/

Blogue

Product Interaction Events

Product Interaction Events

Assurez une cohérence irréprochable entre tous vos produits de données

Assurez une cohérence irréprochable entre tous vos produits de données

Par Minseo (Len) Park

Ensure data from product interaction events is fresh, complete, and reliable before it is used for behavioral analytics, user journey analysis, and product performance measurement.

Data contract description

This data contract is designed for Glassdoor, where high-volume product events such as job_viewed, apply_clicked, review_submitted, comment_posted, and search_performed help the company understand user behavior across jobs, reviews, and workplace conversations on its platform. By putting a one-hour freshness expectation on ingest_time and governing key interaction context such as session, platform, entity, and action outcome, the contract supports Glassdoor’s shift-left approach to data quality by making product telemetry more dependable at the point of production, reducing downstream breakage, and preserving trust in the event data that powers analytics at scale.

Dataset

Defines the dataset this contract applies to, including its source and scope. This is the single point of reference for what data is being validated and protected.

Dataset checks

Validations that apply to the dataset as a whole, including schema stability, data freshness, and overall structural guarantees. This contract includes a schema check that prevents unexpected columns and unexpected column reordering, a row count check to ensure the dataset is not empty, and a freshness check that requires data to be less than one hour old based on ingest_time.

Column checks

Column-level rules to ensure completeness, correctness, and consistency for individual fields. This contract enforces that event_id, event_name, event_time, ingest_time, session_id, entity_type, entity_id, and action_result are not missing, and it requires event_id to be unique. It restricts event_name to approved product events, platform to approved platform values, entity_type to approved entity categories, and action_result to approved outcomes. It also applies length guardrails to key identifiers and version fields, so malformed values are caught early. Also, user_id, anonymous_id, platform, and app_version are constrained when present, but this contract does not require them to be populated.

product_interaction_data_contract.yaml

dataset
checks:
  - schema:
      allow_extra_columns: false
      allow_other_column_order: false
  - row_count:
      threshold:
        must_be_greater_than: 0
  - freshness:
      column: ingest_time
      threshold:
        unit: hour
        must_be_less_than: 1
columns:
  - name: event_id
    data_type: string
    checks:
      - missing:
      - duplicate:
      - invalid:
          name: "Event ID length guardrail"
          valid_min_length: 1
          valid_max_length: 128
  - name: event_name
    data_type: string
    checks:
      - missing:
      - invalid:
          name: "Valid event names"
          valid_values:
            - job_viewed
            - apply_clicked
            - review_submitted
            - comment_posted
            - search_performed
  - name: event_time
    data_type: timestamp
    checks:
      - missing:
  - name: ingest_time
    data_type: timestamp
    checks:
      - missing:
  - name: user_id
    data_type: string
    checks:
      - invalid:
          name: "User ID length guardrail"
          valid_min_length: 1
          valid_max_length: 128
  - name: anonymous_id
    data_type: string
    checks:
      - invalid:
          name: "Anonymous ID length guardrail"
          valid_min_length: 1
          valid_max_length: 128
  - name: session_id
    data_type: string
    checks:
      - missing:
      - invalid:
          name: "Session ID length guardrail"
          valid_min_length: 1
          valid_max_length: 128
  - name: platform
    data_type: string
    checks:
      - invalid:
          name: "Valid platforms"
          valid_values:
            - web
            - ios
            - android
            - backend
  - name: app_version
    data_type: string
    checks:
      - invalid:
          name: "App version length guardrail"
          valid_min_length: 1
          valid_max_length: 64
  - name: entity_type
    data_type: string
    checks:
      - missing:
      - invalid:
          name: "Valid entity types"
          valid_values:
            - job
            - company
            - review
            - post
            - comment
            - message
  - name: entity_id
    data_type: string
    checks:
      - missing:
      - invalid:
          name: "Entity ID length guardrail"
          valid_min_length: 1
          valid_max_length: 128
  - name: action_result
    data_type: string
    checks:
      - missing:
      - invalid:
          name: "Valid action results"
          valid_values

Data contract description

This data contract is designed for Glassdoor, where high-volume product events such as job_viewed, apply_clicked, review_submitted, comment_posted, and search_performed help the company understand user behavior across jobs, reviews, and workplace conversations on its platform. By putting a one-hour freshness expectation on ingest_time and governing key interaction context such as session, platform, entity, and action outcome, the contract supports Glassdoor’s shift-left approach to data quality by making product telemetry more dependable at the point of production, reducing downstream breakage, and preserving trust in the event data that powers analytics at scale.

Dataset

Defines the dataset this contract applies to, including its source and scope. This is the single point of reference for what data is being validated and protected.

Dataset checks

Validations that apply to the dataset as a whole, including schema stability, data freshness, and overall structural guarantees. This contract includes a schema check that prevents unexpected columns and unexpected column reordering, a row count check to ensure the dataset is not empty, and a freshness check that requires data to be less than one hour old based on ingest_time.

