This location intelligence platform helps businesses make pinpointed decisions using bespoke AI

GeoIQ’s algorithms layer government data, satellite imagery, data from public sources, and many other such sources to augment business datasets for better decisioning. The company’s ML engine unlocks patterns in the data to predict user behaviour and business potential down to an address level.

This location intelligence platform helps businesses make pinpointed decisions using bespoke AI

Monday August 15, 2022,

6 min Read

If a burger seller does not know that students from a nearby coaching centre are regular visitors to the joint, the seller may not be able to cash in on the opportunity. This is not an isolated issue but is a problem that many businesses, across industries, face today—lack of access to quality real-world data that is external to them. 

Businesses seek reliable data models that enable efficient decision-making across geographies and customer demographics. But good models need quality data to give the desired results. While a lot of data lies outside company databases, it is scattered, unorganised, and incomplete. 

Data scientists Tusheet Shrivastava, Ankita Thakur and Devashish Fuloria recognised this gap, in 2018, and set up GeoIQ, a location intelligence platform that helps businesses make pinpointed decisions using bespoke location AI. 

Rich data topped with ML and AI

According to the Bengaluru-based startup, the backbone of its platform is data—over 3,000 attributes covering demographics, income, infrastructure, commercial activity, rentals etc., available at street-level granularity, with 100% coverage across the country. 

GeoIQ utilises proprietary algorithms to layer data from over 600 trusted government and public data sources, satellite imagery, and more to create highly granular datasets that differentiate users from one street to an adjacent one.

GeoIQ enables access to thousands of attributes providing a 360-degree view of any location.

The company’s machine learning (ML) engine identifies the ‘why’ behind the ‘where’, thus unlocking patterns in business data. For example, the ML engine can point out that the number of guests in Hotel XYZ increased on September 12 because there was a music festival 100 metres away. Or it can show that the sales of XYZ burger joint doubled in the last month because a new coaching centre opened next to it. 

Not just that, the ML models also help businesses predict user behaviour, affluence, fraud, and business potential at an address level. All this ultimately helps businesses make smart decisions on which customers to target, when, and how. 


Tusheet and Devashish are both alumni of IIT Kanpur. In 2011, their paths crossed on many occasions, leading them to connect with each other professionally. Meanwhile, Ankita and Tusheet, classmates from their school days, were identifying real-world problems that were challenging the Indian market. Soon Devashish, Ankita and Tusheet came together to work on the areas of data accessibility and hyperlocal intelligence across markets and industries. 

In their cumulative work experience of over 30 years in deep tech, the three of them had seen the negative impact of the absence of data on businesses. They realised that most Indian firms rely on internal data to develop solutions for real-world problems. Interestingly, when public/third-party data sets were utilised to make critical business decisions, businesses saw a 25% uptick in critical performance metrics. 

This revelation inspired Devashish, Ankita and Tusheet to develop an easy-to-access tool that delivers hyperlocal intelligence.
GeoIQ Co-founders L:R - Devashish Fuloria (CEO),  Ankita Thakur (CDO) and Tusheet Shrivastava (CTO)

GeoIQ Co-founders L:R - Devashish Fuloria (CEO), Ankita Thakur (CDO) and Tusheet Shrivastava (CTO)

External data and ML attributes

Devashish says, “There is a huge requirement for reliable external data that sits beyond the company’s database. There are multiple problems that would benefit immensely from this information.” 

For instance, external data can help in demand prediction. For example, cough drop sales could increase as the weather becomes colder. So, a company selling cough drop could benefit from external data related to weather changes. 

Apart from access to reliable external data, GeoIQ also provides over 3,000 ML-ready attributes through a single API. Businesses can experiment with these attributes and figure out the most impactful indicators for their use case. 

Businesses can experiment with these attributes and figure out the most impactful indicators for their use case.

Businesses can experiment with these attributes and figure out the most impactful indicators for their use case.

How it works

GeoIQ collects data from over 600 sources—government-released data, public listings, open data, satellite imagery, surveys, and data partnerships. All this data goes through several layers of processing, hygiene, and validation to ensure accuracy and recency. The data is converted into geo-centric information, which then goes through GeoIQ’s ML engines where the data is treated for biases, anomalies, and missing information. 

The ML layer on top of the proprietary algorithms converts this data into intelligent location attributes. This data is available through real-time APIs at the address level (plot no, building name, street name, pincode, city, state). 

“Businesses can do exploratory analysis, shortlist attributes, build ML models with different attributes, and deploy them as real-time APIs in one click. The model that fits their bill perfectly could be used to make decisions,” explains Devashish.

“Businesses can directly use our data APIs or use our no-code ML platform that helps them identify which attribute adds the most value for their use case. They can come in with just an address and a behavioural aspect they want to predict and the no-code ML platform creates a model.”

In the no-code platform, users need to just input data, select the variables (location attributes such as socio-economic, demographic, infrastructure, etc.) they want to test, and the model will start mapping the relationship between the data and the variables and how they are impacting it.

For example, the revenue of a burger joint could be impacted by aspects such as presence of coaching centres and brands and the average meal cost in the vicinity. 

The team

Co-founder and CEO of GeoIQ, Devashish Fuloria holds a bachelor’s degree from IIT-Kanpur and a PhD from Imperial College, London. He was earlier a co-founder at ZeLadder Sports, a consultant at TWI (UK), and a policy researcher at the Royal Academy of Engineering (UK). 

Co-Founder & CDO at GeoIQ, Ankita holds a bachelor of engineering degree from Pune University. She is a data science professional with over 11 years of experience in solving data problems in retail, BFSI, and hospitality. 

Co-founder and CTO at GeoIQ, Tusheet holds a bachelor’s degree from IIT-Kanpur. He has more than 11 years of experience building machine learning capabilities and end-to-end data product development. At GeoIQ, he heads technology, innovation, and product development. 

GeoIQ Team

GeoIQ Team

Growth and revenue

The startup offers users an annual subscription to access data APIs. 

The company says it achieved 10X growth in annual recurring revenue (ARR) over the last four quarters and is set to register an additional 5X growth in ARR in the next two quarters.

Currently, GeoIQ has 30 clients, primarily in fintech, insurance and retail. The company has collaborated with some of the fastest-growing brands in India, including Lenskart, Zepto, DMI finance, Paytm, and Big Basket. 

Funding and way ahead

In May this year, the startup raised $2.25 million from Lenskart. Existing investors, including 9Unicorns and Ecosystem Ventures, participated in the round. In November 2020, the location intelligence startup raised Rs 2.5 crore led by 9Unicorns.

The global location intelligence market is said to reach $51.25 billion by 2030, growing at a CAGR of 15.6%, according to a study by Grand View Research Inc.

GeoIQ has set its eyes on global expansion, starting with the US in FY23. 

Speaking of competition, Devashish says, “This is a niche and nascent market in India. There are only a few players with the ability to provide sophisticated capabilities on a single platform.”


Devashish considers New York’s Carto as a competitor. “Their platform allows organisations to store, enrich, analyse and visualise data to make spatially-aware decisions. Our approach is more data-science-driven, where we let our AI engines predict answers which are delivered via APIs.”

Edited by Swetha Kannan