Putting Climate and Data Into Action

Published in
8 min readSep 9, 2020


Photo from Unsplash from Markus Spiske

It’s getting closer to the launch of our groundbreaking climate change analysis tool. Counting down, Quilt.AI has discussed:

We also travelled to Vietnam and Mexico, advising how you can be a more responsible traveler and leave a lower carbon footprint. How does this nuanced discussion on climate and data lead to action? By building a communications and advocacy strategy linked to both online and offline action in the climate space.

In today’s post, Quilt.AI takes you through each step of building such a strategy. September 7th was the first International Day for Clean Air. Consequently, the climate change challenge for strategy building is inciting action among city level populations on air pollution. The cities we touch on for our communications and advocacy strategy are San Jose (USA), New Delhi (India), Kampala (Uganda), Gaungzhou (China) and Tel Aviv (Israel).

As you embark on crafting your strategy, you will find our analysis on air pollution campaigns, in partnership with the Clean Air Fund, useful.


Each of the cities we have chosen are contextually and culturally different. This is important when it comes to observing sources of air pollution, patterns around its occurrence and communities’ intent to prevent and respond to it.

Quilt.AI developed an air pollution graph, showing PM 2.5 levels in the chosen cities during the past four years. What is observable right away is that some of the cities have spikes in air pollution during the same time each year.

Urbanization, rapid economic development and dense populations play a role in all of these cities. Beyond these factors,wildfires play a role in driving up air pollution levels in San Jose and Tel Aviv. Israel is currently supporting California in its most recent episode of wildfires. Guangzhou accumulates PM 2.5 pollutants during monsoon season, cool to mild months and when variations in wind direction occur.

In Kampala and New Delhi, air pollution levels are historically bad year-round, and have similar sources: transport, industry and waste burning. In New Delhi, seasonal increases occur during the winter months. Air pollution levels in both cities reduced during Covid-19 lockdowns, begging the question among climate activists and policymakers, whether any of these temporary gains can result in longer-term air pollution prevention and response measures.

Other factors that help set the context for each of these cities, include the policy landscape and historical climate data, carbon footprint and patterns in digital conversations on climate change.


After understanding the context, the next step in crafting our action-oriented strategy is analyzing past and ongoing air pollution and climate related initiatives, including campaigns, in these cities. This sets guiding principles for select best practices in behavior shifting communications and advocacy strategy designs.

First, it is useful to look at select country and city level initiatives that aim to incite community level action. In San Jose, the climate smart challenge and in Tel Aviv the imminent city level plan are two such examples. In Guangzhou, an active China Youth Climate Action Network (CYCAN) drives numerous campaigns on air pollution, focusing on how young people can impact sustainable and climate smart cities in the Chinese context. From a policy perspective, Uganda has a positive track record, including the implementation of a National Climate Change Policy in 2015, a Green Growth Strategy and commitment to the Bonn Challenge.

Second, it is instructive to formulate a set of guidelines that make air pollution campaigns more effective. During Quilt.AI’s air pollution campaign research in the India context, analyzing 21 million searches, 7000+ tweets, 5000+ images and 63 climate campaigns and organizations, we found some of the following best practices:

Timing and Usage of Words: The following air pollution-related words rise significantly between the months of October and January each year (based on a trend analysis from 2015–2019): “air pollution,” “causes of air pollution,” “air purifier,” “effects of pollution,” “pm 2.5,” “Diwali pollution,” and “anti-pollution mask.” These searches spike in volume during Diwali time.

The majority of campaigns we analysed, launch in May. However, based on search behaviour, it is more strategic to launch in October and finish by February in the new year. Country wide demographic analysis is important as there is material variation in search behavior across gender and age groups.

Finally, when designing a campaign, it is important to partner strategically with other organisations, in order to achieve the greatest impact and reach. Some of the following organisations are high in search interest and have left an impact with the work they have done: Indian Youth Climate Network (195%), Isha Outreach (200%), Mahindra Rise (18%), Ministry of Youth Affairs and Sports (58%), Niti Aayog (9%), Public Health Foundation of India (4%).


