Search Results

IIT Kanpur’s National Geodesy Centre has verified a benchmarking study, which finds average accuracy of Latlong’s Geo-coding API to be 2x better than Google Maps API. A few months ago, our family was driving in interior Karnataka visiting several temples. We had to get directions from a small village to visit the temple in Goruru, on the banks of Hemavathi in Hassan District. My son opened Google Maps and started giving directions. After driving a couple of kilometres, I sensed that we were heading in the wrong direction. So, I stopped the car and checked the map. I found that we were headed towards another village called Goruru near Bengaluru, in the opposite direction. Before I could open my mouth, my son calmly said that he was blemish-less and Google Maps was to blame. Those maps had Goruru (near Bengaluru) and Gorur (near Hassan) as 2 different villages with different English spelling, when the reality is that both villages have the exact same name in Kannada – ಗೊರೂರು (गोरूरु). If only, the search had shown both as identical possibilities, he argued, he would have stopped to think, rather than choose the one where the English spelling was a better match to the Kannada name. This triggered a conversation with our CTO, Rahul Sindaghatta on how do we use intuitive knowledge about place names. During this conversation, he pointed out that people view addresses telescopically downwards from country to house number, while addresses are written microscopically upwards from house number to country. The way the address is written, impacts converting address strings to location (a latitude-longitude co-ordinate pair or lat-long) in a very big way. So, we decided to translate our ‘tribal’ knowledge (a term coined by a friend of ours) to APIs in our platform. The result was to set an objective for our engineering team, led by Rahul, to better the average accuracy provided by the Application Programming Interfaces (APIs) of the world’s leading maps platforms like Google, Microsoft Bing and TomTom. In a ground-breaking piece of work in deep technology, I am delighted to announce that our engineering team at Latlong, has achieved superior accuracy levels as compared to the world’s leading maps platforms. Specifically, our Geocoding API (Application Programming Interface) has average accuracy levels 2x better than a similar Google Maps API. A benchmarking exercise was carried out in this regard, using several thousand addresses. These addresses represent different possibilities across the country – residential and businesses; hospitals and schools; cities and villages; Tier 1 and Tier 4 cities; apartments and individual houses. I am grateful to Prof Onkar Dikshit, who heads the National Centre for Geodesy at IIT Kanpur. He and his team reviewed the process followed by Latlong and verified the results of the benchmarking effort. We are also grateful to Prof Abhay Karandikar for the opportunity to work with his team at IIT Kanpur. What is geo-coding and why is it important? Process of converting an address into a latitude-longitude co-ordinate pair. This is one of the more complex processes within geospatial work and is a foundational element for all geospatial use cases. It is well-known that Indian addresses are some of the hardest to geo-code. Enterprises use Geocoding API to understand where customers are. This forms the basis for improving their demand generation and fulfilment processes. So, getting the customer location wrong can be disastrous. Note that results from front-end search boxes could be quite different from API results, due to autosuggest and location-preference; while focus for enterprises is API Result of benchmarking exercise In his review report, Prof Onkar Dikshit of National Centre of Geodesy at IIT Kanpur said – “average position error pattern remains: Latlong < Google < Bing < TomTom” Latlong’s Geocoding API has 2x better average accuracy than that of Google Maps API. Latlong’s API accuracy is 35x better than that of Microsoft Bing API. Latlong’s API accuracy is almost 200x better than that of TomTom API. Note: The Table provides average error from actuals, and the lower the error the better the accuracy. What are we doing differently – “Depth of Area” At the heart of the API is the intuitive understanding that humans think of addresses telescopically downwards from the Milky Way to the house number! One can think of each geographical area in an hierarchical fashion: Country (India) -> State (Karnataka) -> City (Bengaluru) -> Locality (Jayanagara) -> Neighbourhood (Yediyuru). The challenge is in implementing the insight due to a host of reasons like overlapping boundaries across these geographical areas: for example, Yediyur belongs to Banashankari and Jayanagara in Bengaluru. The most important reason this has been difficult to implement is that such ‘geospatial shapes’ become compute-expensive to implement. This is where a concept of “Depth of Area” was created by our engineers to be able to quickly select the possibility with the best match. This delivers superior accuracy and performance. Latlong’s strength is having curated area data (called polygons in maps), across India. This helps eliminate all the non-fitting locations very quickly, as they don’t fit into the ‘area’. For example, an address for a village needs the area of the village mapped out, which allows Latlong to quickly zero in on locations with greater degree of accuracy. The ‘depth of area’ concept is made possible by curated area data. It becomes difficult to implement when the ‘area data’ is non-curated and dynamically created through concave / convex hulls. Multiple possibilities of English spelling for Indian names of areas, is a challenge which has been beautifully handled. Latlong uses Indian language scripts appropriately for string matches. For example: Jainagar / Jayanagara / Jayanagar all map to जयनगर (ಜಯನಗರ). Here we acknowledge the APIs of AI4Bharat for this purpose. Such ‘tribal knowledge’, is at the heart of how we have been able create a suite of world-class maps APIs. Try out the Geocoding API and all other APIs at https://apihub.latlong.ai/ A few examples of geocoding API output are presented on Latlong Maps A Tier-1 city address A Tier-2 city address A village address It gives us an immense satisfaction that in a relatively complex and deep technology domain like APIs for maps, we compete with the best and deliver great accuracy. We are excited by the possibilities of what we can deliver going forward.