Column checks

Column-level rules to ensure completeness, correctness, and consistency for individual fields. This contract enforces that event_id, event_name, event_time, ingest_time, session_id, entity_type, entity_id, and action_result are not missing, and it requires event_id to be unique. It restricts event_name to approved product events, platform to approved platform values, entity_type to approved entity categories, and action_result to approved outcomes. It also applies length guardrails to key identifiers and version fields, so malformed values are caught early. Also, user_id, anonymous_id, platform, and app_version are constrained when present, but this contract does not require them to be populated.

product_interaction_data_contract.yaml

dataset
checks:
  - schema:
      allow_extra_columns: false
      allow_other_column_order: false
  - row_count:
      threshold:
        must_be_greater_than: 0
  - freshness:
      column: ingest_time
      threshold:
        unit: hour
        must_be_less_than: 1
columns:
  - name: event_id
    data_type: string
    checks:
      - missing:
      - duplicate:
      - invalid:
          name: "Event ID length guardrail"
          valid_min_length: 1
          valid_max_length: 128
  - name: event_name
    data_type: string
    checks:
      - missing:
      - invalid:
          name: "Valid event names"
          valid_values:
            - job_viewed
            - apply_clicked
            - review_submitted
            - comment_posted
            - search_performed
  - name: event_time
    data_type: timestamp
    checks:
      - missing:
  - name: ingest_time
    data_type: timestamp
    checks:
      - missing:
  - name: user_id
    data_type: string
    checks:
      - invalid:
          name: "User ID length guardrail"
          valid_min_length: 1
          valid_max_length: 128
  - name: anonymous_id
    data_type: string
    checks:
      - invalid:
          name: "Anonymous ID length guardrail"
          valid_min_length: 1
          valid_max_length: 128
  - name: session_id
    data_type: string
    checks:
      - missing:
      - invalid:
          name: "Session ID length guardrail"
          valid_min_length: 1
          valid_max_length: 128
  - name: platform
    data_type: string
    checks:
      - invalid:
          name: "Valid platforms"
          valid_values:
            - web
            - ios
            - android
            - backend
  - name: app_version
    data_type: string
    checks:
      - invalid:
          name: "App version length guardrail"
          valid_min_length: 1
          valid_max_length: 64
  - name: entity_type
    data_type: string
    checks:
      - missing:
      - invalid:
          name: "Valid entity types"
          valid_values:
            - job
            - company
            - review
            - post
            - comment
            - message
  - name: entity_id
    data_type: string
    checks:
      - missing:
      - invalid:
          name: "Entity ID length guardrail"
          valid_min_length: 1
          valid_max_length: 128
  - name: action_result
    data_type: string
    checks:
      - missing:
      - invalid:
          name: "Valid action results"
          valid_values

Data contract description

This data contract is designed for Glassdoor, where high-volume product events such as job_viewed, apply_clicked, review_submitted, comment_posted, and search_performed help the company understand user behavior across jobs, reviews, and workplace conversations on its platform. By putting a one-hour freshness expectation on ingest_time and governing key interaction context such as session, platform, entity, and action outcome, the contract supports Glassdoor’s shift-left approach to data quality by making product telemetry more dependable at the point of production, reducing downstream breakage, and preserving trust in the event data that powers analytics at scale.

product_interaction_data_contract.yaml

dataset
checks:
  - schema:
      allow_extra_columns: false
      allow_other_column_order: false
  - row_count:
      threshold:
        must_be_greater_than: 0
  - freshness:
      column: ingest_time
      threshold:
        unit: hour
        must_be_less_than: 1
columns:
  - name: event_id
    data_type: string
    checks:
      - missing:
      - duplicate:
      - invalid:
          name: "Event ID length guardrail"
          valid_min_length: 1
          valid_max_length: 128
  - name: event_name
    data_type: string
    checks:
      - missing:
      - invalid:
          name: "Valid event names"
          valid_values:
            - job_viewed
            - apply_clicked
            - review_submitted
            - comment_posted
            - search_performed
  - name: event_time
    data_type: timestamp
    checks:
      - missing:
  - name: ingest_time
    data_type: timestamp
    checks:
      - missing:
  - name: user_id
    data_type: string
    checks:
      - invalid:
          name: "User ID length guardrail"
          valid_min_length: 1
          valid_max_length: 128
  - name: anonymous_id
    data_type: string
    checks:
      - invalid:
          name: "Anonymous ID length guardrail"
          valid_min_length: 1
          valid_max_length: 128
  - name: session_id
    data_type: string
    checks:
      - missing:
      - invalid:
          name: "Session ID length guardrail"
          valid_min_length: 1
          valid_max_length: 128
  - name: platform
    data_type: string
    checks:
      - invalid:
          name: "Valid platforms"
          valid_values:
            - web
            - ios
            - android
            - backend
  - name: app_version
    data_type: string
    checks:
      - invalid:
          name: "App version length guardrail"
          valid_min_length: 1
          valid_max_length: 64
  - name: entity_type
    data_type: string
    checks:
      - missing:
      - invalid:
          name: "Valid entity types"
          valid_values:
            - job
            - company
            - review
            - post
            - comment
            - message
  - name: entity_id
    data_type: string
    checks:
      - missing:
      - invalid:
          name: "Entity ID length guardrail"
          valid_min_length: 1
          valid_max_length: 128
  - name: action_result
    data_type: string
    checks:
      - missing:
      - invalid:
          name: "Valid action results"
          valid_values

