Customer Segmentation and Persona Classification

This dataset provides comprehensive behavioral segmentation and persona classification for subscribers, integrating usage patterns, device types, service preferences, spending tiers, demographics, and predictive propensity scores. It is ideal for targeted marketing, personalized product recommendations, and advanced customer analytics to drive engagement and retention.

  • opened 6 times
  • last updated 1 Nov 2025
  • by GoMask
The brief that made it

Targeted marketing campaign optimization

Sample rows

preview · 8 of 1,000 rows · all 22 columns
subscriber_idstringusage_patternstringpropensity_to_churnfloatmarketing_opt_inbooleangenderstringfirst_namestringlast_namestringemailstringdate_of_birthdateaddress_streetstringaddress_citystringaddress_statestringaddress_postal_codestringaddress_countrystringdevice_typestringservice_preferencesstringaverage_monthly_spendfloatspending_tierstringpersonastringpropensity_to_buyfloatlast_activity_datedatetimeaccount_creation_datedate
A0100heavy0.09truefemaleJulietChenjuliet.chen@globeplus.net1999-04-121023 Oak LaneTorontoONM5V3C6Canadasmartphonestreaming,messaging,shopping182.55highurban multitasker0.892024-06-07T19:44:152023-11-18
A0101light0.94falsemaleMohammedAl-Farsimohammed.alfarsi@uaetel.com1981-09-0588 Crescent RdDubaiDU00000UAEtabletnews,finance,education34.4lowretired observer0.082024-04-23T08:30:102013-07-30
A0102heavy0.04truemaleLucasMorenolucas.moreno@argmail.com2003-12-215 Calle 7Buenos AiresBAC1012Argentinalaptopgaming,streaming,education245.7highyoung gamer0.982024-06-06T22:11:172023-09-12
A0103light0.86falsefemaleAvaMüllerava.mueller@t-online.de1972-06-302 Gartenstr.FrankfurtHE60313Germanydesktopnews,shopping,finance19.9lowclassic reader0.132024-05-28T11:12:412012-05-25
A0104moderate0.31truefemaleSophieDuboissophie.dubois@orange.fr1989-03-1722 Rue des LilasParisIDF75019Francesmart_tvstreaming,sports,news66.85mediumentertainment lover0.522024-06-05T15:21:422017-04-10
A0105heavy0.02truemaleXinZhangxin.zhang@163.com1996-11-031018 Nanjing RdShanghaiSH200040Chinasmartphonestreaming,messaging,gaming168.23highdigital native0.922024-06-07T10:45:382024-01-09
A0106light0.92falsefemaleEmilyCarteremily.carter@usamail.com1982-08-2157 Willow RdCharlotteNC28209USAtableteducation,news,finance28.54lowretired observer0.112024-05-14T13:06:022011-08-27
A0107heavy0.07truemaleOscarNguyenoscar.nguyen@singnet.com.sg1995-05-0841 Orchard RdSingaporeSG238859Singaporelaptopgaming,streaming,shopping222.88highpower gamer0.952024-06-07T09:37:562023-06-18

What the 1,000 rows show

from the 1,000-row sample

Light (usage pattern) stands out: mean propensity_to_churn is 0.89, against 0.16 for the rest.

  • 58%marketing_opt_in = true
  • 0.22median propensity_to_churn
  • 4spending tiers
  • 6device types
  • 12personas
  • 44address countries
Mean propensity_to_churn by usage_pattern999 rows
00.510.08heavy405 rows0.27moderate314 rows0.89light280 rows
propensity_to_churn999 rows, in bands of 0.1
013026025919316799100014513500.51propensity_to_churn →

Median 0.22, from 0.01 to 0.99.

gender999 rows with a value · 1 left blank
  1. female509
  2. male448
  3. prefer not to say28
  4. non-binary13
  5. other1
22 columns by typefrom the column list below
  • string 15
  • float 3
  • date 2
  • datetime 1
  • boolean 1

