Insights & Alerts

Pingd doesn't just answer questions — it proactively surfaces insights about your pipeline, accounts, and deals. Insights are generated by advisors, automated AI workflows that continuously analyze your data and deliver findings directly to your team via Slack.

How Insights Work

Insights are created through the Advisor pipeline. Here's the process:

1. An advisor runs on schedule — hourly, daily, or on a custom cron schedule you define
2. Data is fetched from your primary data source (Snowflake, Salesforce, PostgreSQL, etc.)
3. Rules filter the data to surface only what matters (e.g., deals with no activity in 14+ days)
4. AI analyzes each matching record using OpenAI's batch processing API, generating insights with a priority level, title, and detailed body
5. Insights are delivered as Slack DMs to the right people based on the OWNER_EMAIL (or equivalent) column in your data

Each insight includes a priority score (1–10), a title summarizing the finding, and a body with the full analysis and recommended actions.

Advisor-Generated Insights

Advisors are the engine behind insights. You can create advisors for any use case:

🔴 Deal Risk Monitoring

Build an advisor that flags deals showing warning signs:

• Deals stuck in the same stage too long
• Decreased engagement from key contacts
• Close dates pushed back repeatedly
• Champion gone silent

📈 Expansion Opportunities

Create an advisor that spots upsell potential:

• High product usage accounts approaching renewal
• Accounts with growing team size
• Positive engagement trends

🎯 Pipeline Health

Set up an advisor to give daily pipeline summaries:

• Deals expected to close this quarter
• Stage progression velocity
• Coverage ratio analysis

🛡️ Custom Analysis

Build advisors for any data-driven workflow:

• Monitor support ticket trends
• Track NPS scores
• Flag accounts with declining engagement
• Identify competitive threats based on activity patterns

To create an advisor, see Advisors.

Insight Delivery

Insights are delivered as Slack DMs to the people who need to see them. The recipient is determined by the notification delivery column you configure during advisor setup — typically the account owner's email address.

How Delivery Works

1. The advisor generates insights for each matching record
2. Pingd looks up the recipient's email from the delivery column (e.g., OWNER_EMAIL)
3. Pingd resolves the email to a Slack user via users.lookupByEmail
4. The insight is delivered as a DM to that Slack user

Important: The email addresses in your data must match actual Slack workspace member emails for delivery to work. If a recipient can't be found in Slack, the insight is stored but delivery is retried later.

Delivery Status

Each insight has a status:

• Pending — Generated, waiting for delivery
• Delivering — Currently being sent
• Delivered — Successfully sent to the recipient via Slack

Deal Risk Signals

Whether you build a dedicated advisor or ask your agent directly, Pingd can analyze these deal risk signals from your connected data:

• Stale deals — Opportunities with no activity for an extended period
• Engagement drops — Declining reply rates, meeting cancellations, longer response times
• Close date slippage — Deals with repeatedly pushed close dates
• Missing stakeholders — Deals without executive sponsor involvement
• Stage regression — Opportunities moving backward in the pipeline

Acting on Insights

When you receive an insight via Slack DM, you can:

1. Ask your agent for more detail — "Tell me more about the Acme risk"
2. Get recommendations — "What should I do about this deal?"
3. Take action — "Draft an email to re-engage Sarah Chen"
4. Review in dashboard — View all insights and their statuses in the Pingd dashboard

Configuring Alert Preferences

Priority Levels

Advisors assign priority scores (1–10) to each insight. You can configure your advisors to only generate insights above a certain priority threshold, ensuring your team only sees what truly matters.

Scheduling

Control when insights are generated by setting advisor schedules:

• Hourly — For time-sensitive monitoring (e.g., deal risk alerts)
• Daily — For summary-style insights (e.g., morning pipeline briefing)
• Custom cron — For specific timing needs (e.g., every Monday at 9 AM)

Insight Accuracy

Insights are AI-generated assessments based on your data. They're designed to surface patterns and anomalies that deserve attention — but always apply your own judgment. You know your deals and customers better than any AI.

The more structured and complete your data is, the better your insights will be. See Data Shaping for tips on structuring your data for optimal advisor analysis.

See also: Advisors · Configuring Your Agent · Scheduled Tasks