Life in 2030 will be drastically different than what it is today.

Our decisions are going to be defined by a rapidly changing climate - from what we eat, what and where we buy from, to where we decide to invest into (real estate, stock investments).

Some have pushed for "greenwashing" and responding to the unfolding crisis with paper straws and gimmicky sustainable living hacks. But to carry out these decisions, requires us to have access to better data.

How many times have you read the term 'sustainability' while going through a Fortune 500 report that outlines activities that are clearly otherwise.

What we need is to reimagine every sector, every supply chain, and every consumer interaction.

To explain how we at Blue Sky Analytics are working on this transition to climate action, we did an introspective dive to define our objectives with as minimal jargon as possible.

We did this by using the help of DARPA former director George H. Heilmeier's framework: The Heilmeier Catchesim. The list of questions in the catchesim, are what we have answered in the blog below:

  • What are you trying to do? Articulate your objectives using absolutely no jargon.
  • How is it done today, and what are the limits of current practice?
  • What is new in your approach and why do you think it will be successful?
  • Who cares? If you are successful, what difference will it make?
  • What are the risks?
  • How much will it cost?
  • How long will it take?
  • What are the mid-term and final “exams” to check for success?

What are you trying to do? Articulate your objectives using absolutely no jargon.

Blue Sky wants to make environmental data readily available for all.

It wants to do this by building a digital twin of the planet, using remote sensing technology.

To achieve this, we take the help of software using a combination of Artificial Intelligence, Satellite Data and the Cloud. Our setup can be divided into three components:

Geospatial Data Refinery

The Geospatial Data Refinery, similar to an oil refinery, takes in and analyzes massive amounts (terabytes) of raw data. The machine learning algorithms that we use, helps us in generating datasets.

Developer Portal

Our Developer Portal is the gateway through which a user (like a student or insurance firm) accesses our datasets via APIs (basically our store through which we offer our products). If you're still curious about what an API does, then checkout the video below.


SpaceTime, is our version of a platform that will help you visualize the data that we host. We are planning to make it open source (soon), which will help any user visualize data over space and time.

Through these platforms, we aim to make a catalogue of high-resolution datasets on environmental indicators easily accessible and easily pluggable into various use cases.

How is it done today, and what are the limits of current practice?

Most environmental data monitoring happens via ground sensors.

These sensors are very expensive to procure and maintain and are thereby few in number, providing limited spatial coverage. For example, India has 250+ air quality monitors for an area of 3.2 million km². That's almost like the Cyprus government using an air quality monitor installed in Turkey, to measure the quality of air in Cyprus.

An air quality monitoring station in the Indian city of Noida. Given the dearth of good air quality monitors, our AI powered data sets can plug in this gap and provide spatially accurate air quality information. For more check at the end of this blog*. PC: India Education Diary

Further, the data is available via bulky PDFs, complex excel sheets, reports, and poorly designed dashboards. It is difficult to integrate this data and make use of it whether its part of marketing strategies, risk models or insurance underwriting. This is probably why only 0.5% of global data is used for analysis.

In terms of people who use this data, there are two types of players at the moment:

  1. Those that specialize in climate risk modelling
  2. Those that are layering environmental datasets with industry-specific data to draw patterns for asset management, regulation etc.

In both cases, analysis and modelling is laid on top of datasets, which really aren't that great in terms of quality. As accounting for climate-risk gets more critical, high-quality options in this fundamental layer are important to make more accurate decisions.

What is new in your approach and why do you think it will be successful?

We think from the perspective of ease of use - we focus on fundamental issues like will our datasets be reusable, scalable and configurable. This is motivated by the general approach where when a client requires a new dataset, a whole new setup is made to deliver that dataset.

Our approach is to be data and platform agnostic by design, which means that for every new dataset we add to our catalogue, we don’t start from scratch! This allows us to generate and make available new datasets at a quick pace. Further, through our API Gateway, we can cater to diverse, custom requests with negligible back-end involvement.