I started the last essay, writing about the trigger for the essay being the political jousting on which cities are driving economic growth in India. State-level data was examined to understand which states have been driving economic growth, since the pandemic began. The conclusion was that Uttar Pradesh and Karnataka were the largest contributors to economic growth in the last 2 years. Karnataka has also become one of the few large states with a per capita Gross Domestic Product (GDP) in excess of ₹3 lakh. We also examined a couple of hypotheses as to what may be behind the contribution of these states: reverse migration towards states like Uttar Pradesh and the Work-From-Home (WFH) effect of the pandemic. One of the questions, we posed was whether these factors amplified, trends already at play. We delve a level deeper and examine data at district and city level, to see what the data reveals. Indian states have been reporting district-level GDP for several years now. However, the data is reported typically with a year or two’s lag. In order to use comparable data, where the latest year’s data is unavailable, we have used that particular state’s growth for estimating current year’s data for a given district. All data we will refer to will be at current prices. To recap: India’s GDP is estimated at ₹230 trillion for 2021-22 Per capita GDP is ₹1.68 lakhs India’s GDP compounded at 7.2% over the last 3 financial years Growth engine of large cities Urbanization has been an important force behind economic growth of most countries and India has been no exception. Here we look at a few of the large cities of the country and the size of their economy: 6 cities make up 15.3% of India’s economy. Bengaluru is growing at a significantly faster pace than the other cities over the last 3 years at double digits National Capital Territory (NCT) of Delhi is 4% of India’s economy and has grown at the same rate as the country The 3 large cities of Maharashtra have been growing slower than the country While Hyderabad is growing faster than the country, it still lags the growth rate of Bengaluru by a distance The primarily services driven economy of Bengaluru (83%), has been able to grow rapidly in the “Work-From-Home” pandemic world and grow significantly faster than other cities. Notes on city level data in this essay: Mumbai comprises both City and Suburban districts Thane comprises Thane and Palghar districts Hyderabad comprises both Hyderabad and Ranga Reddy districts Data for a few large cities in India, viz. Kolkata and Chennai have not been available for the last several years and hence, these cities are not considered here. While absolute growth is critical, one has to take population into account while comparing economies of different geographies. We will look at the Gross District Income per capita (GDI), for this purpose. In the previous essay, we had pointed out that Karnataka was the only large state with a Gross State Income per capita in excess of ₹3 lakhs. Bengaluru – leader in GDI Bengaluru leads the cities with a GDI of ₹6.8 lakhs. This is more than 4 times the national average per capita GDP. This one number explains the reason behind the large-scale migration to Bengaluru, from both within and outside the state. Bengaluru’s population has grown by 70% in the last two decades, while the country’s overall population has grown about 28%. The other 5 cities in our analysis have between 2-2.7 times the country’s GDP per capita. Faster economic growth coupled with more disposable income (as seen in GDI), clearly differentiates Bengaluru amongst Indian cities. One question which often crops up is, whether disposable income is concentrated in these cities / districts only. Distribution of disposable income Districts of 3 states with shared borders are considered – Maharashtra, Karnataka and Telangana. None of the districts in these three states has GDI less than ₹1 lakh. In Maharashtra, the Division boundaries almost mirror the GDI pattern. Most of the districts in the Amaravati, Aurangabad & Nagpur divisions having GDI less than ₹1.7 lakhs. Kolhapur and Nagpur outside the Mumbai-Pune belt have GDI more than ₹2.5 lakh. Outside of Hyderabad, the districts of Jayashankar Bhupalpally, Jagtial, Hanumakonda, Medak, Sangareddy (has a little overlap with Hyderabad Municipal Corporation), Yadadri Bhuvanagiri, and