Skip to main content

How to interpret the metrics in the General Report in Atom?

How to interpret the metrics of the General Report in Atom?

This article explains how to read the Atom General Report, what each metric means, and how to correctly interpret the data, both inbound and outbound.

❓Frequently asked question or initial context

  • Where can I see a summary of all conversations initiated and assigned?

Where can I see a summary of all conversations initiated and assigned?

  • Why do some conversations appear without a group or agent even though they were assigned?

Why do some conversations appear without a group or agent even though they were assigned?

  • How are campaigns, forms, and webhooks reflected in the report?

How are campaigns, forms, and webhooks reflected in the report?

  • How to identify sales or form responses?

How to identify sales or form responses?

The General Report consolidates all this information and allows you to understand the volume, distribution, and performance of conversations according to their origin, assignment, and closure.

🤓 Explanation

The General Report is divided into three main tabs:

  • General Tab: Shows the global metrics of conversations initiated, assigned, sales, and typifications.

General Tab: Shows the global metrics of conversations initiated, assigned, sales, and typifications.

  • Traffic Source Tab: groups the data by type of origin (campaigns, webhooks, manual message, etc.).

Traffic Source Tab: groups the data by type of origin (campaigns, webhooks, manual message, etc.).

  • Audit Tab: allows you to see the detail by conversation, including all available fields.

Audit Tab: allows you to see the detail by conversation, including all available fields.

Main report filters:

  • Conversation start date

Conversation start date

  • Channel (platform + line)

Channel (platform + line)

  • Type of origin (Click to WA, webhook, manual message, etc.)

Type of origin (Click to WA, webhook, manual message, etc.)

  • Group and assigned agent

Group and assigned agent

  • Typification type and final typification (only shows the last assigned)

Typification type and final typification (only shows the last assigned)

⚙️ Troubleshooting: Step by step

To check that the metrics are correct:

  • Go to Reports > General Report.

Go to Reports > General Report.

  • In the General Tab, observe:Unique contactsConversations startedAssigned to agentsSalesSales funnelTypification distribution

In the General Tab, observe:

  • Unique contacts

Unique contacts

  • Conversations started

Conversations started

  • Assigned to agents

Assigned to agents

  • Sales

Sales

  • Sales funnel

Sales funnel

  • Typification distribution

Typification distribution

  • In the Traffic Source Tab, check if the conversations come from:WebhooksCampaigns (flow or template)Click to WhatsAppAgent messages

In the Traffic Source Tab, check if the conversations come from:

  • Webhooks

Webhooks

  • Campaigns (flow or template)

Campaigns (flow or template)

  • Click to WhatsApp

Click to WhatsApp

  • Agent messages

Agent messages

  • To review forms or surveys connected by webhook:Go to theAudit TabFilter by origin "Webhook"Download the file in.csvo.xlsxCheck if the response fields are complete

To review forms or surveys connected by webhook:

  • Go to theAudit Tab

Go to theAudit Tab

  • Filter by origin "Webhook"

Filter by origin "Webhook"

  • Download the file in.csvo.xlsx

Download the file in.csvo.xlsx

  • Check if the response fields are complete

Check if the response fields are complete

✅ Possible solutions

📌 Scenario 1: Why does the conversation appear in a different group than the one it was initially assigned to?

TheGeneral Report reflects the historical dataof each conversation, from its beginning to its closing.

When a conversation is initially assigned togroup A, but before being closed it isreassigned to group B, the report will take as validthe last action recorded before closing.

🔍 Example:A conversation was assigned by an Automated Task to theNew Medicinegroup, but then the same agent manually reassigned it toMedicine Consults without the management being closed.📌 The report will show that this conversation belongs to the MedicineConsults group, since that was the last assignment recorded before closing.

📌 Scenario 2: Why do conversations appear with the origin "Automated Task" and a red triangle?

This is an expected behavior and is related toAutomated Tasks (TAs)configured so that the bot sends aquick response.

🔎 Example:In the case of ISU University, conversations originate from campaigns, are assigned to the agent, andremain inactive until they close automatically.

  • After20h 30m, a bot assignment TA with a quick response is executed again.

After20h 30m, a bot assignment TA with a quick response is executed again.

  • Sending this response fails becausethe conversation has expired(exceeded 24 h without activity).

Sending this response fails becausethe conversation has expired(exceeded 24 h without activity).

  • The system interprets this new attempt as anew management initiated by TA, that's why it appears asorigin = TAand shows thered triangle.

The system interprets this new attempt as anew management initiated by TA, that's why it appears asorigin = TAand shows thered triangle.

📌 Recommendation:If the objective of the TA is to contact the client again, it is suggested:

  • Modify the TA to use atemplate.

Modify the TA to use atemplate.

  • Or reconfigure it so that itsactivator is a typification, instead of running on conversations exposed to expiration.

Or reconfigure it so that itsactivator is a typification, instead of running on conversations exposed to expiration.

📌 Scenario 3: Why does a conversation that was assigned appear without an agent or group?

In some cases, although from the conversation view it seems that there was an assignment, in reality thiswas not executeddue to a flow condition.

🔍 Example:If the client indicates abudget less than 30 million, the flow does not derive it to any group because it does not meet the condition.

This causes the conversation to remain unassigned, and this is correctly reflected in the report.

📌 If you wantall conversations to be assigned anyway, even if the client does not qualify, it will be necessaryto adjust the flow logic.

📌 Scenario 4: Why are multiple typifications applied to the same conversation not shown?

By design, the General Reportonly shows the last typification recordedin each conversation, as this is what defines the closure.

If you need to see more detail about previous typifications (for example, from the bot and then from the agent), it is suggested to use theBots Report, specifically theTypification Pathstab.

📝 Additional notes

  • Typifications:Only thelast applied typification is shown. If a conversation had several typifications (bot + agent), only the final one will be seen.To visualize the complete path within the bot, it is recommended to use theBots Report.

Typifications:Only thelast applied typification is shown. If a conversation had several typifications (bot + agent), only the final one will be seen.To visualize the complete path within the bot, it is recommended to use theBots Report.

  • Audit:Allows you to see up to 2,000 records on the platform and download up to 1,000,000 in.csv,.xlsx,.json.

Audit:Allows you to see up to 2,000 records on the platform and download up to 1,000,000 in.csv,.xlsx,.json.

  • Differences with “Your Subscription”:The metrics of the General Report do not match those of billing. Atom bills according to Meta's conversation windows, while this report analyzes conversations opened until their final typification.

Differences with “Your Subscription”:The metrics of the General Report do not match those of billing. Atom bills according to Meta's conversation windows, while this report analyzes conversations opened until their final typification.

Did this answer your question?