Skip to main content

Documentation Index

Fetch the complete documentation index at: https://help.daylit.com/llms.txt

Use this file to discover all available pages before exploring further.

The voice agent gives you a hands-free, conversational way to query your AR portfolio. Instead of navigating reports or building filters, you speak a question and the agent responds in real time using your live data — customers, invoices, aging buckets, payment history, and AI-generated insights. It’s built for moments when you need a quick answer while multitasking, or when typing isn’t practical.

​
What the voice agent can access

When you start a voice session, the agent has full visibility into your AR data, including:
  • Customers — names, open balances, overdue amounts, risk signals, and communication history
  • Invoices — invoice numbers, amounts, due dates, aging status, and payment predictions
  • Aging totals — breakdowns by bucket (current, 1–30, 31–60, 61–90, 90+ days)
  • AI insights — conversation summaries, risk signals, payment date predictions, and strategy plans

​
Example questions you can ask

  • “What’s my total AR balance right now?”
  • “How much is overdue across all customers?”
  • “Give me a summary of my aging buckets.”
  • “Which aging bucket has the most exposure?”
  • “Which customers are most overdue?”
  • “Who should I call today?”
  • “Has Acme Corp responded to our last outreach?”
  • “What’s the risk signal on Northstar Industries?”
  • “Which customers have a ‘ghosting’ flag?”
  • “What’s the status of invoice number 1234?”
  • “Which invoices are more than 60 days past due?”
  • “Show me all disputed invoices.”
  • “When does the AI predict invoice 5678 will be paid?”
  • “Which accounts need follow-up this week?”
  • “Are there any promises to pay expiring soon?”
  • “What action does the AI recommend for Riverstone LLC?”
  • “Who has gone quiet despite having an open balance?”

​
Before you start

The voice agent requires microphone access in your browser. When you start a session for the first time, your browser will prompt you to allow microphone permissions. You must grant access for the voice agent to work.
Make sure you’re in a quiet environment for the best transcription accuracy. The agent performs best when background noise is minimal.

​
How to start a voice session

1

Open the dashboard

Log in to Daylit and navigate to your AR dashboard.
2

Open the voice agent

Click the microphone icon in the top navigation bar or the floating action button on the dashboard.
3

Grant microphone access

If prompted, allow your browser to use your microphone. This is required for voice input.
4

Ask your question

Speak your question naturally. The agent processes your speech, queries your AR data, and responds within a few seconds.
5

Continue the conversation

Ask follow-up questions. The agent retains context across turns in a session, so you can refine or dig deeper without restating background.

​
Tips for getting good results

  • Be specific. “What’s the overdue balance for Meridian Tech?” works better than “What about that one customer?”
  • Ask one question at a time. The agent handles focused questions more reliably than multi-part requests in a single turn.
  • Use invoice numbers when you have them. Referencing a specific invoice number (e.g., “invoice 4491”) pulls exact data rather than searching by description.
  • Name the customer. Using the customer name you see in Daylit helps the agent locate the right account without ambiguity.

​
Session transcription

Every voice session is transcribed automatically. The transcript is saved to the call record linked to the customer, so your team can review what was discussed or use it as reference when following up. Transcripts appear in the customer’s communication history alongside emails and manual notes.
You can review or annotate a voice session transcript from the customer detail page after the call ends.