Wanaparthy have GDI greater than ₹2.5 lakh. Dakshina Kannada district in Karnataka, with Mangaluru as its headquarters, has GDI in excess of ₹5.1 lakhs, more than 3 times the national number. Udupi, Chikkamagaluru and Shivamogga also have GDI more than ₹2.5 lakh. Analysis summary · Pandemic has resulted in re-location of growth to certain cities, like Bengaluru, while the older engines of growth like Mumbai and Delhi are growing around the national trend. This is clearly seen in the GDDP and GDI data. · It appears to be an underlying decadal trend which was accelerated by the pandemic, considering the population in cities like Bengaluru and the 4 times national average per capita economic size. A Practitioner’s view Vivek Sunder is the CEO of Cuemath, an edtech company focused on teaching mathematics. He has deep understanding of consumers and marketing in his career, across multiple countries. Key insights from him are: · Location is a great proxy for affluence and income levels · Affordability in a given area and amenities go hand in hand with affluence · Data in hyperlocal areas like postal codes and school districts is used in the western world to determine customer ability and willingness to pay He put a unique market penetration model in place for planning and tracking brand growth in hyperlocal areas, using Aaloka. In summary, the emergence of Bengaluru’s economy with its size, pace of growth and disposable income is borne out by the numbers. Wonder if it is a coincidence that the biggest box office hits of 2022 are movies from Bengaluru and Hyderabad?! In the first part of this blog, we looked at data at state level, to understand growth dynamics in the country. The analysis for this blog has been done using Aaloka, a unique location-analytics platform, with external data benchmarks. Users can try out the platform for free at beta.aaloka.in.

As the financial year, 2021-22 came to an end, one saw a lot of jousting between ministers of different states, about where the economic growth of India is coming from. Triggers for such one-upmanship may be political; however, it is extremely critical for businesses to understand the geographical distribution of income in the country to as granular a level as possible. Towards this end, we decided to examine the budgets and economic surveys of various state governments to understand which areas are growing rapidly. India’s Gross Domestic Product (GDP) is estimated at ₹230 trillion for 2021-22, at current prices. This is a growth of about 18% over the previous year, which was dramatically affected by the pandemic. It represents a compounded growth of 7% over 2019-20 (2 years). In this essay, state-level data is reviewed to look at the geographical spread of growth. In a subsequent write-up, data from key cities and districts will be dwelt on to understand growth dynamics. Most states have already published GSDP estimates for 2021-22. For a few states, where GSDP data is missing, their last GSDP estimate has been multiplied by the national average growth to arrive at the estimated GSDP. Fortunately, these estimates are very few. Impact of pandemic at state-level As with any country, there is economic concentration in a few states: 5 states comprise about half the economy – Maharashtra, Tamil Nadu, Uttar Pradesh, Karnataka and Gujarat What is remarkable is the incremental growth is not coming in the same order as the size of the economy with UP and Karnataka delivering 2-year Compounded Annual Growth Rate (CAGR) in excess of 12%. These rates are significantly higher than the India growth. Maharashtra, in this 2-year period, lags the India growth rate, slightly. Such sharp re-location of growth in a 2-year period, has huge implications for businesses, whether their order management is driven from physical stores or from virtual ones. One can venture to suggest two possible hypotheses for the re-location of growth, driven primarily by the pandemic: ● Massively services driven economy like Karnataka (66%), has been able to take advantage of the Work-From-Home (WFH) environment to grow at similar pace as past years. ● A state like Uttar Pradesh has been able to utilise workers who migrated back, in productive work, especially in infrastructure build-out. While absolute growth is important, one cannot lose sight of the fact that all these states have