Bank Call Center Interaction Logs

This dataset provides granular logs of bank call center interactions, capturing detailed information about each customer-agent conversation, including timing, outcomes, speech analytics, sentiment, and satisfaction scores. It enables comprehensive analysis of customer experience, agent performance, and operational trends, supporting both quality assurance and advanced speech analytics use cases.

  • last updated 22 Jul 2025
  • by GoMask
The brief that made it

Speech analytics and sentiment analysis for quality assurance

Sample rows

preview · 8 of 200 rows · all 18 columns
interaction_idstringcall_statusstringspeech_sentiment_scorefloatissue_resolvedbooleancall_channelstringcustomer_idstringagent_idstringinteraction_datetimedatetimeinteraction_duration_secondsintegercall_directionstringcustomer_satisfaction_scorefloattopics_detectedstringfollow_up_requiredbooleanfollow_up_due_datedatelanguagestringcustomer_feedback_textstringspeech_transcriptstringagent_notesstring
INT00001completed0.82truephoneCUST00001AGT00012024-01-01T09:10:13415inbound4.7password_reset,online_bankingfalseblankenThank you for your help, great service.Hello, I need help resetting my password please.Customer called regarding password reset. Issue resolved during the call. Provided instructions and ensured access was restored.
INT00002droppedblankfalsephoneCUST00002AGT00022024-01-01T11:24:5023outboundblankblankfalseblankenblankblankCall dropped instantly. No customer response.
INT00003completed0.63truechatCUST00003AGT00032024-01-01T13:32:05198inbound4.1statement_request,balance_inquiryfalseblankesGracias, todo bien.Hi, I want to check my last three statements.Customer requested recent statements. Provided download links and confirmed understanding.
INT00004abandoned-0.95falsephoneCUST00001AGT00042024-01-02T10:45:1437inbound0blankfalseblankenblankblankCaller disconnected before an agent could answer.
INT00005completed0.91truephoneCUST00004AGT00012024-01-02T15:08:55720inbound4.5card_issue,online_paymentfalseblankenResolved my issue quickly.I am having trouble with my debit card online purchase.Customer reported card decline online. Walked through troubleshooting and resolved the problem.
INT00006completed0.78truevideoCUST00005AGT00052024-01-02T16:42:30312outbound4.2customer_outreach,account_upgradefalseblankfrMerci pour l'information.Hi, we are reaching out to discuss your account upgrade options.Proactive outreach regarding account upgrades. Customer interested in more info.
INT00007abandonedblankfalsephoneCUST00006AGT00062024-01-03T10:14:0751outboundblankblankfalseblankenblankblankOutbound call not answered by customer.
INT00008transferred-0.62falsephoneCUST00007AGT00072024-01-03T11:25:101680inbound1.5complaint,fees,escalationtrue2024-01-09enStill waiting for resolution.I'm not happy with the recent charges. Can I speak to a supervisor?Customer escalated complaint regarding fees. Call transferred to supervisor. Follow-up required.

What the 200 rows show

from the 200-row sample

Completed (call status) stands out: mean speech_sentiment_score is 0.66, against -0.67 for the rest.

  • 59%issue_resolved = true
  • 0.60median speech_sentiment_score
  • 2call directions
  • 3languages
  • 203.5median interaction_duration_seconds
  • 3.9median customer_satisfaction_score
speech_sentiment_score144 rows, in bands of 0.2
020408882015404032issue_resolved from 0-101speech_sentiment_score →

Every row with speech_sentiment_score of 0 or more has issue_resolved = true. None below it.

call_channel200 rows · 3 values
  1. phone123
  2. chat53
  3. video24
18 columns by typefrom the column list below
  • string 11
  • integer 1
  • float 2
  • date 1
  • datetime 1
  • boolean 2

Columns

18 columns in four groups
blueprint · 18 columns
columntypedescriptionexample
Text 11 columns
interaction_idstringUnique identifier for each call center interactionuniqueINT00001
customer_idstringUnique identifier for the customer involved in the interactionCUST00001
agent_idstringUnique identifier for the call center agent handling the interactionAGT0001
call_directionstringDirection of the call: inbound (customer to bank) or outbound (bank to customer)inbound · outboundinbound
call_channelstringChannel used for the interaction (e.g., phone, video, chat)phone · video · chatphone
call_statusstringFinal status of the call (e.g., completed, abandoned, transferred, dropped)completed · abandoned · transferred · droppedcompleted
speech_transcriptstringFull transcript of the interaction as captured by speech-to-text analyticsoptionalHow do I change my PIN?
topics_detectedstringComma-separated list of main topics discussed during the interactionoptionalalerts,online_banking
languagestringPrimary language used during the interaction (e.g., en, es, fr)3 languagesen
customer_feedback_textstringFree-text feedback provided by the customer after the interactionoptionalGracias, todo bien.
agent_notesstringNotes or summary entered by the agent after the interactionoptionalNo answer, call ended.
Numbers 3 columns
interaction_duration_secondsintegerTotal duration of the interaction in seconds1 or more415
customer_satisfaction_scorefloatScore reflecting customer satisfaction with the interaction (0.0 to 5.0)0 to 5 · optional4.7
speech_sentiment_scorefloatSentiment score derived from speech analytics (-1.0 negative to 1.0 positive)-1 to 1 · optional0.82
Dates and times 2 columns
interaction_datetimedatetimeDate and time when the interaction started2024-01-01T09:10:13
follow_up_due_datedateDate by which follow-up action should be completed, if requiredoptional2024-01-09
True or false 2 columns
issue_resolvedbooleanIndicates whether the customer's issue was resolved during the interactiontrue
follow_up_requiredbooleanIndicates if a follow-up action is required after the interactionfalse

Use it for

  • issue resolved59%118 of 200 rows

    A finance dashboard

    The issue_resolved rate, speech_sentiment_score by call_status and a breakdown of call_channel. Excel, Power BI or Tableau.

  • Why do the 118 completed rows have a mean speech_sentiment_score of 0.66?

    A root-cause class exercise

    Hand out the rows and one question. The answer is in the data, not in the brief.

  • A software demo

    Believable interactions with customer_id, agent_id and interaction_datetime to fill a screen in front of a buyer.

Not quite right?

Make it yours.

Same 18 columns, your size and your rules. See 20 rows before you pay.

Preview 20 rows free

1,000 rows: $1.99 · 10,000: $12.99One-time. No subscription. All prices

This dataset200 rows18 columns
Yours1,000 rows18 columns

blueprint · bank-call-center-interaction-logs

Behind this dataset

Same schema. As many rows as you need.

These 200 rows came out of a blueprint — 18 columns with generation rules behind each one. Open it in Data Factory to retune a column, add your own, wire in foreign keys, and run it at the size you actually need.

Rules it was built with
  • Each record logs customer and agent IDs, timestamp.
  • Interaction type (inquiry, complaint, etc.) classified.
  • Call duration, resolution status included.
  • Sentiment score derived from conversation.
Rows
Open the blueprint in Data Factory

1 credit per row. Preview 20 rows free. Buy rows for the full run.

Exports
CSV, JSON, JSONL, Parquet, SQL, Excel, TSV, XML
Licence
yours to use, including commercially
API slug
bank-call-center-interaction-logs

What should your data show?

Preview 20 rows free
No signup. No card.