Support & Resources

Need help getting started or have questions about Agentforce Decision Lab? We're here to help.

📧

Email Support

Questions about installation, features, or security?

[email protected]

Response time: 24 hours

💼

LinkedIn

Connect with the creator and share feedback

Connect with Anvesh →

Direct messages welcome

Frequently Asked Questions

Q: Do I need an Agentforce license to use this tool?

A: No! That's the whole point. Agentforce Decision Lab lets you test your org's readiness BEFORE buying an Agentforce license. You just need a Salesforce org and 5 minutes.

Q: How does the Real AI Test work?

A: We randomly select 20 records from your chosen object and send them to Claude (Anthropic's AI, same engine as Agentforce) via our secure Cloudflare Workers proxy. We analyze the AI's responses and give you an accuracy score. Your data is never stored.

Q: Is my data secure?

A: Yes. We use Cloudflare Workers to keep your API keys secure (never in Salesforce). We automatically mask PII (SSNs, credit cards, emails, phone numbers). We never store your data—it's processed in real-time and discarded immediately. GDPR & SOC 2 compliant architecture.

Q: What objects are supported?

A: We officially support Case, Lead, and Opportunity out of the box. We also support custom objects—our tool automatically discovers fields and adapts. More standard objects coming in future updates.

Q: How long does an evaluation take?

A: About 60 seconds for Preview mode (estimates), 3-5 minutes for Real AI Test mode (tests 20 actual records). You can run as many evaluations as you want.

Q: What's the difference between Preview and Real AI Test?

A: Preview mode gives you instant estimates based on your org's data patterns. Real AI Test mode actually tests 20 records with Claude to measure real accuracy. Preview is good for quick checks; Real AI Test is what you show to executives.

Q: What does this cost?

A: Public Beta 1.0 is available now. We're focused on collecting feedback and refining the product. Future pricing will be transparent and fair when we're ready to announce it.

Q: Can I use this for multiple orgs?

A: Yes! Install the managed package in as many orgs as you need. Great for consultants and agencies evaluating multiple clients.

Q: I'm getting an error during installation. Help?

A: Common issues: (1) RecordTypes not enabled—we handle this gracefully, shouldn't cause errors. (2) Lightning Web Security not enabled—required setting. (3) Remote Site not added—required for AI testing. Check our Quick Start Guide for detailed setup instructions. Still stuck? Email [email protected]

Q: My readiness score is lower than expected. What now?

A: That's exactly what this tool is for! Look at your dimension breakdown and top 3 priority fixes. We give you effort estimates (hours/days/weeks) for each fix. Address the blockers, then re-run the evaluation to see your improved score.

Q: Can I download the evaluation results?

A: Yes! We generate two PDF reports: (1) Executive Summary for leadership, (2) Technical Deep Dive for architects. Both are downloadable directly from the app.

Q: Who built this?

A: Anvesh Chodavarapu, a Salesforce Architect with 11 years of experience and 12 Salesforce certifications. Built independently with zero vendor bias. Connect on LinkedIn.

Q: I have a feature request. Where do I submit it?

A: We love feedback! Email [email protected] or message Anvesh directly on LinkedIn. During beta, we're actively iterating based on user input.