very different population.So, one needs to look at GDP Per Capita (Gross National Income or GNI), to get a better sense for spending power.India’s GNI stands about ₹1.68 lakhs (about $2.2k). The famous $4k Experts in consumer behaviour mention that once the GDP per capita (Gross National Income or GNI) crosses $4,000 (about ₹3 lakh) consumption basket changes significantly.One can keep that in mind, while looking at GSDP Per Capita or Gross State Income (GSI). ● 9 states / Union Territories (UTs) have GSI in excess of ₹3 lakhs. ● Of these, the largest economy is Karnataka, followed by Gujarat and then, Delhi. Here is a look at the large 5 states comprising about half of Indian economy: Economic growth coupled with more disposable income (as seen in GSI) is clearly seen in Karnataka, closely followed by Gujarat. Tamil Nadu and Maharashtra have very close GSI, with the slower growth in the last 2 years hampering GSI in Maharashtra. While the absolute incremental amount of growth in Uttar Pradesh has been the highest in the last 2 years, the vast population of the state keeps the GSI quite low. Analysis summary ● Pandemic has caused significant re-location of growth over the last 2 years. This is clearly seen in the GSDP and GSI data for the last 2 years ● Perhaps, it was an underlying trend going back to earlier years, which was exacerbated by the pandemic-related migrations and economic activity. This will be known over the next few years, as more economic activity confirms and rejects this hypothesis. ● Credit growth data provides some corroboration of the re-location of economic growth. I had commented about this last year. In this light of the GSDP growth and last year’s credit growth, practitioners could start paying serious attention to RBI’s credit and deposit growth data, which is available at good granularity and frequency, for leading indicators of growth in different parts of the country. ● Uttar Pradesh and Karnataka have emerged as drivers of the Indian economy A Practitioner’s view Anand Bhatia, CMO of Fino Payments Bank, has a ring-side view of all the changes happening in the country, with their 8+ lakh merchant network. He has not only seen the changes during his visits to the heartland but has been able to take advantage of a massive build-out in the state of Uttar Pradesh. This was done by mapping every single village (across the country) and its desirability for an immediate merchant location, using Aaloka extensively for this purpose. This kind of structured approach to the distribution network has delivered sustainable profitable growth, to the newly listed entity. In summary, the emergence of Karnataka’s economy with its size and disposable income is quite evident. It is quite poetic, that a Kannada movie, the Yash-starrer KGF2 (with all its dubbed avatars), is setting box-office records in India. In the next part of this blog, we will examine data at a more granular level – district and city level, by considering some of the states discussed in this essay. The entire analysis for this blog has been done using Aaloka, a unique location-analytics platform, with external data benchmarks. Users can try out the platform for free at beta.aaloka.in.

Over the last couple of months, India’s vaccination programme has picked up pace. 3 times in the last 12 days, India administered 10+ million doses to its adult population. India has administered 710+ million doses to its adult population, with about 543+ million people getting at least 1 dose of the vaccine. This compares with about 376+ million doses being administered in the United States, with 207+ million people having received at least 1 dose. A lot of commentators use percentage of population vaccinated to compare different geographies; but this is not a very meaningful way for a couple of reasons: Different countries have different guidelines on which age group is eligible for vaccination. For example, in India only 18+ are eligible, while a few countries in the west, 12+ year olds are eligible. Each country has substantially different age demography. India has a third of its population under 18 years, while the US / UK have less than 20% of their population under 18. Hence, it is best to use percentage coverage of "eligible population", to arrive at meaningful conclusions. In this context, Team Latlong (https://www.latlong.ai/) has