Second, we are running on the thesis that environmental data will be valuable across sectors this decade. Hence, we are building our expertise in disseminating fundamental level ‘n-1’, ‘n-2’ datasets efficiently to all.  

These datasets can be easily integrated into solutions and moulded according to clients needs.  designed for further analysis, rather than the end dataset that needs to be reverse engineered first before further analysis to be done on top of it

For all other steps, collaboration with upstream data providers, industry-specific risk modelers, researchers, data scientists and the community at large is central to our approach.

Simply put, we will be successful because:

  • we are providing data that is ready-to-use
  • our product development is driven by our community

Who cares? If you are successful, what difference will it make?

  1. ESG Analysts: While thrusted with the responsibility of scoring companies on sustainability metrics, there is little data for them to play around with. A lot of the data is self-disclosed or is sourced while shifting through mounds of reports, excel sheets, articles, and websites.
  2. Amazon, Microsoft, Unilever- anyone going carbon neutral: How does one go carbon neutral? One way is through using our data, which will help in identifying a company's carbon footprint across the value chain and thereby identifying ways to cut down on emissions, waste etc. Indirectly, we hope our data can enable low-carbon businesses, processes to come up that can support the transition to a low-carbon economy.
  3. Risk Modelers: Modelers need good quality data to feed into the models that they are building and perfecting. Often, they have to start with raw data and process it, which is time-consuming and deviates from their value-add. With Blue Sky’s data, that hassle is dealt with.
  4. Insurance Providers, Hedge Funds: To identify low-risk assets and determine how climate-changes can be incorporated in insurance underwriting and determining premiums.

The truth is we have a fairly limited list on our end and that’s why we are building a malleable offering that can cater to an evolving set of use cases. Our goal is straightforward: the data should be simplified and put on a platform that is easily accessible (through our API) so that analysts can actually focus on analyzing data instead of cleaning it up!

What are the risks?

Firstly, the major risk is not identifying the product-market fit in due time. We are focusing on First-Principles rather than anchoring ourselves to a particular-industry and with time, can pivot if the need arises. However, with this approach there is greater risk in not gaining early traction and steady first-customers. It’s a case of moon-shot vs sure-shot, and we are definitely aiming for the moon 🙂

We are focusing on fundamental issues - i.e. as the climate crisis intensifies, businesses and other entities would want to understand and track factors that are putting their assets at risk!

Secondly, risk lies in not being able to attract capital- monetary and human. Both depend on how the ecosystem around Climate-Tech and Geospatial Data shape up. This points the finger back to you!

How much will it cost?

We do not have extensive machine cost and the bulk of our cost lies in sourcing the best talent in Data Science, Climate Science, GIS Analysts, Developers and Community Engagement. Further, some capital is allocated for purchasing privatised satellite-data and executing bespoke client requirements.

How long will it take?

Ah, the million dollar question!

At Blue Sky, we say a tech product is never complete! So while our data refinery is perfecting in the background, we are happy to share that our first two datasets- BreeZo (Air Quality) and Zuri (tracking farm & forest fires) are ready for you to explore on our Developer Portal. BreeZo and Zuri are also a testament to the progress made by the Refinery, where BreeZo took 8 months to wrap up and Zuri 2.

In the immediate pipeline, we have datasets on Surface Water Levels, Soil Health and Industrial Emissions coming up.

You can reach out to Tej or subscribe to our newsletter to stay posted. Your engagement is going to define the length of this journey, in a big way.

What are the mid-term and final “exams” to check for success?

The mid-term exam would entail identifying the product-market fit and have a growing community. An indicator would be the number of recurring users we have on our Developer Portal, SpaceTime and Discussion Forums.

The final exam would be if Blue Sky becomes a household name and is synonymous with environmental indicators, such as BreeZo for air quality. By this stage, we should have bottoms-up brand loyalty, where members on our forums are rooting for us and buying from us.