Post-Covid has proven to be a golden age for retail investors to get into stocks. Thanks to the proliferation of fintech applications and stricter disclosure rules imposed by the Securities and Exchange Commission of India (Sebi) on businesses, most of the data and related information is now just a click or a swipe.
According to market watchers, the “democratization of data,” as they say, now allows retail investors to compete on a level playing field with high net worth individuals and even institutional investors.
But is he really?
Retail investors are pleased that they can now easily view key financial metrics and ratios such as price-to-earnings, price-to-book, cash flow, return on equity, and profitability. on the equity of the companies they hold shares or that they want to invest in and even compare these numbers with their competitors’ respective figures, all on one website. But unbeknownst to them, the game went up a few notches.
With so much data and information available for free, the race is now on at what isn’t available that wealthy individual investors are willing to pay for. Alternative datasets, as they are called in market parlance, are gaining popularity with this group of investors, whose much larger sums of money are at stake and desire an edge. can help them improve their bottom line or else, at least protect their bottom line. portfolio against any unpleasant surprises. Moneycontrol learned this while talking to a number of wealthy investors using these datasets. Fees for these various services range from Rs 1.5 lakh to Rs 2 lakh per year and in some cases Rs 1,500,000 per month. What exactly are these alternative tuples?
Thanks to advances in data analytics, it is now possible to obtain a wealth of data and information that is different from data and information from traditional sources such as financial statements and official reports. They are gathered from a variety of sources and provide insight into different aspects of companies, industries and economic trends. At a basic level, web scraping and social media data provide analysis of sentiment, brand mentions, customer reviews, and social media trends.
At a higher level, alternative economic indicators are generated from non-traditional sources such as shipping (exports) and shipping data. Further down the chain, you have satellite imagery and geospatial data delivery services that provide information on economic activity, traffic patterns, agricultural productivity, and infrastructure changes.
Companies with such capabilities then add a layer of machine learning to this layer, which helps generate insights into how investors using it can make decisions.
How does it all work?
For example, an automobile manufacturer has high monthly sales. But these are the cars that are delivered to the dealer. The question is how well that translates into actual retail sales. The Vahan portal provides numbers based on data compiled by Regional Transportation Offices (RTOs) across the country. But not all RTOs upload data to Vahan at the same time. So there may be a delay of a few days. Another data set provider might compile data when individual RTOs upload it to their own websites first. And if there is a lag there, the algorithms will be trained to extrapolate trends from past data and come up with estimates.
One quick-service restaurant says its new store is in good working order. Satellite imagery can give an idea of how many people frequent this point of sale. And if you have data on each store, you can get a rough idea of the sales.
Similarly, the number of cars entering the parking lot of a shopping mall can give an idea of the number of visitors to this mall.
Construction companies often store building materials outdoors. Satellite imagery can provide material volume information that can indicate the level of construction activity on a particular project. In the case of a mining company, a larger external inventory may reflect a higher level of production, while a smaller inventory may indicate a slowdown in mining activity. Similarly, the movement of vehicles in the mine will also provide clues about activity levels.
It is possible to monitor the profitability of the exchanges almost daily using software that will quickly analyze the trading volume and the number of trades in each segment, as these figures are obtained by the exchanges. published daily.
Specialized software can crawl e-commerce sites’ websites, track shopping carts, and see which brands people buy the most.
big picture
Investors using these data sets agree that analyzing a company’s fundamentals is at the heart of any good investment. In other words, based on the number of additional data points an investor may have, fundamental research and understanding the company’s business model are most important.
And yet, alternative tuples can help add an extra layer of profitability in certain cases.
“It can work on two levels,” one wealthy investor using several services told Moneycontrol. “First, it can help you gauge a company’s financial performance to a certain extent. Second, it can help cross-check what management is saying, as there is often a gap between management’s assertion and reality,” the investor said. “If you have a portfolio worth Rs 50 crore or more, then you should spend Rs 10-20 lakh a year if it gives you a good picture of the business situation at all times.”
And yet, the effectiveness of surrogate tuples can be limited.
“Quant funds regularly use such datasets all the time because their very model is based on analysing every possible data that is available. In that sense, HNIs have only got into this game recently,” the investor said.
This investor requested that one of the datasets he uses not be named in the story because not many were aware of it right now, even though the service provider has a website with details of its offerings spanning multiple sectors.
“The upshot of all this is that stocks are reflecting all the information at a much faster pace than before,” the investor said.
Piecing it together
The founder of one of the alternate dataset service providers has said in a presentation to students from his alma mater that the best way to get the most of out of such data was not to use more of the same kind of data, but to use data points from related sectors and then connect the dots.
He cited an example where he was able to capitalise on an opportunity in the tractor segment some years back though the rest of the auto sector was faring poorly. He was able to do that by studying the activity in the agriculture sector, looking at fertiliser demand and the water level in reservoirs.