been publishing maps every day, on both LinkedIn (https://www.linkedin.com/in/pnbhatta/) and Twitter (https://twitter.com/BhattaPN) showing the eligible population covered by 1st dose, as also total number of doses administered. Latlong has used eligible population estimates combined with data from Cowin (https://dashboard.cowin.gov.in/) site for this purpose. As of date India has administered 1st dose to 58% of eligible population; while, the United States has administered 1st dose to 75% of 18+ population and 73% of 12+ population, per data from CDC (https://covid.cdc.gov/covid-data-tracker/#vaccinations_vacc-total-admin-rate-total). United States has administered 139 doses per 100 adults, while India has administered 77 doses per 100 adults. One of the data points, which has been greatly speculated is ‘when’ will India reach 100% vaccination of adults and the rate of vaccination to get there by December 31st. Typically, maps are visualised based on numbers; Team Latlong used this opportunity to create a map of India, of when different states would reach certain vaccine level. For this map, the following data was : 80% of eligible population coverage with 1st dose was considered, since the US has been around the 70% mark for several weeks now. Vaccination rate for the last 2 weeks (14 days) was considered, as different states are following different strategies to vaccinate: some are doing massive 1-day camps covering 5-6%+ eligible population, while others are doing a steady 0.8% coverage every day. Based on this, it is estimated that India will reach 80% coverage of its eligible population with first dose of vaccine, by 13th October. The most populous state of Uttar Pradesh is likely to reach 80% coverage of eligible population by 26th October, while economically most critical state of Maharashtra is estimated to do so by 3rd November. 15 states / Union Territories already exceed 80% coverage. Telangana and Chhattisgarh, among states with 10+ million eligible population, are likely to see 80% coverage after mid-December. 3 states in the north-east are outliers: Meghalaya, Manipur and Nagaland are projected to reach 80% first dose coverage only in middle of next year. These numbers could change in 1-2 days, as the last 14 days has seen very little vaccination in these states. This visualisation of maps using a date range, is a new feature being implemented in Latlong’s No-Code Analytics platform, Aaloka. There are some more exciting features, which will be released for use, in a few days. Next Team Latlong will produce a map of estimated date by which India and different states are likely to reach 150 doses per 100 adults. This number has also been chosen looking at the US data. For now, we all have our eyes peeled on 13th October!

Financial year 2020-21 (FY21: April 1st 2020 to March 31st 2021) has been a challenge for most individuals and businesses. Headline GDP number shrank by 7.3% in India, due the severe impact of the pandemic. An important indicator of economic health is credit growth. The Reserve Bank of India (RBI) reported a growth of 5.6% in credit provided by scheduled commercial bank in FY21. The stock of credit grew to 110 lakh crores from 104.5 lakh crores, in this period. During this time, deposit growth was healthier at 12.3% and the stock of deposits with scheduled commercial banks grew to 154 lakh crores from 137.5 lakh crores. So, the Credit to Deposit (C / D) ratio worsened from 0.76 to 0.71 in the last financial year. Here, we will examine credit growth. While credit growth was quite low in FY21, the credit growth across the vast nation provides a very fascinating picture. Credit Growth Across States The map of India with annual credit growth across states and Union Territories (UTs), paints a remarkable picture. Three states / UTs (Maharashtra, Delhi & West Bengal) accounting for 42% of the credit, grew by a total of 0.7% in FY21. Credit in Delhi (DL), with a 12% share in credit, shrank by 3%. Credit in the remaining part of the country grew by 9.3%. Among states with more than 4% credit share, 3 states (Andhra Pradesh, Uttar Pradesh and Telangana) stand out with double digit growth. Credit in Andhra Pradesh (AP) grew by 16%. Economically smaller states and UTs reported uniformly high credit growth, as can be seen from the map. Newly formed UTs of J&K and