Columns

22 columns in four groups
blueprint · 22 columns
columntypedescriptionexample
Text 15 columns
subscriber_idstringUnique identifier for each subscriberuniqueA0100
first_namestringSubscriber's first nameJuliet
last_namestringSubscriber's last nameChen
emailstringSubscriber's email addressuniquelucas.moreno@argmail.com
genderstringSubscriber's gendermale · female · non-binary · prefer not to say · optionalfemale
address_streetstringSubscriber's street addressoptional1023 Oak Lane
address_citystringSubscriber's cityoptionalToronto
address_statestringSubscriber's state or provinceoptionalON
address_postal_codestringSubscriber's postal codeoptionalM5V3C6
address_countrystringSubscriber's countryoptionalCanada
device_typestringPrimary device type used by the subscriber6 valuessmartphone
service_preferencesstringComma-separated list of preferred services (e.g., streaming, messaging, gaming)optionalnews,finance,education
usage_patternstringCategorization of subscriber's usage pattern (e.g., heavy, moderate, light)heavy · moderate · lightheavy
spending_tierstringSpending tier classification (e.g., low, medium, high, premium)low · medium · high · premiumhigh
personastringAssigned persona label based on segmentation (e.g., 'tech enthusiast', 'budget user', 'family planner')12 personasurban multitasker
Numbers 3 columns
average_monthly_spendfloatAverage monthly spend by the subscriber in USD0 or more182.55
propensity_to_buyfloatPropensity score for likelihood to purchase recommended products (0-1 scale)0 to 1 · optional0.89
propensity_to_churnfloatPropensity score for likelihood to churn (0-1 scale)0 to 1 · optional0.09
Dates and times 3 columns
date_of_birthdateSubscriber's date of birthoptional1999-04-12
last_activity_datedatetimeTimestamp of subscriber's last recorded activityoptional2024-06-07T19:44:15
account_creation_datedateDate when the subscriber account was created2023-11-18
True or false 1 column
marketing_opt_inbooleanIndicates if subscriber has opted in to marketing communicationstrue

Use it for

  • marketing opt …58%582 of 999 rowsmean propensity to ch…0.08heavy0.27modera…0.89light

    A marketing dashboard

    The marketing_opt_in rate, propensity_to_churn by usage_pattern and a breakdown of gender. Excel, Power BI or Tableau.

  • Why do the 280 light rows have a mean propensity_to_churn of 0.89?

    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 subscribers with first_name, last_name and email to fill a screen in front of a buyer.

Not quite right?

Make it yours.

Same 22 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 dataset1,000 rows22 columns
Yours1,000 rows22 columnsaddress_street: UK only

blueprint · customer-segmentation-and-persona-classification

Behind this dataset

Same schema. As many rows as you need.

These 1,000 rows came out of a blueprint — 22 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
  • Subscriber MSISDN and assigned segment/persona
  • Segmentation model: RFM (Recency, Frequency, Monetary), clustering, decision trees
  • Usage persona: data-heavy, voice-centric, balanced, IoT/M2M, roamer, family
  • Device persona: premium flagship, mid-tier, budget, feature phone
  • Spending tier: high-value (top 10%), mid-tier, low-value (bottom 25%)
  • Demographic segments: age group, urban/rural, business/consumer
  • Service preference: 5G early adopter, video streamer, gamer, traveler
  • Propensity scores: upgrade likelihood, add-on purchase, international roaming
Rows
Open the blueprint in Data Factory

1 credit per row. New accounts start with 25 free credits.

Exports
CSV, JSON, JSONL, Parquet, SQL, Excel, TSV, XML
Licence
yours to use, including commercially
API slug
customer-segmentation-and-persona-classification

More in Marketing

Browse all
Also in MarketingSynthetic Retail Promotion UpliftEmail Campaign LogsMarketing Attribution Path DatasetAll 1,954 datasets

What should your data show?

Preview 20 rows free
No signup. No card.