adaptmap

* * AdaptMap * * is a German-language, mobile-first web application that helps cities understand and respond to heat stress at neighborhood level. It combines an easy-to-use citizen questionnaire, an anonymised public heatmap, and optional AI-generated advice for residents. The app is designed to work well as a ‘street-ready’ tool: people can a QR code or open a short link, complete a focused set of questions about their heat situation, and immediately see a simple result and further options.

​

# # # * * Primary Goals * *

— * * lower the threshold to reporting heat problems * *: People should be able to report their situation in under a few minutes, on a smartphone, without creating an account.
— * * created a privacy-respecting evidence base * *: Collected information is aggregated and geospatially Generalised so that individual households cannot be identified, while still giving planners and policymakers a clear picture of where heat stress is concentrated.
— * * offer immediate, practical value to respondents * *: After filling out the questionnaire, users receive a simple “heat problem index” plus the option to generate AI-based recommendations tailored to their answers and neighborhood.
— * * Empower non-technical staff * *: City staff and project teams can adjust wording, legal texts, informational pages, and knowledge-base entries through the CMS and admin dashboard instead of changing code.

# # # * * core User Groups * *

— * * residents/Citizens * *
People living in Köln who experience heat stress (e.g. in their apartment, on their street, in their neighborhood) and want to:

— Report their situation,
— Understand how ‘severe’ their situation is in simple terms, and
— Receive contextual tips and recommendations on how to cope or what to change.
— * * city administration & project teams * *
Staff responsible for climate adaptation, urban planning, or public health, who needs:

— A way to * * collect structured data * * that goes beyond ad-hoc complaints,
— Visual tools (heatmap, dashboards) to * * spot patterns and hotspots * *, and
— A maintainable * * knowledge base * * that feeds AI-generated advice.
— * * Researchers/partners * *
Researchers and project partners who analyse spatial patterns of heat stress and evaluate interventions using the provided data and tolling.

# # # * * High-Level User Journey * *

1. * * discover the app * *
Residents typically reach AdaptMap Köln by scanning a QR code on printed materials, seeing a city campaign, or following a shared link. They land on the home page, which briefly explains the purpose and shows a preview of the heatmap.
2. * * start the questionnaire * *
A clear call-to-action (“Fragebogen starten” or similar) invites them into a guided questionnaire flow. The questionnaire is split into steps with visual progress and section covers, so the process feels manageable and structured.
3. * * provide location and context * *
Early in the questionnaire, users specify where they are affected:

— Either by letting the browser determine their * * GPS location * * and automatically resolving it to a postal code and city.
— Or by manually entering an address (street, house number, postal code, city) with search assistance.
— This location information is then normalised and later aggregated; it is essential both for meaningful recommendations and for building the heatmap.

4. * * answer targeted questions * *
Users answer a small number of questions about:

How often and how strongly they experience heat problems,
— What their living situation is like (e.g. building type, Greenery, shading),
— How they perceive their neighborhood in terms of heat,
— Optionally, some basic demographic information such as age, gender, and household size.
— The app uses different input components-sliders, multiple choice cards, icon-based selections, etc. – to keep the interaction intuitive and mobile-friendly.

5. * * optional free-text feedback and consent * *
At the end of the questionnaire, users can add an open-ended comment describing their situation or suggestions (‘Was wünschen Sie sich, was sich ändern sounding?’). At the same time, they must accept that their data will be collected and processed anonymously for analysis purposes. Legal text and links to privacy policy and terms are prominently available.
6. * * submission and results * *
After submitting, the backend calculates a * * heat problem index * * between 0 and 100. The results page shows:

— A clearly labeled index with a color-coded severity category (low, medium, high),
— An explanatory text describing what the score means in everyday language,
Optional sub-scores for different dimensions (e.g. housing, outdoor environment).

7. * * AI-generated recommendations (optional) * *
Users can then choose to click a button to receive a * * “KI-Empfehlung” * *. Only when they click:

— Their anonymised submission is sent to an external AI workflow (via n8n),
— The AI returns a German summary plus a handful of concrete recommendations,
— These recommendations are shown in a structured card layout and stored for that submission so they can be revisited without re-calling the AI.

8. * * explore the public heatmap * *
From the results page or main navigation, users can open the * * heatmap * *. This map:

— Show a grid or postal-code-level Coloring based on averaged problem indices,
— Provides tooltips with statistics when hovering or tapping Tiles (number of reports, average index, distribution),
— Optionally highlights the user’s own approximate area (e.g. postal code), without exposing exact locations.

9. * * optionally restart or share * *
Users can start a new survey, share the link, or simply leave. No account is required at any point.