Ladakh, both had credit growth of 18% and 20%. The map clearly points to a significant deviation in growth rates of credit across the country. Contribution to Credit Growth To understand impact of this geographical spread of growth rates on overall credit, it is critical to look at the contribution of the regions to the total growth in credit in this period. Tamil Nadu, with a 9.4% credit share, contributed 12% to the country’s growth, marginally higher than Maharashtra. AP with a 4.3% credit share, contributed 11% to the growth. Maharashtra with a credit share of 26%, contributed to 12% of the credit. If one examines data across a few years, couple of regional shifts can be noticed: Credit in India’s has growth at a Compounded Annual Growth Rate (CAGR) of 8.6% in 4 years. In this period, CAGR of credit growth in Maharashtra (MH) is 5.9% and the states share in credit has reduced from 29% to 26%. West Bengal (WB) has seen a CAGR of credit at 0.9% in the last 4 years, and its share in credit has dropped from 5% to 3.8%. Three states (MH, DL, WB) had a share of credit of 46%, which has dropped to 42% in 4 years, as credit in all 3 states has been below the national average. UP credit has seen a CAGR of 10.6% and its share of credit has inched up from 4.4% to 4.8%. Clearly, growth dynamics across the country are changing across the country. One wonders if the picture is similarly changing within a state. Credit Growth Within State Credit in the state grew by 11.7% in FY21. 4 year credit CAGR was 10.6%, indicating an acceleration in credit growth. Interestingly, while C/D ratio of India dropped from 0.74 to 0.71 in 4 years, UP’s C/D ratio went up from 0.39 to 0.41. So, opportunities for credit absorption have increased in this time. A picture says a thousand words. Map of credit growth in FY21 in UP is a sea of green, with more pronounced growth in the central (Awadh region) and eastern (Poorvanchal) parts of the state. 4 districts closest to Delhi (Gautam Buddha Nagar, Ghaziabad, Agra & Meerut), with a state credit share of 23%, contributed only 15% to the state credit growth. The state capital of Lucknow saw credit growth of 28% in credit, with incremental credit of 13,300 crores. To put this in context, this is the same as the increase across the 2 districts comprising Mumbai and marginally higher than the increase in Bengaluru Urban district. Those districts have orders of magnitude higher amount of credit stock than Lucknow. Credit growth has been consistently high for many districts in Awadh and Poorvanchal regions, for the last 4 years. Jaunpur, Varanasi, Ghazipur, Prayagraj, Chandauli, Kaushambi, Rae Bareli, Ayodhya, Pratapgarh, Ambedkar Nagar – have all seen credit CAGR of 15%+ in this time. Lucknow credit CAGR was 14% in the last 4 years. Districts nearest to Delhi have grown at the pace of the state or lower. Gautam Buddha Nagar, Ghaziabad, Meerut have seen a credit CAGR of 8-11%. Analysis Summary Credit growth is a good indicator of economic activity. Different states have varying degrees of economic activity, reflected in credit data. Two key trends seem to be pushing credit growth in FY21: infrastructure push worker migration Trend of credit growth moving to newer regions and centres is quite clear. It is quite likely that the pandemic driven changes to migration accentuated this trend. As newer centres of growth emerge quite quickly, enterprises need to be alert to these and plan their distribution networks to cater to new customers. View From The Practitioner Sanjeev Moghe is the head of credit cards and payment solutions at one of India’s largest banks, Axis Bank. It has millions of customers and merchants spread across the country. Geographical distribution of network and transactions is extremely crucial for Sanjeev. Understanding regional shifts which help identify new hot spots of growth. Informed decisions on planning distribution network. Growth projects can be planned at local level, keeping in mind local market share. To use a phrase of the great Indian poet, Bhavabhuti (भवभूति), (from a slightly different context!), “विपुला च पृथिवी”(the world is vast). So, financial companies need to start spreading their network across the nation, as growth becomes more distributed. If you want to know more about hyperlocal analytics platform, Latlong Aaloka, write to abhiyach@latlong.in.

