Forecasting tools for retail and e-commerce are booming. Not surprising, because it’s essential in the fashion industry to predict what you expect to sell as accurately as possible. In other words: the goal is to gain insight into consumer buying patterns, preferably even before the marketing season starts. With the help of machine learning and artificial intelligence, it’s possible to make these predictions with extreme accuracy. For example, over the expected number of customers in the stores, the number of articles sold and the coming fashion trends.
Distorted view
Predictions about the future are always made based on data from the past and present – known as predictive analytics. “Due to the Corona crisis, fashion companies are now confronted with ‘polluted data’. Data that gives a distorted view”, says Xin Zhang, business consultant at Chainbalance. While many webshops are unexpectedly working overtime, retail stores have seen their turnover come to a complete standstill during the lockdown, leaving them holding surplus stock. “That means you cannot trust the calculations made based on this data, because it gives a completely different view than usual”, adds Zhang. “For fashion stores it’s important to keep this in mind, otherwise they will always make inaccurate forecasts.”
Personalised product range
The ‘personalised stores module’ from Chainbalance allows you to fine-tune your assortment at each step based on consumer data, taking into consideration the distorted data from the last year. “Stores are frequently initially stocked based on a standard profile. But this can be made more specific per location: retail is detail”, says Maaike Poelmans, operations director at Chainbalance. “Say a certain bra at Triumph is a strong seller in most of the stores, but this item is not included in the assortment at the Maastricht store. Why wouldn’t you stock this article in Maastricht as well? If this article is a best-seller in most of the stores, even if there is limited stock in the distribution center, it would be smart to produce more of them as fast as possible. The personalised stores module gives our customers these sorts of prompt insights and recommendations, so they can adapt the product range in the stores to the actual demand.”
Flexible approach
Insight into the actual local consumer sales, enriched with data from that specific retail area, allows retailers to personalise their product range per store. “Our philosophy is that this is a very dynamic process”, says Poelmans. “Our software generates orders and takes into account questions such as: how much time passed before the article was sold? Is there enough stock to make it through the week? If not, the system sends an order to replenish this article. In short: we choose a flexible approach by taking into account many different factors. The goal is to make sure that there is always enough stock in the store or webshop, but also not too much.”
Optimal assortment
This often results in changes to stock target levels that retailers would never come up with on their own. Poelmans gives an example from an outlet store in Roermond. “This customer of ours always had three pairs of a certain model of jeans in a specific size in stock. They were always sold within a few days. That’s why we increased this number to six pairs. What happened? The store also sold the six pairs within no time! Even when we raised the number to twelve, the stack of jeans on the shelf was gone in no time. The merchandise team was amazed. This shows how valuable it is to be able to act promptly based on the actual demand of the consumer.”
Further improve results
Improving sales by adapting the product offering to the consumer demand works for webshops and physical stores. “The fashion industry is complex: you are dealing with different models, sizes and colours. Especially for complex products, such as bras, blue jeans and shoes, our system makes a big difference for our customers. Do you really need all of the sizes in all of the stores? Or can you limit your size curve to free up more square meters for other products? Which assortment is the best to offer online and how do you make sure you have enough stock for your online channel?” There’s a reason the slogan of Chainbalance is: reinventing your inventory. That is the most important starting point for this Dutch company: using data in a smart way to make sure that fashion companies achieve the best results with their inventory.