The constant and proceeding emergence of deep learning wouldn’t have been possible without the soaring availability of big data. According to a Vouchercloud – survey 90% percent of the world’s data has been created in the last 2 years. That means a daily 2.5 quintillion bytes that could fill more than 10 million blu-ray discs. Let’s have a look at how this tremendous amount of data empowered strategic marketing and advanced customer service in the e-commerce field.
Gone are the days when the proverbial saying prevailed: „half of the marketing costs are wasted, however we don’t know which half”. Processing the huge amount of data assists e-commerce businesses in optimising prices, forecasting demand and predicting customer trends on the corner. Strategic decisions get more and more backed by big data amounts.
Big data helps businesses to determine the most favorable and convenient price by keeping an eye on competitor transactions, transaction frequency and pricing, inventory and forecasted inventory run-out. The reaction ( falling or rising demand) of increased and decreased price can be captured immediately and corrective actions can be started in matter of seconds.
The visibility and „graceful eventlessness” in an e-commerce supply chain management is a great asset. With the help of big data the store can pre-define the anomalies hence evades running out of stock and can be prepared for the waterfalls of seasonal orders. The unexpected delays in shipping or the rapidly dwindling stock of a wanted product is detected and can be corrected immediately.
The question to answer: what are the items you must have to ride the approaching trend wave? Sentiment (also known as opinion mining or emotion AI) analyses are driven onto the online discussions – around what products or product categories are there the most robust conversations? Are they negative or positive? The advertisement buying data is also analysed to figure out what are the hottest product categories being marketed and the forgotten or sinking ones that are non- or less and less marketed.
According to an American Express Survey 78% of the leaving customers leave the intended purchase because of the poor customer service experience. „Understanding Customers” by Ruby Newell-Legner delivers an experiment proving that it will take 12 positive experiences to make up for one unresolved negative experience. Needless to say that a very few customers will give further 12 chances to a web store when hundreds of other stores compete to fulfill the customers’ needs. With the help of big data e –stores can comb the various communication channels ( email, phone call, chat ) together so that customers don’t have to file their problems and personal information several times. Also, big data reveals which issues are the most accumulated and explores the time periods when the most intensive customer support is needed.
Let’s have a quick example how significant the difference is: 1. a cart abandoner customer gets a flat notification, or 2. a notification enriched with a personalized offer. A large amount of customers turn back in the last minute without conversion, but they leave a fair amount of personal data even if they are first-time customers. Their profile gets created in seconds from their footprints left onsite, their browser/referrer/location and device info, their preceding and even their successive social events… and the first recalling letter can be released right after the abandoned purchase or after a few hours or days as well.
Keeping stock on shelves burns money therefore demand forecasts should be more and more accurate. Using historical data Amazon is able to forecast the versatility map of the year broken down to holiday seasons and geographical locations. Not only the traffic volume but the possible conversion rate can be predicted and significant spikes can be covered by the Amazon Web Services cloud servers.
Big data ushers e-commerce businesses into the golden age of more personalized offers. From now on stores can connect the tiny data spots and swatches of recorded user events like browsing habits, logins, purchase history etc. to draw a more and more realistic user profile. For example Cora Romania using Yusp’s solutions (personalized couponing, online personalization initiatives, target customer selection (user-to-item) and in-store personalization) has attracted + 10% converting visitors and experienced a 5 million USD increase in annual revenue.