Test before going live

Why test first

The test runner simulates a real message through your entire pipeline without actually sending anything. No SMS goes out, no CRM gets updated, no tasks are created. It's completely safe and free: test runs don't count toward your signal quota.

Testing catches configuration mistakes before they affect real leads.

Run your first test

1. Go to the Test Runner in your dashboard
2. Enter a message: paste or type a realistic customer message
3. Set contact details (optional): name, phone, email. These fill in template variables.
4. Choose a channel: SMS or email. This determines which templates match.
5. Click Run Test

Read the results

Results appear in pipeline order:

Classification (Pro and Enterprise only)

• Intent: what the customer wants (booking request, complaint, general inquiry)
• Urgency: how time-sensitive (low, medium, high, critical)
• Sentiment: emotional tone (positive, neutral, negative)
• If AI isn't configured, classification is skipped and a note explains why.

Lead score (Pro and Enterprise only)

• Score: 0–100 with a color-coded bar
• Tier: hot (red), warm (orange), cool (blue), cold (gray)
• Recommended actions based on the tier

Matched rules

• Every rule that would fire, with a green checkmark
• The actions each rule would trigger, shown as badges
• If no rules match, you need to add or adjust your rules

Generated reply (if AI reply generation is enabled)

• The actual text that would be sent
• Whether it requires human approval before sending
• The channel (SMS or email)

Signal Guard evaluation (if AI is enabled)

• Whether the reply passed, was flagged for review, escalated, or blocked
• Which violations were detected (if any)

Actions triggered

• Summary of everything that would execute
• Actions show as "success" or "dry-run" (dry-run means the adapter isn't connected)

What to test

Run these scenarios before going live:

After-hours inquiry: run the test outside your configured business hours:

"Do you have any availability this week for a consultation?"

Missed call: set the source to missed_call:

"Hi, I missed your call. Can you call me back?"

Hot lead with budget: test lead scoring:

"I need a kitchen remodel done by March. Budget is $40k. I'm the homeowner and ready to sign a contract this week."

Compliance-sensitive message: test Signal Guard:

"What medication should I take for my skin condition? I've been having allergic reactions."

Booking request: test your booking template:

"I'd like to schedule an appointment for next Tuesday at 2pm if possible."

Checklist before going live

Before enabling live responses, verify:

• Business hours are correct (timezone and schedule)
• At least one channel is connected (check Integrations)
• After-hours rules fire correctly when tested outside hours
• Templates render with the right variables; check the reply text
• Signal Guard catches sensitive content (test with a compliance scenario)
• Lead scoring tiers feel right (test with high-intent and low-intent messages)
• AI replies match your brand voice (if using AI)
• No unexpected rules match: check for rules with no conditions that might fire on everything

Tips

• Test after every configuration change. Changed a rule? Adjusted scoring weights? Updated brand voice? Run a test.
• Use real customer messages. Copy-paste from your inbox for the most realistic results.
• Test without AI too. Even if you're on Pro, verify that your rules work on channel/source/business hours alone. This is your fallback if you hit your AI signal limit.
• Compare before and after. Run the same message before and after a change to see the impact.