# UNDP BiH | Envisioning

Source: https://www.envisioning.com/services/impact/undp-bih

---

# Scanning the Green Transition in Bosnia and Herzegovina

Foresight radars for UNDP's five-year Green Transition programme in Bosnia and Herzegovina, surfacing citation-backed signals of change across decarbonisation, circularity and environmental justice to feed directly into portfolio reviews and programming decisions.

2025

[Back to impact](/services/impact)

![UNDP BiH](/_next/image?url=https%3A%2F%2Fres.cloudinary.com%2Fenvisioning%2Fimage%2Fupload%2Fv1772722384%2Fgyilxs9msfsrcxyyj008.png&w=3840&q=75)

Envisioning partnered with the [United Nations Development Programme](https://www.undp.org/) to build a horizon scanning capability for their five-year programme **"Catalysing the Green Transition in Bosnia and Herzegovina"**, which aims to enable a just shift to a green, zero-carbon economy.

The programme adopts a portfolio approach across five pillars — decarbonisation, depollution, nature and biodiversity, circularity, and environmental justice — and is underpinned by continuous learning through Monitoring, Evaluation and Learning (MEL) and Dynamic Portfolio Management (DPM). Within this context, Envisioning's **Signals platform** was used to establish a lightweight, repeatable foresight-to-MEL workflow: surface signals, organise them around the portfolio's pillars, and feed intelligence into regular sensemaking and programming decisions.

The delivery — two interactive radars, full data exports and an AI assistant — is designed to outlast the engagement itself. UNDP teams can reuse the datasets, run follow-up scans at set intervals, and embed the radar directly into quarterly portfolio reviews, closing the loop between foresight and action.

### Our Work

-   Built **2 purpose-built interactive radars** (Green Transition and Circular Economy) containing citation-backed signals of change.
-   Applied a **6-stage AI-enhanced methodology**: ensemble generation across **5 frontier models**, semantic deduplication, impact assessment, narrative synthesis, source verification, and interactive visualisation.
-   Orchestrated **5 AI models in parallel** (Gemini 3 Pro, GPT-5 Mini, Claude Sonnet 4.5, Mistral Large, Sonar Pro) to increase signal diversity and reduce single-model bias.
-   Integrated a **5-level verification system** (Grounded → Fabricated) with linked external sources for every signal.
-   Delivered **full downloadable datasets** and an **AI assistant** for natural-language exploration of the signal data.
-   Designed the workflow to plug directly into UNDP's existing **MEL and Dynamic Portfolio Management** processes.

### Impact

-   Established a **repeatable foresight-to-MEL workflow** embedded in a 5-year programme — foresight outputs feed directly into quarterly portfolio reviews and programming decisions.
-   Covered **5 portfolio pillars** (decarbonisation, depollution, nature/biodiversity, circularity, environmental justice) in a single scanning engagement.
-   Delivered within a **6-week timeline**, including model configuration, results refinement, and 2 stakeholder webinars.
-   Provided **6 months of hosted access** to interactive radars for ongoing use.
-   Builds on prior UNDP collaboration, demonstrating the **platform's adaptability** across different country contexts and thematic areas.

[Signals Work Session](/services/impact/services/signals-work-session)

Envisioning ran a series of working sessions with UNDP to configure and operationalize a repeatable horizon-scanning workflow in the Signals platform—aligning teams on how to generate, deduplicate, assess, and (where enabled) verify signals, then organize them into interactive radars that feed into regular sensemaking, portfolio reviews, and programming decisions.

This made my year!

A.D.

Head of Exploration, UNDP

![UNDP](/api/client-logos/232a0083-f131-4e7c-8ce3-96425f3beaea/svg?v=1cbl5dq)

## Interested in our work?

Reach out to learn how we can help you with your foresight needs.

Contact Envisioning

## Newsletter

Follow us for weekly foresight in your inbox.

Sign up

Browse the latest from Artificial Insights, our opinionated weekly briefing exploring the transition toward [AGI](https://envisioning.ai/AGI+(Artificial+General+Intelligence)).

![Aug 24, 2026 · Issue 145](https://res.cloudinary.com/envisioning/image/upload/f_auto,q_auto:best/v1787597428/hzrsjpxn2jxag8k4ruuy.png)

Aug 24, 2026 · Issue 145

The Country of the Future

[/newsletter/145-the-country-of-the-future](/newsletter/145-the-country-of-the-future)

![Aug 10, 2026 · Issue 144](https://res.cloudinary.com/envisioning/image/upload/f_auto,q_auto:best/v1786371190/az6kgat0wnzvgf89ehsy.png)

Aug 10, 2026 · Issue 144

Indistinguishable from Magic

[/newsletter/144-indistinguishable-from-magic](/newsletter/144-indistinguishable-from-magic)

![Jul 27, 2026 · Issue 143](https://res.cloudinary.com/envisioning/image/upload/f_auto,q_auto:best/v1785141074/qexku3lbuvvmmz3tx6l2.png)

Jul 27, 2026 · Issue 143

This Is Fine

[/newsletter/143-this-is-fine](/newsletter/143-this-is-fine)

[View all issues](/newsletter)