Insurance Claims Severity Analysis

This dataset provides detailed records of insurance claims, including severity ratings, claim amounts, incident types, policy details, and customer demographics. It is ideal for risk modeling, severity analysis, fraud detection, and insurance pricing optimization, supporting both operational and analytical needs in the insurance industry.

  • opened 5 times
  • last updated 20 Aug 2025
  • by GoMask
The brief that made it

Insurance risk modeling and severity analysis

Sample rows

preview · 8 of 200 rows · all 18 columns
claim_idstringseverity_levelstringcustomer_ageintegerfraud_suspectedbooleanincident_typestringpolicy_idstringcustomer_idstringclaim_datedateincident_datedateclaim_statusstringclaim_amountfloatapproved_amountfloatpolicy_typestringcustomer_genderstringcustomer_postal_codestringpolicy_start_datedatepolicy_end_datedateadjuster_notesstring
CLM-2024-AUTO-001medium28falseaccidentP-2021-AUTO-001C-9011-M282024-03-182024-03-17open5420.75blankautomale941052021-03-012025-03-01Customer reported collision at intersection, minor injuries, awaiting police report.
CLM-2023-HEALTH-002medium44falseliabilityP-2022-HEALTH-010C-9022-F442023-10-052023-10-03pending3080.25blankhealthfemale100012022-06-012024-06-01Medical reimbursement claim for workplace injury, awaiting employer verification.
CLM-2022-HOME-003high64falsefireP-2019-HOME-005C-7991-F642022-07-212022-07-15closed125000120000homefemaleV5K3Z12019-04-102024-04-10Kitchen fire caused by stove malfunction, repair estimates verified, approved payout.
CLM-2021-LIFE-004high59falseotherP-2010-LIFE-021C-8993-M592021-02-102021-01-30closed8000080000lifemale750012010-02-012021-02-01Life insurance claim, required death certificate submitted, payout processed.
CLM-2024-BUSINESS-005critical61falsenatural_disasterP-2020-BUSINESS-009C-8822-F612024-01-252024-01-20under_review450000blankbusinessfemaleW1A0AX2020-08-132030-08-13Flooding damaged warehouse, awaiting structural engineer's report.
CLM-2023-AUTO-006medium34falseaccidentP-2022-AUTO-012C-8015-F342023-05-142023-05-13open6190blankautofemale902102022-05-122025-05-12Rear-end collision, vehicle at repair shop, photos and police report pending.
CLM-2022-HEALTH-007low18falseaccidentP-2021-HEALTH-024C-7733-M182022-11-042022-11-02closed2150.52100healthmaleA1A1A12021-01-012023-01-01Minor injury from sports activity, doctor’s note submitted, reimbursement approved.
CLM-2023-BUSINESS-008highblanktruefireP-2018-BUSINESS-011C-8888-PNT2023-08-292023-08-28denied1200000businessprefer_not_to_say001002018-06-012028-06-01Claim denied due to insufficient fire safety compliance documentation.

What the 200 rows show

from the 200-row sample

Low (severity level) stands out: mean customer_age is 8.3, against 48.3 for the rest.

  • 16%fraud_suspected = true
  • 44median customer_age
  • 4customer genders
  • 5claim statuses
  • 6policy types
  • 48customer postal codes
Mean customer_age by severity_level191 rows
040808.3low23 rows27.1medium68 rows57.9high54 rows68.3critical46 rows
customer_age191 rows, in bands of 20
030603851474717060120customer_age →

Median 44, from 0 to 120.

incident_type200 rows · 6 values
  1. accident57
  2. natural_disaster39
  3. fire37
  4. liability30
  5. other30
  6. theft7
18 columns by typefrom the column list below
  • string 10
  • integer 1
  • float 2
  • date 4
  • boolean 1

Columns

18 columns in four groups
blueprint · 18 columns
columntypedescriptionexample
Text 10 columns
claim_idstringUnique identifier for each insurance claimuniqueCLM-2024-AUTO-001
policy_idstringUnique identifier for the insurance policy associated with the claimP-2021-AUTO-001
customer_idstringUnique identifier for the customer making the claimC-9011-M28
incident_typestringType of incident leading to the claim (e.g., theft, accident, fire)6 valuesaccident
claim_statusstringCurrent status of the claim (e.g., open, closed, pending, denied)open · closed · pending · denied · under_reviewopen
severity_levelstringSeverity rating of the claim for risk modelinglow · medium · high · criticalmedium
policy_typestringType of insurance policy (e.g., auto, home, life, health)6 valuesauto
customer_genderstringGender of the customermale · female · other · prefer_not_to_say · optionalmale
customer_postal_codestringPostal code of the customer's addressoptional94105
adjuster_notesstringFree-text notes from the claim adjusteroptionalCustomer reported collisi…
Numbers 3 columns
claim_amountfloatTotal monetary amount claimed by the customer0 or more5420.75
approved_amountfloatAmount approved by the insurer for payout0 or more · optional120000
customer_ageintegerAge of the customer at the time of claim0 to 120 · optional28
Dates and times 4 columns
claim_datedateDate when the claim was filed2024-03-18
incident_datedateDate when the incident occurred2024-03-17
policy_start_datedateStart date of the insurance policyoptional2021-03-01
policy_end_datedateEnd date of the insurance policyoptional2025-03-01
True or false 1 column
fraud_suspectedbooleanIndicates if the claim is suspected to be fraudulentoptionalfalse

Use it for

  • fraud suspected16%32 of 200 rowsmean customer age by …8.3low27.1medi…57.9high68.3crit…

    An insurance dashboard

    The fraud_suspected rate, customer_age by severity_level and a breakdown of incident_type. Excel, Power BI or Tableau.

  • Why do the 23 low rows have a mean customer_age of 8.3?

    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 claims with policy_id, customer_id and claim_date 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 columnscustomer_postal_code: UK only

blueprint · insurance-claims-severity-analysis

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
  • Claim incident type categorized
  • Severity rated on scale
  • Claim value provided
  • Settlement timeline recorded
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
insurance-claims-severity-analysis

What should your data show?

Preview 20 rows free
No signup. No card.