India is a demographically ‘young’ country is well-known. The median age of Indian population is about 28 years, while that of China is 37 years. The presence of a young workforce is also referred to as ‘demographic dividend’. Of course, there are concomitant challenges of finding employment opportunities for the crores of people entering the workforce every year, substantial migration to areas with higher opportunities and associated matters. However, any average invariably misses the layered information beneath that one salient number. One specific aspect that curiously receives very little comment is the geographic heterogeneity of data. In a complex and socio-economically diverse country such as ours, the geographic dimension plays a significant role in what is happening to population age cohorts. In this context, it is very heartening to note that the Economic Survey 2019 devotes an entire chapter to discuss age-wise demographics of Indian population. Chapter 7 estimates that Indian Total Fertility Rate will reach 1.8 by 2021, with a range of 1.5 to 2.5 for different. Based on the very different age cohorts in different states, the Survey makes a case for different policy suggestions in these states. One of the examples quoted is in the area of education: elementary schools may have to consolidate to survive in some states, while in other states there is need for more elementary schools. Not only is age cohort data important for policymakers, it is a vital attribute of consumer profile for a company. Products are created with a certain age demography in mind and awareness of where such population is in higher numbers, helps in targeted marketing as well as in planning inventory in different parts of the marketplace. So, knowledge of age demographics is critical for any consumer-focused company. Age demographics vary significantly within a state: Latlong estimates that the proportion of Karnataka population which is 60+ years old is about 11.9%, while that of Uttar Pradesh is about 8.6% and Tamil Nadu’s is about 13.5%. On the other hand, the proportion of population in colleges and fresh into the workforce, i.e. 18–29 cohort in Karnataka, Uttar Pradesh and Tamil Nadu is estimated to be about 19.7%, 21.7% and 19.1%. Economic Survey highlights some of these differences in age cohorts across states. But, what it doesn’t talk about is the fact even within a given state population distribution across age cohorts is quite heterogenous. A map of Karnataka with Latlong estimates of proportion of population in the 18–29 and 60+ age groups is presented below. A typical assumption would be that a large metropolis like Bengaluru would attract a large number of young workers and hence, would have a higher share of 18–29 year olds, while other districts away from such growth pockets would have a higher proportion of 60+ year olds. This turns out to be only partially true. The estimates show that Bengaluru indeed has a lower share of 60+ year olds at about 8.9%, substantially lower than Karnataka average and much closer to Uttar Pradesh levels. However, Bengaluru also has a lower proportion of 18–29 cohort at 16.8%, again significantly lower than Karnataka average. The 60+ cohort has a huge range from Bengaluru’s 8.9% to Udupi’s 16.7%. The 18–29 cohort has a smaller range between Udupi’s 18.4% to Raichur’s 22.5%. Geographically, while the Hyderabad-Karnataka region has a higher proportion of 18–29 year olds, the districts immediately surrounding Bengaluru are estimated with a higher proportion of 60+ year old. This data supports the anecdotal information that many of the youth from districts surrounding Bengaluru tend to move to the city in search of employment opportunities. Also, Hyderabad-Karnataka region being one of the lesser developed parts of Karnataka, has per capita income levels similar to the averages of lesser developed states of India. The age cohorts also indicate the similarity. What is abundantly clear is the fact that districts within a state aren’t homogeneous at all. Now, let us examine an even smaller area, a city. Age cohorts vary even across a city: As pointed out earlier, Latlong estimates the proportion of Bengaluru population in the 60+ cohort at about 8.9%; while that of 18–29 cohort at 16.8%. These are very different from the state averages, though Bengaluru forms a dominant 18% of the state population. We again see the heterogeneity of age cohorts within a city like Bengaluru. However, there is a more marked pattern at play in the case of Bengaluru than was the case at a state level. The map clearly shows that the central part of Bengaluru (old Bengaluru?!) having lesser proportion of 18–29 year olds and a much higher proportion of 60+ year olds, than the outer areas of the city. One of the older parts of Bengaluru like Malleswaram has 10.8% 60+ cohort and about 15.3% 18–29 cohort, while Kengeri, one of the newer parts of the city municipality has 8% and 22% of the same cohorts. Factors like presence of colleges in areas like J P Nagar and Banashankari, also cause non-homogeneous distribution of age cohorts. One of the trends in areas in Bengaluru like Jayanagar and Malleswaram is parents living in the city whose children are typically in another country. Such factors also show up in the age distribution across the city. To understand how geographical diversity in age cohorts impacts branding and marketing, I spoke to an expert in this field. Implications for a marketer: Vivek Sunder, COO at Swiggy and a veteran from the Fast Moving Consumer Goods (FMCG) space, summarises brand implications in two key takeaways: The heterogeneous nature of age cohorts means that consumer profile for a marketer needs to be truly hyperlocal. One cannot operate with a typical profile for Karnataka and one for Tamil Nadu. Consumer profiles need to be at Jayanagar level and Chembur level or even more hyperlocal. This will help brand outreach be very specific to an audience. What product will get consumed in what locality will vary quite sharply, every few hundred meters, with the sharply different age profiles. So, having the right mix of inventory (or restaurants in the case of Swiggy) becomes very critical. ‘Heterogeneity’ of geospatial distribution of population, across any geography is a given, with all its attendant impact on brands. So, brands need to use geospatial analytics to think and execute with ‘hyperlocal’ consumers in mind.