By Alex Bordei
When you start a company, Venture Capital ignores you, crowdfunding does not help as you need to pay for expensive marketing for your product to get attention, you are alone, and you base your business decisions on your intuition alone. You have an ally you do not know about, it is called Big Data, but you keep avoiding it. I am here to change your mind.
Breaking the “Big Data, Only for Big Businesses” Myth
It may very well be the adjective ‘Big’ that confuses people, but when people hear ‘Big Data,’ they mind-connect it to big corporations like Facebook or Google. Yes, these corporations take advantage of big data and machine learning the most because they need to. Otherwise, their business models would not work, and they would drown in data. For example, Google performs over 40,000 searches each second, so targeting ads to users based on their searches is impossible without big data.
Let’s follow a simple scenario: you buy 10 machines, and you connect them to sensors connected to the internet (IoT sensors). Let’s say these sensors send all the info about the machinery a couple of times per second. One year passes, and you are already drowning in data. Look at your hard-drive, at the pictures you took, at the memes you saved, or at the music you stored. If you are a data hoarder like I am, you may even have video card drivers from 1998 on a CD somewhere.
Today, gathering data happens, especially if you are a company. You may label that data useless, but first of all, it will clog your operation in time. Moreover, that data is wasted: you collect it and, instead of processing it and making use of it, you deposit it, and it eventually clogs your operation.
A Simple Use Case
Let’s use a simple example for this small piece: a small online bookstore startup. This raises a couple of questions. Amazon sells Kindle devices, runs powerful big data, and can give people suggestions. Still, your online store begins to sell books, in part because you have a specific niche that you address, or maybe you cover more of a particular language that Amazon does not include.
For you to survive, you have to provide similar services in time, targeted at your customers, and boosting your sales, or they will take that market as well. It is imperative that you know what your top selling books are and how often they are purchased, or to record searches on your website to find out what books people are looking for and you do not sell.
Market-basket analysis helps you suggest new books to your customers, better fit to their needs. Let’s say two customers John and Susan buy the same three books in the romance genre. A week later John also purchases a fourth book, “The State of Desire,” and Susan does not. Doing big data and applying simple machine learning on that data can suggest “The State of Desire” to Susan, making it highly likely that she purchases that book from you. That is one less book sold by Amazon, more data about your customers that can boost the next sale, one happier customer, Susan, who found a new cheesy romantic book to read, and one extra book you sold.
All of the above used to be done with rough estimations, but now big data stepped in and took over, especially if combined with simple machine learning algorithms for prediction. The data you collect can help our book startup use case in so many ways: you estimate book stocks, your resupply times improve, you rarely run out of stock for bestselling books, and you prevent overstocking.
As a new online bookstore, you have little data compared to the big players like Amazon, so you better make the most of it.
Conclusion
Even if doing big data on what ‘little’ you have as a startup may seem useless, it is not. It forces you to think in a data-driven way that will be crucial to your operation in time. You make use of the data you have at first, learn how to analyze it, and build a framework that helps you grow.
In time your data collection capabilities will increase, but your data structure is already in place, growing alongside your company. When your operation rises, you need only scale it, and many cloud services are extremely scalable.
Alex Bordei is Director of Product and Development at Bigstep, a company that empowers organizations determined to make sense of their data, by providing a full-stack big data ecosystem running in a high-performance bare metal cloud. A highly technical professional with over 10 years of experience in architecting and developing high performance distributed services for the cloud market, he has an MSc in Computer Science and has always been keen on research in advanced software technologies.
Big data stock image by GarryKillian/Shutterstock