How to Enforce Data Contracts with Soda

Embed data quality through data contracts at any point in your pipeline.

Embed data quality through data contracts at any point in your pipeline.

# pip install soda-{data source} for other data sources

# pip install soda-{data source} for other data sources

pip install soda-postgres

pip install soda-postgres

# verify the contract locally against a data source

# verify the contract locally against a data source

soda contract verify -c contract.yml -ds ds_config.yml

soda contract verify -c contract.yml -ds ds_config.yml

# publish and schedule the contract with Soda Cloud

# publish and schedule the contract with Soda Cloud

soda contract publish -c contract.yml -sc sc_config.yml

soda contract publish -c contract.yml -sc sc_config.yml

Check out the CLI documentation to learn more.

Check out the CLI documentation to learn more.

How to Automatically Create Data Contracts.
In one Click.

Automatically write and publish data contracts using Soda's AI-powered data contract copilot.

Qualité des données IA basée sur la recherche

Nos recherches ont été publiées dans des revues et conférences de renom, telles que NeurIPs, JAIR et ACML. Les mêmes lieux qui ont fait progresser les fondations de GPT et de l'IA moderne.

Demande une démo

Explore more data contract templates

One new data contract template every day, across industries and use cases

Demande une démo

Produit

Solutions

Tarifs

Templates

Blogue

Log in

Demande une démo

4,4 sur 5

Commencez à faire confiance à vos données. Aujourd'hui.

Trouvez, comprenez et corrigez tout problème de qualité des données en quelques secondes.
Du niveau de la table au niveau des enregistrements.

Demande une démo

Adopté par

ⓒ 2025 Soda Data Inc. Tous droits réservés.

|

Conditions Générales

|

Politique de Confidentialité

AICPA SOC logo
Giga OM logo
TW logo
Gartner logo
High Performer 2025 badge
Best Roi 2025 badge
Momentum Leader 2025 badge

4,4 sur 5

Commencez à faire confiance à vos données. Aujourd'hui.

Trouvez, comprenez et corrigez tout problème de qualité des données en quelques secondes.
Du niveau de la table au niveau des enregistrements.

Demande une démo

Adopté par

ⓒ 2025 Soda Data Inc. Tous droits réservés.

|

Conditions Générales

|

Politique de Confidentialité

AICPA SOC logo
Giga OM logo
TW logo
Gartner logo
High Performer 2025 badge
Best Roi 2025 badge
Momentum Leader 2025 badge

4,4 sur 5

Commencez à faire confiance à vos données. Aujourd'hui.

Trouvez, comprenez et corrigez tout problème de qualité des données en quelques secondes.
Du niveau de la table au niveau des enregistrements.

Demande une démo

Adopté par

Qualité des données native AI

Product

Soda AI

Observabilité des données

Contrats de Données Collaboratifs

Alertes Intelligentes

Diagnostics au Niveau des Enregistrements

Traçabilité et impact

Entrepôt de diagnostics

Résolution automatisée au niveau des enregistrements

Réparation Automatisée des Pipelines

Solutions

Personas

Gouvernance des données

Architecte de Données

Consommateur de Données

Ingénieur Data

Responsable de la gouvernance des données

Cas d'utilisation

Opérationnaliser la Gouvernance des Données

Test de pipeline

Maillage de données

Migration vers le Cloud et Réconciliation

Conformité BCBS 239

Company

Tarifs

Opportunités de carrière

Boutique d'articles promotionnels

Boutique d'articles promotionnels

Ressources

Blogue

Documentation

Documentation

Webinaires

Événements

Notes de version

Slack

Contactez le Support

Blogue

Quelles sont les Dimensions de la Qualité des Données ?

Introduction à la Traçabilité des Données 101

Prévention des problèmes avec les Data Contracts

Observabilité et Tests des Données sur Databricks

ⓒ 2025 Soda Data Inc. Tous droits réservés.

Conditions Générales

Politique de Confidentialité

AICPA SOC logo
Giga OM logo
TW logo
Gartner logo
High Performer 2025 badge
Best Roi 2025 badge
Momentum Leader 2025 badge