After collecting information on the city’s context and on previous and ongoing initiatives and campaigns — it is time to layer this climate information with nuanced data from Quilt.AI’s climate change analysis tool and create a strategy of action. Democratizing climate relevant data, this interactive dashboard focuses on three crucial indices:

  • Climate Variability Index (CVI) — takes into account various climate/ weather related parameters (e.g. flood, storm, temperature changes) that affect climate change and provides a risk factor in terms of the scale of climate change in the last few decades.
  • Socio Economic Risk Index (SERI) — This index is an indicator of how well a city can handle large scale climate change events by considering the economic status, population, social conditions and the level of income inequality in the society.
  • Search Deniers Index (SDI) — is built by understanding the search behavior of people online, the volume of these keywords and terms. It is based on how much of the searches are around climate change denial and how much of this search behavior translates into political support for climate change.

Looking at just two of the focus cities, the climate change analysis tool provides crucial information for an effective and action-oriented strategy. In San Jose, climate variability may be low, but it is seasonal (as we know from the air pollution graph). However, the socio-economic risk is medium as is the search denier score. The majority of San Jose’s population are UNACTIVATED ALLIES, meaning that they are aware of climate impact on their city (e.g. through rising air pollution during California wildfires) and already participate in some community level initiatives (such as the climate smart challenge), but their action at a city level, could be amplified.

The key action to integrate in our strategy is to highlight how existing community initiatives are changing the air pollution landscape. Additional city level initiatives with specific calls to action and tangible goals should be started in San Jose.

In New Delhi, we see a different picture. Socio-economic risk is high, with high population density, higher levels of income inequality, and factors such as caste and education leading to a larger scale of marginalized populations, who are more acutely affected by the city’s air pollution problem. Used to rising air pollution levels for years, many New Delhi residents consider climate change to be SOMEONE ELSE’S PROBLEM.

The key action to integrate in our strategy is to illustrate how air pollution reduced during the Covid-19 lockdown. Small actions such as rotational work from home schedules and taking public transport can reduce air pollution spikes in the future. Giving specific examples of Indian cities, such as Rajkot, Pune and Nagpur, that are using interventions to reduce climate change in their communities may also create heightened awareness and knowledge in Delhi.


Understanding the city level context, previous and ongoing initiatives and getting a nuanced view from the Quilt.AI climate change analysis tool gives us rich information to test messages in an online campaign. This includes targeting the different behaviour clusters with air pollution prevention and response messages, using three methods:

  1. Re-direction of search behavior to legitimate information on air pollution issues. For example, in Kampala, we know that a big cluster of the city level population is SKEPTICAL of climate change. Redirecting climate related searches to alarming but scientifically backed information is important.
  2. “Real-estate” takeover of social media timelines. This includes the usage of existing, repurposed content, which is dropped multiple times a day into someone’s timeline with the intent of educating them.
  3. Buying strategic ad space on video platforms with audience specific message targeting.

For more information on these methods, please read about how we shifted online behaviors among 13–19 year old boys in the Indian state of Rajasthan on patriarchal beliefs on women and girls.


An action-oriented communications and advocacy strategy is not complete without the right partnerships. For organizations actively working at the community level in Delhi, please see this recent Quilt.AI blog post. In Kampala, organizations such as the Uganda Environmental Education Foundation, Greenwatch, Deniva, Environmental Alert and Ugandan Youth Fighting Climate Change may be a good place to start for insights and on-ground impact measurement.

In Tel Aviv, there is a plethora of organizations working on a cross-section of climate change issues, including but not limited to the Heschel Center for Sustainability, the Porter School for the Environment and Earth Sciences, Green Course, Life and Environment, EcoCinema, the Association for Sustainable Economy, the Green Collar, the Green Network and the Society for the Protection of Nature in Israel.

In Guangzhou, Eco Canton, and the China Youth Climate Action Network (CYCAN) are a good entry point into community level organizing on specific climate change issues.

Using this five step approach will get us one step closer to cleaner air, climate smart cities, and more sustainable collective action to prevent further climate crisis. Remember, there is no Planet B.

Join us for the launch of our Climate Change Analysis Tool on 10th September 2020. Register here.




We are a culturally rooted, AI powered insights firm that converts millions of data signals into human understanding. Visit us: https://quilt.ai