# Frequent-Contacts Workbook — Repeat User Analysis The Frequent-Contacts workbook is the third workbook in the Conversations gallery. It is designed to identify users (identified by their ANI — Automatic Number Identification) who initiate the highest volume of conversations within the selected period. Monitoring frequent contacts is a critical operational metric: in an AI assistant platform, a single user or endpoint generating an unusually high number of conversations may indicate an unresolved issue, a looping bot behavior, automated testing traffic, or a specific customer segment with persistent unmet needs. The workbook contains six sub-views accessible from the sub-tab bar: Dashboard | Top Frequent Conversations | Frequent Conversations by Date | Frequent Conversations by Date, ANI | Frequent Conversations by Date, Connection Disposition | Top Frequent Conversations by Date | Details Page. ### 1. Dashboard — Top Frequent Conversations (with Data) The Dashboard sub-view is the primary view with meaningful data. It consolidates three sections: a Top N Frequent identifier bar, a time-series volume chart, and a tabular daily breakdown.

Figure 1: Frequent-Contacts Dashboard — Top 10 Frequent Conversations (Employee Platform)

### 2. Section 1 — Top 10 Frequent ANI Bar The uppermost section displays a horizontal bar chart titled “Top 10 Frequent” with a configurable “Top:” control (set to 10 by default, adjustable by the user). Each row represents one ANI — the unique identifier of a caller or user endpoint. In the visible data, only one ANI row is present: INB-CHAT-USER-ENDPOINT. This is the system-generated identifier used for inbound chat sessions on the Employee platform (IX Hello Employee). The bar spans nearly the full width, indicating this single endpoint accounts for virtually all conversation traffic in the selected period and environment. The label format “INB-CHAT-USER-ENDPOINT” confirms this is a chat channel (not voice) and an inbound session type. ⚠ Note: INB-CHAT-USER-ENDPOINT is a generic pooled identifier used for all inbound chat users in the Employee platform. In production with real user data, individual employee identifiers (e.g., employee IDs or email addresses) would appear as separate ANI rows, enabling per-user frequency analysis. In the current dataset this reflects aggregated chat traffic under a single endpoint label. ### 3. Section 2 — Frequent Conversations by (Time-Series Chart) The middle section displays a time-series line chart plotting daily conversation counts for the top frequent ANI(s) across the entire available date range. This chart uses a significantly wider date window than other dashboards, spanning from Wed 3/11/26 through Tue 4/7/26, and reveals the full conversation volume trajectory for the top identifiers. Data points visible on the chart (INB-CHAT-USER-ENDPOINT):
DateConversation Count
Wed 3/11/2610
Fri 3/20/2642
Sat 3/21/2644
Sun 3/22/268
Mon 3/23/26114
Tue 3/24/26142
Wed 3/25/26114
Thu 3/26/26105
Fri 3/27/26109
Sat 3/28/2677
Sun 3/29/2615
Mon 3/30/269,783 — major spike begins
Tue 3/31/2653,595 — peak value (labeled on chart)
Wed 4/1/26127
Thu 4/2/26408
Fri 4/3/26180
Sat 4/4/26280
Sun 4/5/2667
Mon 4/6/2664
Tue 4/7/26101
The spike to 9,783 on Mon 3/30/26 and 53,595 on Tue 3/31/26 is a dramatic anomaly — the conversation count on 3/31 is approximately 375 times the typical daily volume of 100–150 conversations seen in the surrounding days. This pattern is characteristic of one of three scenarios: (1) an automated load test or stress test executed against the Employee platform on those dates; (2) a data ingestion issue that caused records from multiple days to be timestamped on 3/31; or (3) a real production event such as a company-wide push notification or employee onboarding campaign that drove a massive surge in platform engagement. The sharp return to normal volumes from 4/1 onwards supports the first two interpretations. ### 4. Section 3 — Frequent Conversations by Date (Tabular) The lower section of the Dashboard shows a simple two-column data table: Day of Conversation Date and ANI with conversation count. This is the tabular companion to the line chart above, listing each date-ANI combination with its exact count. The visible rows confirm the data points charted above (10, 42, 44, 8, 114, 142, 114, 105 for the period 3/11 through 3/26). ### 5. Sub-View Reference — Remaining Sub-Views The following sub-views are available within the Frequent-Contacts workbook. Each provides a distinct analytical lens on the same frequent-contact dataset. In the current STG environment with Channel and Assistant filters set to (None), these views render without data but become active once filters are applied or in environments with populated data.
Sub-ViewWhat It Provides
Top Frequent ConversationsA ranked list of the top N ANIs (configurable via the Top: control) sorted by total conversation count across the entire selected period. Enables quick identification of the highest-volume users or endpoints. Filters: Date range, Channel, Stage, Assistant, Business Day.
Frequent Conversations by DateA time-series chart showing daily conversation counts for frequent contacts, filterable by Channel, Stage, and Business Day. Useful for tracking whether repeat-contact behavior is consistent over time or concentrated on specific dates.
Frequent Conversations by Date, ANIExtends the “by Date” view by adding the ANI dimension — enabling analysts to track individual users’ conversation frequency day by day. Useful for customer journey analysis: when did this specific user start contacting the platform repeatedly?
Frequent Conversations by Date, Connection DispositionAdds Connection Disposition to the date-level frequent contact view. Enables analysis of whether high-frequency users are experiencing a specific outcome (e.g., repeated Timeouts or AppDisconnects) that may explain their repeat contact behavior.
Top Frequent Conversations by DateCombines the Top N ranking with a date breakdown — showing which top frequent ANIs were active on which dates. The Top: control (default: 10) limits the display to the highest-volume identifiers. Useful for identifying whether a spike in frequent contacts is driven by a single user or multiple users simultaneously.
Details PageRecord-level drill-down showing individual conversation records for the selected frequent contacts, with full metadata (Organization, ConversationId, Start/End timestamps, ANI, DNIS, Channel, Stage, Assistant).
⚠ Note: In the current screenshots, the sub-views for Frequent Conversations by Date, Frequent Conversations by Date ANI, Frequent Conversations by Date Connection Disposition, and Top Frequent Conversations by Date all show empty charts because Channel and Assistant are set to (None) in the STG environment. Selecting (All) for Channel and a valid Stage (PROD or STG with data) will populate these views. The Dashboard sub-view (Section 26.1) displays data because it uses a broader query that does not require a channel selection. ### 6. Business Value The Frequent-Contacts workbook addresses one of the most important quality signals in AI assistant operations: repeat contact rate. In a well-functioning AI assistant, users should resolve their needs in a single conversation. When the same user returns multiple times, it typically indicates the assistant failed to fully resolve their intent on the first (or subsequent) attempts. Key use cases for this workbook include: (1) Identifying chronic unresolved intents — users who return 5+ times in a week likely represent an intent the assistant cannot handle, providing direct input for bot improvement priorities. (2) Detecting automated traffic — extremely high ANI counts (as seen with the 53,595 spike on 3/31) may indicate load testing, bot traffic, or data issues requiring investigation. (3) Customer success flagging — in production, flagging specific customers who contact support repeatedly enables proactive outreach by human agents before the customer escalates. (4) SLA and compliance reporting — some contracts include commitments on first-contact resolution rates; this workbook provides the raw data needed to calculate and report on those metrics. --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.ixhello.com/ix-hello-reporting/premium-reporting/total-conversations-with-contained-and-transferred-dedicated-sub-view/frequent-contacts-workbook-repeat-user-analysis.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.