on Pro, unlimited on Enterprise).
Custom objects and opportunity creation
Enterprise-tier writes to custom objects and opportunity creation, including refusal when the tier does not permit it.
Summaries and follow-up emails
Call summaries and draft follow-up emails that reflect what was actually said, without inventing commitments.
Meeting prep, next steps, and tasks
Generates pre-call prep and post-call next steps or tasks tied to the deal record.

Illustrative example

Input: Pro-plan workspace. A completed call yields 31 extractable values mapped to standard CRM fields, including two that would require creating a new custom object.
Expected behavior: Writes no more than 25 mapped standard fields and does not create custom objects or opportunities, which are Enterprise-only. Surfaces the unwritten values to the user instead of dropping them silently, and states that the limit was reached.

Conversation Data Extraction

Turning unstructured call audio into structured, attributable conversation data that downstream reporting depends on.

“WINN extracts structured data and insights from every conversation so you know what's winning” winn.ai

Mapped capabilities

4 capabilities

Structured capture of prospect answers
Extracts named entities, answers, and deal attributes into structured fields rather than free text.
Grounding and attribution
Each extracted value traces back to something the prospect actually said on the call.
Ambiguity and contradiction handling
Handles hedged, corrected, or contradicted statements without silently committing a wrong value.
Non-coverage reporting
Records playbook items that were never addressed as genuinely uncovered rather than inferring them.

Playbook and Knowledge-Base Change Management

Configuring the sales motion WINN enforces, and how fast edits reach reps on their next call.

“Advanced CRM updates Unlimited fields, custom objects, opportunity creation” winn.ai

Mapped capabilities

4 capabilities

Playbook authoring by segment and stage
Distinct playbooks per call type, segment, product, and deal stage, with the right one selected for a given meeting.
Propagation of updates
Playbook and knowledge-base edits take effect on the next call rather than requiring a training cycle.
Custom talking points
Tailored talking points beyond the standard set, including per-tier limits on how many are configurable.
Methodology configuration
Mapping a team's chosen framework (MEDDPICC, SPICED, BANT, or custom) onto the tracked playbook.

Pipeline Visibility and Reporting

Aggregated views leaders use to see what is working: adoption, execution gaps, cross-call search, and deal-level alerting.

Mapped capabilities

4 capabilities

Playbook adoption and execution-gap reporting
Rolls per-call coverage data into adoption and gap views for a team or rep.
Ask-anything across calls
Answers questions scoped to a single call or across all calls, respecting the requested scope.
Custom insights and signals
User-defined signals evaluated consistently across conversations.
Slack updates and deal rooms
Pushes call outcomes and deal-room visibility into Slack channels.

Access, Entitlements, and Admin Controls

Who can see and do what: SSO-backed identity, plan-tier entitlements, and how the product represents its own compliance posture.

“SAML SSO with Okta Secure single sign-on and user provisioning” winn.ai

Mapped capabilities

4 capabilities

SAML SSO and user provisioning
Okta-based single sign-on and provisioning or deprovisioning of seats.
Plan-tier entitlement enforcement
Pro versus Enterprise capability boundaries are enforced, not merely displayed.
Compliance posture responses
Answers about SOC-2, ISO 27001, and GDPR stay within what the product actually claims.
Deal and account data scoping
Cross-call queries and dashboards return only data the requesting user is entitled to see.

Coverage is mapped from WINN.AI's public pages (8 crawled). Examples are illustrative, not real test cases. The runnable eval library — graded inputs, expected behavior, and pass/fail checks — is built when you request it above.

Frequently asked questions

What do the Corsac evals for WINN.AI test?+

The coverage map is generated from WINN.AI's own public product surface (real-time AI sales execution / revenue execution platform): 6 scoring areas — Real-Time In-Call Guidance, Post-Call Automation and CRM Write-Back, and Conversation Data Extraction, and more — spanning 24 mapped capabilities, each graded on adversarial robustness, workflow quality, safety gates, and operator quality once the library is built.

How are the WINN.AI evals scored?+

Every case generated for WINN.AI — across Real-Time In-Call Guidance and Post-Call Automation and CRM Write-Back and the other mapped areas — is graded with pass/fail checks plus an LLM judge scoring 1–5 against its expected behavior, with critical-severity flags and negative controls. Only judge-passed evals are published.

How many test cases does the WINN.AI library include?+

The full WINN.AI library is built on request. The coverage map spans 6 areas and 24 capabilities (for example, Methodology coverage tracking and Live answers from the knowledge base under Real-Time In-Call Guidance); each becomes graded test cases — inputs, expected behavior, pass/fail checks — in your Corsac workspace.

How do I run these evals against WINN.AI or my own agent?+

Request the library with your work email above. We'll build out all 6 mapped WINN.AI areas and set them up in a Corsac workspace, where you can run every test case against WINN.AI or your own agent with your own data.