View a markdown version of this page

Bot analytics data in the Connect Customer analytics data lake - Amazon Connect Customer

Bot analytics data in the Connect Customer analytics data lake

This topic details the content in the Connect Customer data lake bot tables. The tables list the column, type, and description of the content.

There are two ways to access the analytics data lake and configure data to be shared:

If you are unable to access the scheduling tables by using Option 1, try using Option 2.

Bot conversations

Table name: bot_conversations

Description: Records bot conversation sessions including conversation outcomes, turn counts, and associated flow and contact context for Lex bot interactions.

Primary key: instance_id, event_id

Partition key: bot_conversation_start_timestamp (daily)

Join keys:

  • instance_id — Joins to all tables

  • contact_id — Joins to Contact Record

  • flow_resource_id — Joins to Contact Flow Events

  • bot_originating_request_id, bot_session_id — Joins to bot_intents

Column Type Description
event_id String No This is the primary key of the table. This will be a hash of originating request identifier and session identifier.
bot_originating_request_id String Yes A unique identifier for a specific bot request.
bot_session_id String Yes The identifier of the user session that is having the conversation.
aws_account_id String Yes The identifier of the AWS account that owns the contact.
instance_arn String Yes The ARN of the Connect Customer instance.
instance_id String No The identifier of the Connect Customer instance.
invoking_resource_type String Yes Can be flow or module.
flow_resource_id String Yes Flow identifier.
module_resource_id String Yes Module identifier.
invoking_resource_start_timestamp Timestamp Yes Time at which flow started.
parent_flow_resource_id String Yes Flow Id from which the module was invoked from. This field will only be populated for modules.
contact_id String Yes The identifier of the contact.
flow_action_id String Yes Identifier for action that was executed. An Action is a single step of a flow's run.
invoking_resource_published_timestamp Timestamp Yes "Creation" or "revision" date of the flow itself.
flow_type String Yes Connect Customer includes a set of nine flow types. For more information, see Choose a flow type.
channel String Yes The method used to contact your contact center: VOICE, CHAT, TASK.
sub_type String Yes This subtype for the contact. For example, connect:Guide or connect:SMS.
initiation_method String Yes Indicates how the contact was initiated.
flow_language_version String Yes Flow language version.
invoking_resource_version String Yes Version of the contact flow used.
bot_id String Yes The identifier of the bot.
bot_alias_id String Yes The alias identifier of the bot that the session was held with.
bot_version String Yes The version of the bot that the session was held with.
bot_locale String Yes Language configuration of the bot.
bot_conversation_start_timestamp Timestamp Yes The Timestamp marking the start of the conversation with the bot.
bot_conversation_end_timestamp Timestamp Yes The Timestamp marking the end of the conversation with the bot.
bot_conversation_outcome String Yes The final state of the conversation. Values: Success | Failure | Dropped
bot_number_of_conversation_turns Number Yes The number of turns that the session took.
data_lake_last_processed_timestamp Timestamp Yes Timestamp, which shows the last time the data lake processed the record. This can include transformation and backfill. This field cannot be used to determine reliably data freshness.

Bot intents

Table name: bot_intents

Description: Records intent-level outcomes for bot conversations, tracking intent names, levels, and resolution status for each bot interaction turn.

Primary key: instance_id, event_id

Partition key: bot_conversation_start_timestamp (daily)

Join keys:

  • instance_id — Joins to all tables

  • bot_originating_request_id, bot_session_id — Joins to bot_conversations

  • bot_originating_request_id, bot_session_id, bot_intent_name, bot_intent_level — Joins to bot_slots

