Table of Contents

    COVID-19 has hit almost every industry hard, and we’ve been left grasping at straws trying to make sense of our new reality. One of the biggest blows we’ve suffered is that our old data has become, to put it nicely, mostly useless. So what can we do, in this new landscape? The first step is to understand our new reality with fresh eyes and data. Finding the right ways to frame the situation involves obtaining the right data and the correct signals to interpret economic movements. 

    One of the biggest challenges our customers face today is that many of their old models are no longer functional in today’s landscape. At the heart of the matter is that they can’t properly factor in the current pandemic into their forecasts and strategy. Our customers have repeatedly come to us with broken models, looking for ways to reset them and offer new insights to keep them running.  

    To help our customers better navigate these tough times, we knew we had to build a new trove of data that could add context to their existing models, and provide much-needed clarity. Our solution? To build an entirely new set of COVID-19 signals in our platform that let organizations understand their risk derived from the current pandemic. Even so, the proof is in the pudding, as they say, so we needed to see how they work in the real world. The results? See for yourself. 

    COVID-19 signals in action

    Some of our customers in the B2B landscape are flying blind as most of the instruments they use to measure risk previously were seriously hampered by the new economic realities. They needed to understand how COVID-19 impacts their risk models but had no real barometer or tools to quantify the “Coronavirus effect”.

    Once they connected to Explorium’s COVID-19 signals, they were able to not only understand their new pool of viable loan candidates but to effectively score their potential borrowers based on a brand new risk metric — one that can effectively account for the pandemic’s impact on industries. We build the risk score by combining a variety of signals that include internal company data, policy factors that can impact their operability, and even geographic factors that can affect a company’s repayment risk. 

    Risk Score

    A look at what a company’s results and overall risk score look like.

    Let’s dig a little deeper to break down our new risk score model for small businesses and lenders. 

    Under the hood – a look at Explorium’s COVID-19 signals

    The big question is, what goes into Explorium’s COVID-19 risk score? To understand, we can look under the hood at Explorium’s new COVID-19 signals, which blend our existing datasets with proprietary signals developed exclusively to help organizations understand their businesses in our current landscape.

    Sample Risk Signals

    Once we have your dataset, we can quickly populate it with our unique COVID-19 signals.

    First, we take a company’s data — which, for example, consists of company names and the state they’re in — and identify each potential customer’s industry to start a deeper risk assessment. Once we know where they are and what industry they work in, we can build a new risk score that our customers can quickly reference in their own models. 

    The industry and policy signals

    It’s crucial to understand how COVID-19 has affected different industries, and what that means for companies in each industry. The first thing we did was to focus on adding an industry to each company in our customers’ datasets. This lets us quickly extract other vital bits of data that can give us much greater insight. 

    Perhaps the most important signal is our “market effect” measure, which tells you the impact COVID-19 has had on the industry each company is in. This lets you better understand how an individual company may be affected by a quick glance.

    COVID industry effect

    A visual look at the average market effect on each industry.

    This lets our customers adjust their risk models to account for their potential borrowers’ industry, and better define what their new risk thresholds should be based on external factors. 

    Another thing we understood pretty quickly was that simply knowing if a company’s industry was affected wasn’t enough, as some companies may have other things going for it that help it stay afloat. We added another layer, one that examined whether each company had an online presence, if it had a business rating, whether it was still open, and even the number of social networks on which it was active.

    market effect vs. online presence

     A comparison of market effect and a company’s online presence indicators.

    With internal factors in hand, we also knew that we had to account for the external — after all, this crisis has placed burdens on every single industry. We built a signal that let us track whether a company was considered essential (and therefore remained open), and another one that checked if they were eligible for federal funding (which would impact its survivability and ability to pay back a loan). So, for instance, a healthcare company that is considered essential and receives federal aid may have a significantly lower risk than a hotel which is closed, and not considered critical. 

    The geographic factors

    Policies are a great indicator, but they’re also shaped by location. Our next layer is geographic, adding signals about a company’s location that could impact its risk level. For example, even if a company scores low on other risk factors, it may find itself impacted if it’s in a region that is currently being strained or at higher risk of incidence of COVID-19 cases. Making smarter decisions means being able to predict where COVID-19 may strike next and correcting course based on the data. 

    Our external data includes several features focused specifically on the spread of the pandemic. This includes the number of COVID-19 cases near a data point (or a company), the growth factor of COVID-19 cases by county, and the predicted growth rate. Using these data points, we can also create a virtual heatmap that shows users the most high-risk areas on average. 

    COVID-19 heatmap

    A heatmap showing the COVID-19 risk by US state.

    We also incorporated footfall data to map customers’ movement patterns and identify areas and regions that may have a higher business activity rate than others. Another important consideration, especially for retailers and offline companies, is whether other businesses near them remain open or closed, which is an indicator about a region’s overall activity and therefore risk. Finally, we included data about users’ online and offline shopping habits, which help us paint a better picture of companies’ profitability, especially those that rely on offline sales. 

    Adapting faster with smarter signals

    More than just providing these new signals, however, we believe in helping our customers make sense of them and incorporate them into working, relevant risk models. By combining our new COVID-19 signals with their existing models, our customers have been able to hit the ground running by using risk predictions that can account not just for a company’s pre-existing risk factors, but the economic and societal implications of the pandemic we’re facing. As a whole, our platform can help our customers by injecting their existing models with the factors that will help them determine risk today and tomorrow. By combining our policy, industry, and geographic signals, companies can build a new risk score that is more accurate and relevant. 

    Unfortunately for businesses today, it doesn’t matter how well they did yesterday, or how much data they managed to collect. The reality of the pandemic means that the calculus of risk and understanding business opportunities has radically shifted. Risk has taken on a new meaning, and the factors that define it must now include COVID-19 and its impact. Instead of guessing at how it affects your business, our COVID-19 signals and risk analysis tools let you make smarter decisions based on new data that can help you navigate these turbulent times.