A few weeks ago, our team was at the Big Data & AI Paris tradeshow to present our products and meet prospects. One question was on everyone’s lips when reading our “Small Data, Big Impact” motto: what are the differences between Big & Small Data?
First, let’s state the obvious: when it comes to data, size matters. Big Data relies on petabytes, while Small Data is much lighter. Cloud or on-premises infrastructures host Big Data infrastructures. Meanwhile, computers or local servers of Business experts are full of 300-row, 15-column spreadsheets, no matter their department or industry.
Business Experts are familiar with Small Data
Tabular structured data, i.e. properly filled and labeled spreadsheets with rows and columns, is what makes Small Data. Hence, it is transparent, actionnable and easy to understand by non-data people. On the contrary, Big Data sets consist of very different types of data. From structured data to images or videos, they require a lot of preprocessing to be analyzed by data scientists.
Furthermore, Business experts are already familiar with Small Data sets. For example, customer care teams know how to interpret a satisfaction survey and what conclusions to draw from it.
Structured, transparent sets of Small Data are present on every business expert’s computer and updated quickly. Business experts can visualize them through Excel, and some of the key benefits start to appear, although additional solutions can be used to understand their inner workings:
- Small Data is simple: it focuses on one topic at the time to provide detailed information in a structured way. It is natively more understandable than Big Data;
- As it is close to business experts’ knowledge, Small Data provides immediate actionable insights.
Now that we’ve seen some of the key characteristics of Small Data, let’s have a few words about Big Data. There is a lot to read on the topic, so we’ll keep it short – we specialize in Small Data, don’t we?
Big Data is all about huge amounts of data and infrastructure.
Insane volumes of structured, unstructured and semi-structured data, complex Cloud-hosted infrastructures form a maze of information to untangle.
That is why Big Data can’t be natively understood by Business experts. Data analysts and scientists have to refine, clean and engineer data through tailored data pipelines and visualization tools. They then automate the process to deliver Data insights to Business experts.
Companies gain tremendous advantages from their data: better understanding of their customers, new market opportunities and trends, demand shift anticipation… Thus they can adapt their overall strategy according to these new insights.
Some companies actually rely almost exclusively on Big Data for their Products and Solutions. For instance, the Netflix algorithm recommendations are based solely on user data so that you find a show you like and you stay on the platform.
Of course, building and maintaining a Big Data infrastructure, hiring the right data persons and delivering the right insights takes time and money. This is why organizations only have a few Big Data projects running at Enterprise level. On the other hand, Small Data projects rely on already existing data and infrastructure and are easier to set up.
Small Data’s beauty resides in its simplicity while Big Data performs in complexity.
Small Data will change business experts’ everyday life by helping them answer the business questions they face. Meanwhile, Big Data will help companies evolve, adapt their strategies and create new products. Both approaches are complementary: they address different ladders of the company with a similar purpose of becoming a data-driven organization.
To help understanding and remembering, we created this small chart with the 6 Vs of Data and how Big and Small Data position themselves within this grid:
