๐Ÿ“Š Sample workspace โ€” Telco Customer Dataset (7,043 rows)

Analyze your own data

Demo workspace ยท read-only

Telco customer churn โ€” what's driving it

A completed analysis of 7,043 telco customers (26.5% churned). Explore the real insights, charts, and a pinned dashboard โ€” no signup needed.

Generaldataset type

Rows

7,043

Columns

21

Numeric

3

Categorical

3

Missing

0.2%

88/100

Data quality report

Grade BGeneral

Strong โ€” ready for analysis

Missing data ยท raw upload: 0.16% of cells (~11 rows with any gap)ยทafter cleaning: 0.16% of cells (~11 rows with any gap)

7,043 rows ยท 21 columns

Score breakdown

Completeness

99.0/30

Uniqueness

95.0/20

Consistency

86.0/20

Validity

82.0/15

Structure

78.0/15

Client readiness

โœ“

Missing data is under 1% โ€” minimal impact on analysis

โœ“

No duplicate rows detected

โœ“

No critical structural issues detected

Score deductions

11 rows have blank TotalCharges (all brand-new, tenure = 0)

SeniorCitizen is stored as a 0/1 integer rather than a boolean flag

What looks healthy

โœ“

Completeness is strong โ€” very few missing values

โœ“

Uniqueness is strong โ€” minimal duplicates

โœ“

Data formats are consistent

โœ“

Value distributions look normal

โœ“

Column structure is clean

Cleaning Summary

Cleaning summary

Metrics from this run only.

Cleaning steps

4

Type conversions

1

Normalized dtypes in pipeline.

Suspicious columns

1

Review below or in Cleaning.

Top highlights

#1 finding
โœ“ Report-ready

Contract type is the #1 churn driver โ€” month-to-month churns 8ร— more

Month-to-month customers churn at 42.7% versus just 5.1% on one- or two-year contracts โ€” a 38.4pp gap. A logistic model confirms Contract is the single strongest predictor of churn (AUC 0.84), ahead of charges, tenure, and services.

#2 finding
โœ“ Report-ready

Fiber-optic customers churn the most (41.8%)

Fiber optic internet customers churn at 41.8%, more than double DSL (19.0%) and 5ร— customers with no internet service (7.4%). Despite higher revenue, fiber's churn risk erodes lifetime value.

#3 finding
โœ“ Report-ready

Electronic-check payers churn at 45%

Customers paying by electronic check churn at 45.2% โ€” roughly triple the rate of automatic payment methods (bank transfer 16.7%, credit card 15.2%). Payment friction is a strong retention signal.

All insights

6 findings5 report-ready1 informational

For your report

5 findingsSafe to include without manual review
SegmentHigh priorityโœ“ Report-ready100% confidence

Contract type is the #1 churn driver โ€” month-to-month churns 8ร— more

Month-to-month customers churn at 42.7% versus just 5.1% on one- or two-year contracts โ€” a 38.4pp gap. A logistic model confirms Contract is the single strongest predictor of churn (AUC 0.84), ahead of charges, tenure, and services.

SegmentHigh priorityโœ“ Report-ready100% confidence

Fiber-optic customers churn the most (41.8%)

Fiber optic internet customers churn at 41.8%, more than double DSL (19.0%) and 5ร— customers with no internet service (7.4%). Despite higher revenue, fiber's churn risk erodes lifetime value.

Segmentโœ“ Report-ready100% confidence

Electronic-check payers churn at 45%

Customers paying by electronic check churn at 45.2% โ€” roughly triple the rate of automatic payment methods (bank transfer 16.7%, credit card 15.2%). Payment friction is a strong retention signal.

Correlationโœ“ Report-ready100% confidence

Shorter tenure is strongly linked to churn (r = โˆ’0.35)

Churn falls steeply as tenure grows. Churned customers have a median tenure of 10 months versus 38 months for retained customers โ€” most churn happens in the first year.

Segmentโœ“ Report-ready100% confidence

Senior citizens churn at nearly double the base rate

Senior citizens churn at 41.7% versus 23.6% for non-seniors. Seniors skew toward fiber and electronic check, compounding their churn risk.

5 findings marked report-ready โ€” head to Report to select which ones to include.

Ready to analyze your own data?

Upload any CSV and get insights, charts, and a shareable dashboard in 90 seconds.

Analyze your own data