Context
The NGO needed to compare environmental narratives on Facebook and Instagram to guide communications investments.
Approach
LLM methodology (ChatGPT + tailored prompts) scoring tone/topics/perspectives, correlated with Facebook/Instagram KPIs by audience type (age, country, interests) and yielding comparative visuals.
Impact
Budget reallocation per country/platform and an internalized analysis framework.
Posts analyzed
60 k
Languages
8
Visualizations
50+
LLMOpenAINLPTypeScript
What we delivered
- Prompts capturing tone, actors, topics and emotions on Facebook/Instagram.
- Automated pipeline: social data → LLM → scoring of topics/tones/perspectives per audience type.
- Cross-analysis with post performance to recommend priority themes.
- Documented methodology replicable by local teams.
- Enablement sessions to interpret recommendations and adjust budgets.
Want to explore a similar use case?
Let's align on your next data, app or AI priority and draft an actionable roadmap in days.