Analysts like Forrester, Gartner and IDC are predicting that many industries are being fundamentally changed by “Big Data”. The basic idea of the Big Data trend is to employ better tools that analyze information that you already have (e.g. website stats, ERP data) and combine that with additional data that you can get (either free or paid for) such as social media, market research and other information. But how should Finance teams take advantage of Big Data?
In many respects I agree with Bruno Aziza in his blog “Why I don’t buy the hype about ‘big data’“, that in order for Big Data to truly be justifiable, the costs need to plummet (both for the technology and the external data). But at the same time there is a need to be more realistic about how little data is needed to gain a benefit. At IFS, as the developers of the IFS Applications ERP tool, we can help you streamline the collection of data from your business system to support these initiatives.
For a real world example let’s look at Wal-Mart who are taking social media data about local sports teams to their nearest stores. Wal-Mart judges the level of its store inventory based on the number of mentions sports teams get in social media. When there is a lot of social media activity about a particular team the relevant merchandising is increased in the nearby store. In a similar but more internal way, Amazon make suggestions at the bottom of product pages about items “Frequently Bought Together,” in the hope that you may buy a related item as well. Some other uses can be seen in this article by Gigaom.
So the Big Data concept has the potential to combine multiple data sources (with structured and unstructured data) into a single tool that provides your business better information. Below is a breakdown of example types of data that companies are collecting for Big Data analysis. (Infographic from The Raconteur)
From an operational perspective it seems to make a lot of sense to look at a wider set of information, so that you can make suggestions about extra things customers may want to purchase, thereby maximizing your sales and better satisfying customers’ requirements.
For more administrative functions (like Finance, HR and IT/IS) there are a huge number of opportunities where Big Data could make a difference. Here are a few simple ideas that spring to mind:
- Compare what is the average cash receipt timeline for customers (by entity, geography, customer type etc.) against the wider market so that it is possible to target improvement activities. IFS Applications allows you to use this type of information so that better cash forecasts can be generated.
- Investment decisions based on the social media popularity of a company. An example of financial market sentiment analysis is Derwent Capital Markets.
- Stronger budgetary forecasts by adding additional demand predictors (e.g. social media sentiment, Google search trends, seasonal demand from entry counters on your business doorways and analyzing the relationship between product clicks on your website versus subsequent product demand).
In the examples above the costs vary widely. These costs are made up of two basic components:
- How to capture the data to analyze.
- How much technology and resource is needed to be able to analyze the data (here I count servers, storage, software as well as potentially additional people to perform data analysis).
One of the big challenges for all Big Data projects is clearly cost. This is especially important when you are looking at Finance, as the data is very structured and is already compared across industries in stock markets.
To finish, if the costs of big data analyses could be driven down by sharing an analysis platform (making analysis into a commodity where you pay per use) would you feel comfortable with sharing some sub-set of your internal data with other platform users (so that the data cost can be driven down for all)?