Social Media Moderation Workflow Metrics

This dataset provides granular metrics on social media content moderation workflows, capturing task assignment, moderator details, decision outcomes, compliance flags, and quality review status for both human and AI-driven moderation teams. Designed for operational leads, it enables deep analysis of workflow efficiency, resource allocation, and policy compliance, supporting transparency and optimization in remote and hybrid moderation environments.

  • last updated 16 Dec 2025
  • by GoMask
The brief that made it

Monitor and optimize moderation workflow efficiency and staffing allocation

Sample rows

preview · 8 of 65 rows · all 24 columns
moderation_task_idstringdecisionstringai_confidence_scorefloatpolicy_violatedbooleanmoderator_teamstringplatform_namestringcontent_idstringcontent_typestringreported_reasonstringreport_sourcestringmoderator_idstringmoderator_typestringmoderator_location_countrystringtask_assigned_datetimedatetimetask_started_datetimedatetimetask_completed_datetimedatetimehandling_time_secondsintegerescalation_levelintegerpolicy_categorystringcompliance_flagbooleanquality_review_requiredbooleanquality_review_completedbooleanquality_review_outcomestringnotesstring
MT10001removedblanktruein-houseFaceSpaceCID2001posthate speechuserMOD_A_001humanUS2024-05-14T09:22:15Z2024-05-14T09:24:52Z2024-05-14T09:29:11Z2590hate speechtruetruetruepassClear violation; user warned.
MT10002removed0.95trueblankTwiddleCID2002commentspamautomatedAI_SYS_01aiblank2024-05-14T10:05:31Z2024-05-14T10:05:33Z2024-05-14T10:05:35Z20spamtruefalseblankn/aAI flagged mass advertising link.
MT10003removed0.74trueEMEAInstaPicCID2012imagenudityuserMOD_C_021humanDE2024-05-13T13:11:44Z2024-05-13T13:13:10Z2024-05-13T13:17:08Z2380nudityfalsetruefalsen/aPolicy violation; flagged for review.
MT10004escalated0.63falseAPACPicSnapCID2020videomisinformationexternal_partnerMOD_B_047hybridIN2024-05-13T17:26:12Z2024-05-13T17:28:00Z2024-05-13T17:39:48Z7082blanktruefalseblankn/aUnclear policy; sent to legal.
MT10005removed0.98trueblankTwiddleCID2002commentspamautomatedAI_SYS_02aiblank2024-05-14T10:07:10Z2024-05-14T10:07:12Z2024-05-14T10:07:13Z10spamtruefalseblankn/aRepeat offender; content deleted.
MT10006approvedblankfalseEMEAFaceSpaceCID2017messageharassmentuserMOD_A_021humanUK2024-05-15T11:02:19Z2024-05-15T11:05:30Z2024-05-15T11:09:22Z2320blanktruetruetruepassMessage context checked; no violation.
MT10007removed0.81trueEMEAInstaPicCID2013imagenudityuserMOD_C_022humanIT2024-05-12T09:13:32Z2024-05-12T09:14:36Z2024-05-12T09:18:21Z2250nuditytruetruefalsefailDisputed by user; pending appeal.
MT10008escalated0.55falseAPACTwiddleCID2018othercopyrightinternal_reviewMOD_B_051hybridPH2024-05-15T17:55:01Z2024-05-15T17:56:40Z2024-05-15T18:02:13Z3331blanktruefalseblankn/aCopyright unclear; sent to legal.

What the 65 rows show

from the 65-row sample

Removed decision stands out: 51 of its 51 rows have policy_violated = true, against 3 of 14 for the rest.

  • 83%policy_violated = true
  • 0.86median ai_confidence_score
  • 3moderator types
  • 3quality review outcomes
  • 5report sources
  • 6content types
Policy violated rate by decisionpolicy_violated = true
0%50%100%0%approved0 of 6100%removed51 of 5150%escalated3 of 60%no_action0 of 2
ai_confidence_score36 rows, in bands of 0.1
081612468150.40.71ai_confidence_score →

Median 0.86, from 0.49 to 0.99.

moderator_team44 rows with a value · 21 left blank
  1. EMEA16
  2. contractor11
  3. APAC10
  4. in-house6
  5. US1
24 columns by typefrom the column list below
  • string 14
  • integer 2
  • float 1
  • datetime 3
  • boolean 4

