The Challenge
The Wyland Foundation needed to transform their 14-year-old water conservation campaign from a simple pledge platform into an engaging, personalized experience that would drive meaningful behavior change at scale across thousands of U.S. cities.
Since 2011, the Wyland National Mayor's Challenge for Water Conservation has helped over 1,000 U.S. mayors inspire residents to take small, sustainable actions. However, by 2024, the foundation faced a critical question: How could they help participants see the tangible impact of their individual actions?
Abstract Impact
Participants struggled to understand how their personal pledges translated into real environmental benefits for their communities.
Generic Messaging
One-size-fits-all content didn't resonate with participants' local contexts across diverse regions from drought-stricken Arizona to flood-prone New Jersey.
Engagement Gap
Without personalized feedback, maintaining long-term commitment beyond the initial pledge was difficult.
Scale Challenge
Serving thousands of cities with relevant, localized content manually was operationally impossible.
Our Solution
DoozerAI designed and implemented a secure web service that automatically curates environmental content, matches user pledges with relevant local stories using ZIP code data, and delivers personalized impact narratives showing how individual actions benefit specific communities. For the 2026 Challenge, DoozerAI layered on two new AI-generated deliverables for every pledge confirmation: a custom-voiced audio news story and a retrieval-augmented, LLM-written impact report — generated concurrently and returned as ready-to-embed content with zero client-side API calls.
Leveraged DoozerAI's existing Hunter digital worker infrastructure (processing 1,000+ stories weekly) to build an intelligent content engine with AI-powered categorization, geographic matching, and real-time personalized delivery through seamless integration. On top of that foundation, a new single-pass service pairs Azure neural text-to-speech with a Qdrant-backed retrieval pipeline: it synthesizes a two-voice 'breaking news' segment about the participant's pledge, mixes it with background music using pure-Python audio processing (no ffmpeg, so it runs in restricted sandboxes), and simultaneously has an LLM write a localized impact report grounded in real local water, energy, and infrastructure context — all rendered server-side into the final page in one pass.
AI-Narrated Personal News Story
Every pledge triggers a custom 'breaking news' segment voiced by two distinct AI voices trading lines about the participant's name, city, gallons pledged, and dollar savings — mixed live with background music into a ready-to-play audio clip, generated in seconds and cached automatically with no studio or manual editing required.
Retrieval-Augmented Impact Reports
A vector search pipeline embeds each pledge and retrieves the most relevant local context from a state-filtered knowledge base, then an LLM writes a structured, personalized report connecting the participant's specific pledge to local infrastructure, drought conditions, and household cost savings — no two reports read the same.
Automated Content Curation
Leveraged DoozerAI's existing news collection infrastructure processing 1,000+ stories weekly with AI-powered analysis to categorize and tag environmental content, building a robust pipeline for water conservation topics.
Intelligent Personalization Engine
Matched user pledges (water, plastic, energy) with relevant local stories using ZIP code data to deliver hyperlocal environmental news with inspiring synopses tailored to each conservation commitment.
Impact Quantification
Calculated specific benefits—water saved, energy conserved, CO₂ reduced—translating abstract concepts into tangible metrics and connecting individual actions to community-wide environmental improvements.
Scalable, Real-Time Architecture
Secure, scalable design with enterprise-grade authentication, handling up to 100 requests per minute with built-in error handling and standardized data format for seamless website integration. The voice narration and impact report are generated concurrently and returned in a single response for a fast, seamless pledge confirmation experience.
Results & Impact
The 2026 Challenge featuring DoozerAI's Impact Insight Initiative achieved record-breaking results. Over 2,800 cities participated across 23 states, with residents pledging to save 559 million gallons of water annually—representing $7 million in utility savings, 2.2 billion pounds of CO₂ eliminated, and massive reductions in waste and pollution.
"By incorporating DoozerAI, we're providing participants with a powerful tool to see the direct connection between their actions and the health of their communities. When residents see how their actions impact everything from local water quality to the cost of living, it motivates them to stay engaged and do even more for their communities."
Project Gallery
