A beginner’s guide to alternative data

What is data?

For a newcomer to our space, the term ‘alternative data’ can sound overly technical, but it can actually be explained very simply. Let’s start by first defining ‘data’ itself.

Broadly, data can be thought of as information that’s collected and then used for analysis. Almost every action that humans take, particularly when we act as consumers in a digital world, leaves behind a trail of data in our wake. Data helps us explain why and how things happen and helps us inform our decision-making processes.

Data is particularly useful in industries such as financial services, where investors use this information to make decisions about the types of investments they should make. These investors may use data to decide whether an individual company, industry or sector is worth investing in, or how they should bet on the health of an individual country or the global economy.

Traditionally, investors have used data derived from the historical prices of stocks, bonds and other financial instruments. This type of data is highly structured and is backward-looking, as it incorporates historical price information.

What makes data alternative?

So, what makes a dataset ‘alternative’? Alternative data typically refers to data used within the financial services industry that is gathered from non-traditional or unofficial sources. Users of alternative data are looking to improve their investment decisions and gain a competitive edge when trading financial products like stocks and bonds in markets, or when investing in private-market deals.

This type of data can be recorded in real- or near-real time and is typically derived from unorthodox sources, such as satellites, sensors connected to the internet and consumers’ smartphones. Alternative data is mostly unstructured and has to be analysed by a data science professional to extract value.

Alternative datasets can also be combined with other data, both traditional and alternative, to form a fuller picture of the world.

While traditional data has only limited sources, alternative data collection is only limited by your imagination. Think about it – in our interconnected world, we can obtain useful data from medical equipment, live traffic updates, online browsing activity and even your daily step-count.

The COVID-19 pandemic also played a significant role in both the supply of and demand for alternative data. As more activity moved online, more data about consumers’ and businesses’ daily activities was generated. At the same time, many new types of users, such as banks, consulting firms and corporations, turned to alternative data to help them understand changing human behaviour. Advances in technology have also made it easier to gather data and monetise it.

When applied specifically to investing, alternative data can provide insight into the future performance of target organisations or industries, which allows investors to identify the most likely avenue for making a profit.

What are data providers?

In recent decades, a huge increase in computing power, combined with the enormous adoption of personal computers and mobile devices around the world, has caused massive growth in data generation. As a direct result, companies have begun to collect and analyse this vast expanse of data and offer it as a product to inform investment decisions. These companies are called data providers or data vendors.

A data provider sources and collects datasets, which it then provides or sells to hedge funds and other data users. An alternative data vendor may simply collect the data, or it may structure, clean and even analyse the data before providing it to the end user.

Alternative data providers use external partners like Neudata to help them form relationships with data buyers. At Neudata, we specialise in helping data vendors by introducing them to relevant data buyers interested in unique sources of data. Because Neudata sits between data buyers and data sellers, we’re in a unique position to help you market your data and provide market intelligence about your product.

Since 2016, Neudata has been the only provider of data scouting services that does not require vendors to pay a commission or enter into a revenue-sharing agreement to list their data.

How is alternative data changing the finance industry?

Alternative data usage has been important in the financial services industry for several decades, but its adoption has expanded rapidly in the past 10 to 15 years.

Quantitative hedge funds were the original innovators, but now practically everyone in the financial services industry utilises alternative data in some form. It’s now being used to provide competitive insights that would be unavailable when only using traditional sources of investment information.  

Leveraging alternative data in this way has allowed investment managers, funds and firms to generate alpha, a term that’s used to describe an investment strategy’s ability to beat the market. Alternative data is enabling investors to uncover new investing opportunities and helping to de-risk their portfolios, truly transforming the way finance professionals conduct business.

What types of datasets are there?

Neudata provides research on over 7,000 datasets across 19 proprietary data categories: crowdsourced, economic, ESG, event, financial products, fund flows, fundamental, internet of things, location, news, primary research, satellite/aerial, search, sentiment, social media, transactional, weather, web-tracking and web-scraping.

Neudata’s proprietary data also shows that companies in North America, Europe and the Asia-Pacific region are the top three geographical users of alternative data globally.

Below, we’ve provided more information on the categories of data that buyers have historically found most beneficial.

Consumer transaction data

As the name suggests, consumer transaction data provides visibility into financial transactions made by consumers. For example, when a user purchases a product/service or makes a deposit into/withdrawal from an account.

This type of information is extremely valuable to many different types of businesses, primarily because it provides insight into the behaviour of consumers and shows the products and industries that are gaining traction or falling out of favour. In other words, it can provide a window into company performance or be used to analyse macroeconomic trends.

ESG data

ESG data tracks how well companies are fulfilling environmental, social and governance goals. Investors typically use ESG data to both assess a company’s performance on ESG metrics and its risk profile. Alternative sources of ESG data tend to be extremely helpful to users who otherwise have to rely on self-reported metrics that can lead to ‘greenwashed’ results.

Environmental data typically provides insights into a company’s sustainability metrics, while social data explores the treatment of people (employees, suppliers, individuals and groups). Governance data allows users to delve deeper into the structure of rules, practices and processes used to direct and manage a company, potentially uncovering things like corruption and how well a company is adhering to regulations.

Web- and app-tracking data

Web- and app-tracking data measures consumer behaviour online, specifically through tracking metrics such as site visits/app sessions, traffic, engagement and user activity.

This type of data is extraordinarily useful across a wide variety of use cases, including the internal or external assessment of a company’s web presence and performance, as well as identifying certain patterns in consumer behaviour.

Geolocation/location data

Geolocation data is any type of data that determines the precise (or as close to precise as possible) location of a person, object or place. This type of data is typically collected through smartphones and other internet-enabled devices.

Often cited as the quintessential investment management use case, location data sources can help users model consumer behaviour at physical locations and forecast future financial performance, assess migration patterns that shed light on macroeconomic trends and more.

Sentiment data

Sentiment data attempts to quantify people’s feelings and emotions towards certain events or products.

Vendors that operate in this space typically use machine learning techniques like natural language processing to analyse news events, reviews and social media posts to understand what topics people find interesting. Then, vendors will assign scores that reveal whether people’s opinions are changing positively or negatively towards a specific security or event.

In addition to being used to understand sentiment on a micro level, this data can be used in a macro context to understand how broader human sentiment is driving world events.

If you’re a data provider or data buyer, Neudata can help you unlock the world of alternative data. To get started, request a trial and see how our industry-leading research can help your company.

‍