  • contact_id — Joins to Contact Record

Column Type Description
event_id String No This is the primary key of the table. This will be a hash of originating request identifier, session identifier, intent name, and intent level.
bot_originating_request_id String Yes A unique identifier for a specific bot request.
bot_session_id String Yes The identifier of the user session that is having the conversation.
account_id String Yes The identifier of the AWS account that owns the contact.
instance_arn String Yes The ARN of the Connect Customer instance.
instance_id String No The identifier of the Connect Customer instance.
invoking_resource_type String Yes Can be flow or module.
flow_resource_id String Yes The flow identifier.
module_resource_id String Yes The module identifier.
invoking_resource_start_timestamp Timestamp Yes The time when the flow started.
parent_flow_resource_id String Yes The flow identifier where the module was invoked. This field is only be populated for modules.
contact_id String Yes The identifier of the contact.
action_id String Yes The identifier for action that was executed. An Action is a single step of a flow's run.
invoking_resource_published_timestamp Timestamp Yes The "creation" or "revision" date of the flow itself.
flow_type String Yes The type of flow. Connect Customer includes a set of nine flow types. For more information, see Choose a flow type.
channel String Yes The method used to contact your contact center: VOICE, CHAT, TASK.
sub_type String Yes The subtype for the contact. For example, connect:Guide or connect:SMS.
initiation_method String Yes How the contact was initiated.
flow_language_version String Yes The flow language version.
invoking_resource_version String Yes Version of the contact flow used.
bot_id String Yes The identifier of the bot.
bot_alias_id String Yes The alias identifier of the bot that the session was held. with
bot_version String Yes The version of the bot that the session was held with.
bot_locale String Yes The language configuration of the bot.
bot_conversation_start_timestamp Timestamp Yes The Timestamp marking the start of the conversation with the bot.
bot_conversation_end_timestamp Timestamp Yes The Timestamp marking the end of the conversation with the bot.
bot_intent_name String Yes The name of the intent.
bot_intent_level Number Yes The number of intents up to and including the requested path.
bot_intent_outcome String Yes The end state of the intent. Value of Success, Failed, Switched, or Dropped.
data_lake_last_processed_timestamp Timestamp Yes The Timestamp, which shows the last time the data lake processed the record. This can include transformation and backfill. This field cannot be used to determine reliably data freshness.

Bot slots

Table name: bot_slots

Description: Records slot-level outcomes for bot intent fulfillment, tracking slot resolution status and retry counts for each slot within an intent.

Primary key: instance_id, event_id

Partition key: bot_conversation_start_timestamp (daily)

Join keys:

  • instance_id — Joins to all tables

  • bot_originating_request_id, bot_session_id, bot_intent_name, bot_intent_level — Joins to bot_intents

  • bot_originating_request_id, bot_session_id — Joins to bot_conversations

  • contact_id — Joins to Contact Record

Column Type Description
event_id String No This is the primary key of the table. This will be a hash of originating request identifier, session identifier, intent name, intent level, slot name, and slot level.
bot_originating_request_id String Yes A unique identifier for a specific bot request
bot_session_id String Yes The identifier of the user session that is having the conversation.
account_id String Yes The identifier of the AWS account that owns the contact.
instance_arn String Yes The ARN of the Connect Customer instance.
instance_id String No The identifier of the Connect Customer instance.
invoking_resource_type String Yes Can be flow or module.
flow_resource_id String Yes Flow identifier.
module_resource_id String Yes Module identifier.
invoking_resource_start_timestamp Timestamp Yes Time at which flow started.
parent_flow_resource_id String Yes Flow identifier from which the module was invoked from. This field will only be populated for modules.
contact_id String Yes The identifier of the contact.
action_id String Yes Identifier for action that was executed. An Action is a single step of a flow's run.
invoking_resource_published_timestamp Timestamp Yes "Creation" or "revision" date of the flow itself.
flow_type String Yes Connect Customer includes a set of nine flow types. For more information, see Choose a flow type.
channel String Yes The method used to contact your contact center: VOICE, CHAT, TASK.
sub_type String Yes This subtype for the contact. For example, connect:Guide or connect:SMS.
initiation_method String Yes Indicates how the contact was initiated.
flow_language_version String Yes Flow language version.
invoking_resource_version String Yes Version of the contact flow used.
bot_id String Yes The identifier of the bot.
bot_alias_id String Yes The alias identifier of the bot that the session was held with.
bot_version String Yes The version of the bot that the session was held with.
bot_locale String Yes Language configuration of the bot.
bot_conversation_start_timestamp Timestamp Yes The Timestamp marking the start of the conversation with the bot.
bot_conversation_end_timestamp Timestamp Yes The Timestamp marking the end of the conversation with the bot.
bot_intent_name String Yes The name of the intent.
bot_intent_level Number Yes The number of intents up to and including the requested path.
bot_slot_name String Yes The name of the slot.
bot_slot_level Number Yes The number of slots up to and including the requested path.
bot_slot_outcome String Yes The end state of the slot. Values of Success, Failed, Dropped, or Retry.
bot_slot_retry_count Number Yes The number of times the bot tried to elicit a response from the user for the slot.
data_lake_last_processed_timestamp Timestamp Yes Timestamp, which shows the last time the data lake processed the record. This can include transformation and backfill. This field cannot be used to determine reliably data freshness.