Literature review
AuraScore 81/100

Streaming Retention Algorithmic Research Protocol Plan

Develop an exhaustive academic literature review plan evaluating recommendation algorithms and streaming subscriber churn dynamics.

Use this template when planning an end-to-end systematic literature review on user engagement, churn modeling, and recommender systems across digital broadcast and streaming platforms. It guides senior audience researchers in structuring query protocols, thematic extraction, and synthesis roadmaps.

Template

Role: Principal Media Data Scientist and Audience Research Director with twenty years of experience in entertainment consumption modeling.

Context

  • Target Platform Ecosystem: {{streaming_platform}}
  • Primary Research Question: {{churn_hypothesis}}
  • Target Audience Cohort: {{target_demographic}}
  • Algorithmic Architectures Under Review: {{algorithmic_models}}
  • Academic Indexing Window: {{publication_window}}
  • Primary Commercial Metric: {{key_business_kpis}}

Task

Produce a systematic literature review project plan that evaluates empirical studies on recommendation systems and predictive audience attrition for {{streaming_platform}}. The resulting plan must deliver an actionable roadmap to synthesize behavioral science, algorithmic design, and churn mitigation strategies against {{key_business_kpis}}.

Method

  1. Deconstruct {{churn_hypothesis}} into boolean search strings across peer-reviewed databases (ACM, IEEE, JMM, Communication Research).
  2. Formulate inclusion and exclusion criteria tailored to {{publication_window}} and {{target_demographic}} consumption patterns.
  3. Establish a multi-coder screening protocol to classify papers by methodology, algorithmic paradigm ({{algorithmic_models}}), and validity.
  4. Design a standardized data extraction matrix to capture sample sizes, neural architectures, behavioral predictors, and metric impact.
  5. Define a quality assessment framework assessing statistical rigor, platform bias, and reproducibility in streaming contexts.
  6. Structure a narrative and meta-analytic synthesis methodology comparing algorithmic performance across churn intervention stages.
  7. Map literature findings directly to commercial intervention milestones affecting {{key_business_kpis}}.
  8. Establish risk controls for publication lag, conflicting empirical findings, and proprietary dataset divergence.

Constraints

  • MUST cite specific empirical metrics (e.g., NDCG, MRR, Hazard Ratios) in the data extraction criteria.
  • MUST NOT include non-peer-reviewed whitepapers or unverified industry blog posts.
  • All database search queries MUST be fully articulated with field tags and boolean operators.
  • Timeline milestones MUST allocate defined buffer periods for inter-rater reliability calibration.

Output format

Provide the review plan structured under five mandatory headings:

  1. Search Protocol and Query Architecture (exact boolean strings for 4 distinct databases)
  2. Inclusion, Exclusion, and Quality Screening Criteria (tabular layout with 6 parameters)
  3. Thematic Extraction Matrix (field definitions and synthesis categories)
  4. Synthesis and Meta-Analysis Roadmap (4 sequential analytical phases)
  5. Operational Review Timeline and KPI Governance (gantt-style milestone table) Total length must be between 900 and 1400 words.

Self-review

  • Ensure search queries specifically address both {{algorithmic_models}} and {{target_demographic}}.
  • Confirm extraction parameters track direct correlations to {{key_business_kpis}}.
  • Verify all four constraints and exact section titles are strictly preserved.
AuraScore breakdown
81/100Provisional
Instruction clarity15/15 · Strong

Explicit role, a named task, and discrete steps the model can follow.

Context architecture12/12 · Strong

Background, inputs and variables the model needs before it starts.

Constraint engineering12/12 · Strong

Hard boundaries — what the model must and must not do.

Output specification6/14 · Thin

A named, field-level shape for the response.

Reasoning structure10/10 · Strong

Ordered work items that force analysis before an answer.

Model compatibility10/10 · Strong

Length and structure that travel across frontier models.

Token efficiency5/10 · Thin

Signal density — instruction weight without padding.

Reusability7/7 · Strong

Documented variables so the scaffold adapts to new inputs.

Robustness3/5 · Adequate

Quality bar, assumptions and behaviour when inputs are thin.

Observed performance1/5 · Thin

How much real usage the template has behind it.

research-analysis
research-literature
media-entertainment
streaming
literature-review
audience-analytics