The Berlin technology company 7Learnings offers AI-based software for forward-looking pricing for retailers: inside. On the NRF Retail Big Show, Fashionunited spoke to Eiko van Hettinga, co -founder of the company. Van Hettinga explained why pricing is one of the most important levers for profitability. He also explained how forward -looking pricing works in practice and how fashion companies such as the Tamaris brand belonging to the Wortmann Group achieve results.
Why should brands practice a forward -looking pricing?
The pricing is the biggest lever for profitability. Many companies are thinking about cost reductions when they talk about profit. However, the truth is that price movements have a much greater influence. That is why analyst describes: Inside, like Gartner, price optimization is one of the most attractive AI applications in retail: it has the greatest business effect and is one of the most practical.
For retailers: Inside who are asking where to start your AI trip, the price optimization should be at the top of the list. With forward -looking pricing, you use data to predict the effects of prices on your KPIs – such as sales, margin and sales – and then optimize accordingly. This is exactly what we demonstrate together with Tamaris.
How exactly did that work with Tamaris?
At Tamaris, the challenge was clear: they expanded their online business into 26 countries. The complexity of determining prices across markets and channels was enormous. There was a lot of manual work and the need to optimize along the entire product life cycle.
Together we carried out a five -month Proof of Concept. The results were impressive: profitability rose, the average discount rate dropped by five percent. The manual expenditure of time for price optimization was halved. Today Tamaris uses this AI-controlled setup in all markets and controls prices and margins automatically.
So is a flexible pricing instead of just pulling discounts?
Exactly. It’s not just about hanging a large red sales sign in the window. You can also cut the prices strategically and still keep your margins. In fashion retail we do everything for fashionable reasons, so we continue to think of discounts.
What can you say about the complexity and the challenges of online price design?
There are additional levels online such as vouchers and coupons. There is a risk of easily losing control of your profitability if you stack too many advertising campaigns. That is why we also feed this type of data into the system to predict how many of these advertising campaigns are used and what effects they have on the profit.
In fashion retailing, we also predict the return quotas together with the prices. Over the entire range, our forecasts achieve an accuracy of more than 90 percent, for a two -week horizon. We believe that the right approach is a combination of highly precise short -term forecasts and long -term planning.
What about short -term vs. long -term forecasts?
One could say: Why not make every decision based on a 40-week forecast? The problem is that such long -term forecasts are very imprecise. You just don’t know what will happen so far in advance. That is the big challenge in fashion.
We use long -term forecasts to set limits, not to dictate every decision. The algorithm could, for example, calculate the price that optimizes long-term profit and then allow us to act within a 20 percent range around this point. Within this range we can make short -term decisions, such as boosting the sale faster. The system also prevents us from going so far that we affect long -term profitability. Technically speaking, we believe that this is the best way to solve the problem – and practicers in this area confirm this approach.
Can the collection development be determined with AI?
Not really. As soon as the collection is on the market, we can help with the initial pricing, but this part usually has more human influences. When a dress comes onto the market for the first time, we can look at its properties and compare it with similar articles to propose a price. But if you believe that it is an outstanding piece, the human judgment comes into play, because the machine will not see it. Over time, while the product goes through its life cycle, the system is increasingly learning from transactions and attributes to improve its price decisions.
How did 7Learnings be founded?
Our CEO, Felix Hoffmann, spent his entire career in the Pricing area. He initially worked for consulting companies like AT Kearney and was later responsible for the price algorithm near Zalando in Berlin. At some point he realized that you can’t work with Excel forever – you need something more technical. This is how the idea for 7Learning came into being. Today we are an independent company.
7Learnings also worked with retail companies such as Tom Tailor and Mister Spex and helped them implement forward -looking pricing. The start-up was founded in Berlin in Berlin by Felix Hoffmann, Eiko van Hettinga and Martin Nowak.
This article was used with digital tools translated.
Fashionunited uses artificial intelligence to accelerate the translation of articles and improve the end result. They help us make the international reporting of fashionunited a German -speaking readership quickly and comprehensively accessible. Articles that have been translated using AI-based tools are read and carefully edited by our editor: Correcting inside before they are published. If you have any questions or comments, please contact me by email to [email protected]