Columns

24 columns in four groups
blueprint · 24 columns
columntypedescriptionexample
Text 14 columns
moderation_task_idstringUnique identifier for each moderation task or case.uniqueMT10001
platform_namestringName of the social media platform where the moderation task originated.8 namesFaceSpace
content_idstringIdentifier for the content (post, comment, image, etc.) under review.CID2001
content_typestringType of content being moderated (e.g., post, comment, image, video, message).6 valuespost
reported_reasonstringPrimary reason the content was flagged for moderation (e.g., hate speech, spam, nudity, misinformation).9 reasons · optionalhate speech
report_sourcestringSource of the report (user, automated system, external partner, etc.).5 values · optionaluser
moderator_idstringUnique identifier for the moderator (human or AI system) assigned to the task.MOD_A_001
moderator_typestringType of moderator handling the task (human, ai, hybrid).human · ai · hybridhuman
moderator_location_countrystringCountry where the moderator is located (for compliance and resource allocation).optionalUS
moderator_teamstringTeam or group to which the moderator belongs (e.g., APAC, EMEA, US, contractor, in-house).5 teams · optionalin-house
decisionstringOutcome of the moderation task (approved, removed, escalated, no action, etc.).approved · removed · escalated · no_action · otherremoved
policy_categorystringCategory of policy violated (e.g., hate speech, nudity, spam, misinformation).6 categories · optionalhate speech
quality_review_outcomestringOutcome of the quality review (pass, fail, n/a).pass · fail · n/a · optionalpass
notesstringAdditional notes or comments from the moderator or reviewer.optionalMultiple links detected.
Numbers 3 columns
handling_time_secondsintegerTotal time in seconds taken to complete the moderation task (from start to completion).0 or more · optional259
escalation_levelintegerLevel of escalation if the task was escalated (0 = not escalated, 1 = first level, etc.).0 or more · optional0
ai_confidence_scorefloatConfidence score assigned by AI (if applicable), between 0 and 1.0 to 1 · optional0.95
Dates and times 3 columns
task_assigned_datetimedatetimeTimestamp when the moderation task was assigned.2024-05-14T09:22:15Z
task_started_datetimedatetimeTimestamp when the moderator started working on the task.optional2024-05-14T09:24:52Z
task_completed_datetimedatetimeTimestamp when the moderation task was completed.optional2024-05-14T09:29:11Z
True or false 4 columns
policy_violatedbooleanIndicates if the content was found to violate platform policy.true
compliance_flagbooleanIndicates if the moderation process complied with internal/external regulations (e.g., GDPR, local laws).optionaltrue
quality_review_requiredbooleanIndicates if this moderation task requires a secondary quality review.optionaltrue
quality_review_completedbooleanIndicates if the quality review has been completed.optionaltrue

Use it for

  • policy violated83%54 of 65 rowspolicy violated by de…0%appr…100%remo…50%esca…0%no_a…

    A social media dashboard

    The policy_violated rate, ai_confidence_score by decision and a breakdown of moderator_team. Excel, Power BI or Tableau.

  • Why do 54 of 65 rows have policy_violated = true?

    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 moderation tasks with platform_name, content_id and content_type to fill a screen in front of a buyer.

Not quite right?

Make it yours.

Same 24 columns, your size and your rules. See 20 rows before you pay.

Preview 20 rows free

10,000 rows of yours: $12.99One-time. No subscription. All prices

This dataset65 rows24 columns
Yours10,000 rows24 columnsmoderator_location_country: UK only

blueprint · social-media-moderation-workflow-metrics

Behind this dataset

Same schema. As many rows as you need.

These 65 rows came out of a blueprint — 24 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 row reflects a distinct moderation task and its outcome
  • Includes AI-assisted and human-only actions distinctly
  • Tracks intervention time, escalation frequency, and automated decision override
  • Flags tasks involving regulatory compliance or flagged sensitive content
  • Quantifies resource utilization per moderation workflow
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
social-media-moderation-workflow-metrics

More in Social Media

Browse all
Also in Social MediaRemote Moderator Productivity MetricsSocial Media Influencer Metrics StreamsGeotagged Social Sentiment DataAll 1,947 datasets

What should your data show?

Preview 20 rows free
No signup. No card.