The alternative advantage of data
Making money in the share markets has always been geared around information and who can use it to their advantage. In the early years, well before the advent of computers and phones, there would be those who would purposely spread negative information about certain stocks and position themselves to take advantage of the inevitable fall in prices. In modern times (and more legally), gaining an informational edge has involved hiring the smartest statisticians (quants & data scientists) as well as having faster technology for either algorithms or positioning networks closer to the stock exchanges networks within a data centre (on this latter point, for those interested in this there is a great documentary from VPRO called ‘Flash Crash’ - see below).
The latest trend in the creation of informational edge has seen the rise of what's commonly referred to as alternative data. This has gained in popularity so much so that it now has events and conferences dedicated to it and some of the largest hedge funds in the world devote serious resourcing amounts of resources towards taking advantage of it. The key reason behind this has been akin to finding a needle in a haystack, but this time, it's finding what insights the human eye cannot see. It's made me wonder, and the reason for this note, whether or not non-investment companies can also take similar advantage of this type of information?
So what is alternative data? Well firstly, it’s not the “alternative” that’s been making headlines in the political news media for notorious reasons (fake news anyone?). Rather, it refers to information that institutional investors use to gain insights about a company that are not readily available in traditional datasets (like market, financial and economic data) . Traditonal data comes from sources like the financial services terminals like Bloomberg, Thomson Reuters Eikon (now Refinitiv), Morningstar, Factset and S&P Capital IQ (just to name a few). These can be quite powerful in the investing world, especially if they are not common and if you are the first to exploit the advanced insights that can give you an edge over your competitors. Examples of this have included satellite imagery (although that’s been around for years), credit card transactions, geolocation information (foot traffic in stores) and others. Every year there seems to be more information created like this and packaged up for institutional investors to take advantage (and it's mostly the large asset managers who can do this since access to this data can be quite costly). Anyway, one of the more popular ones I noticed last year was vessel tracking and shipping information data. Using information like this can give investors an edge over those who stick to using traditional sources. So if this works in the investing space, how can we apply that line of thinking for other industries?
Firstly, those within the investments game who are able to take advantage of alternative data, tend to be those with dedicated resources (hardware/software/people) analysing the large amounts of data that are typically made available. They are likely to have hired data scientists and run large-scale statistical tests of their data to find the best ways of exploiting it to increase their profits. This often includes seeing how well it performed in back-tests (i.e. seeing how well an alternative datasets signal was at predicting future price movements).
Non-finance companies might think that this is a game that they can’t play in but the truth is far from it. In fact, they’re likely closer to taking advantage of alternative data because, in most cases, the type of data that gets classified as alternative for investors is typically the same type that is used within the confines of normal business activity for non-investment companies.Here is an example of what institutional investors consider as alternative datasets:
Geolocation (foot traffic)
Credit card transactions
Web site usage
Obscure city hall records
Social media posts
Online browsing activity
Shipping container receipts
Flight and shipping trackers
Companies with this data are likely to be those generating it and using it to measure their own performance but many cannot take advantage of this because they don’t have the information systems in place that would enable them to do so. Unfortunately, many are still stuck in the dark ages of data where IT controls access to the data that the business needs to do their jobs better. This isn’t IT’s fault but rather it’s likely to be in their control because there are hardly enough skills within the business to manage/massage/wrangle or munge the data into the ways needed to best gain insights from it. Additionally, even if the skills are there to understand what to do with data, many organisations are stuck with software that doesn’t enable them to take full advantage of their own information. There can be a myth that doing the type of analytics that many information/data managers read about is expensive but this does not have to be the case. There are ways to fit a data analytics ecosystem to an organisations profile and all companies with all sorts of budgets can take advantage of this. Taking those first steps is also easier than you think because great insights have often been gained just by doing simple tasks such as joining different databases of information together and visualising them. I’ve seen first-hand the story of data savings being discovered thanks to simple efforts like this so it’s surprising that it’s not done more often.
Getting back to the alternative data spiel, I think companies in many industries can take advantage of this if they get their own data and insights in order first. They can do this by talking to data experts who are dedicated to the cause of solving data problems with software and relevant industry skillsets like those seen at ABM Systems.
I hope that companies start to realise how valuable their own data is and how they owe it to themselves to take advantage of this. Then, when its time to level up, they can look at ways in which information from outside sources can be brought in to augment their analysis of their customers, competitors and themselves.
For those interested in alternative data in finance there are some great conferences put on around the globe (especially if you're in New York or London). Check out Battle of the Quants run by Bartt Kellerman and team: https://battleofthequants.com/
For those interested in how to improve data organisational literacy then you should check out the great work of Jane Crofts and team at Data to the People (http://datatothepeople.org). Follow her on Twitter and start talking about data at your organisation today.
There is a growing need to have people become more data literate but they can't do it without tools and some of the best are provided by the software companies like Qlik. I've had a bias towards using their software for many years but it's been with good reason since it has strong capabilities at its price (https://www.qlik.com/us/). Additionally, they've grabbed the data literacy flag with both hands and have been leading the charge on this, even creating a free classroom where you can learn about data. Big thanks to Jordan Morrow for the initiative on this and you can check out the free courses his team created here (https://qcc.qlik.com/course/view.php?id=811&_ga=2.122054539.1020865877.1538919118-1418984300.1534401081).
Other people might be using some of the other very capable data analysis tools out there such as Tableau and Power BI. These are great and a perfect fit depending on the organisation. If you're not already using these then it's your chance to get involved (you owe it to yourself if you've read this far). So go ahead, and put your hand up and ask the question. If you aren't getting the help inside your organisation then ask me as I've used most of these (and many other tools) and can likely help you to get started.