19.8 C
Wednesday, July 24, 2024

No products in the basket.

HomeSales and MarketingDigital MarketingHow giftware retailers boost revenue with the help of machine learning

How giftware retailers boost revenue with the help of machine learning

What do you do when you need to reprice some 4,000 gifts sold online? The £1 billion UK gifts market, let alone its e-commerce segment, is very dynamic — so, your decision is required immediately. What’s more, you probably have to follow suit of the market leaders when it comes to pricing. All you can rely on is your own expertise and the endless lines of Excel spreadsheets. Lacking time and necessary data, you’ll probably try to calculate optimal prices for the best and worst-performing products, while the rest of your product portfolio would be left with suboptimal prices. That’s problem number one.

Problem number two is the market becoming more intensive. As a result, retailers would do anything to attract and retain customers. They may tap into a deep 30% promo for a group of products and boost sales significantly, especially in the weeks leading to holidays like Christmas. For some gift sellers like UK-based Find Me a Gift this is an extremely high season which accounts for some 50% of its annual turnover. But even after making £15.89 million in sales in 2018, up £3.16 million from the previous year, the company realized that soaring sales do not necessarily convert into growing revenue.

By changing a shelf price or launching a promo for a specific product, retailers inevitably influence the revenue and profit of the whole product category. Therefore cutting sales blindly and pricing only a part of assortment optimally are not an option anymore. Pressured by the growing competition and the ineffectiveness of the old pricing methods, Find Me a Gift decided to cooperate with Competera, a retail price optimisation company. The partnership aimed to help the retailer revamp its pricing strategy with the help of machine learning algorithms and allow for maximising revenue and sales. “We were running around selling lots of stuff, but we wanted to find a way to make each pound work harder for us,” commented Jean Grant, purchasing and product development senior manager for the company.

Throughout a market test, the retailer applied price recommendations generated by the self-learning algorithm for a group of nearly 600 SKUs. A dense neural network powering the algorithm would calculate optimal regular and promo prices based on a wide range of pricing and non-pricing parameters like price elasticity, website traffic, competitors’ prices, and seasonality.

During the five weeks of the market test, Find Me a Gift managed to improve its item sales by 24.7% and increase revenue by 9.3% for the selected 600 products versus the rest of the retailer’s assortment which was repriced traditionally. It happened thanks to the algorithm factoring in thousands of hidden relationships inside a product portfolio and suggesting individual prices for every product, which altogether boosted sales and revenue of the total portfolio. Previously retail managers simply could not consider all the necessary interconnections between products to make proper pricing decisions as they lacked a computational power to process vast amounts of data in real-time.

“I wanted somewhere people can come to and find perfect gift ideas for all occasions,” Adam Gore, CEO and founder at Find Me a Gift, stated in an interview. With algorithmic data-driven pricing, this venture becomes not only pleasant but profitable.

All things considered, staying relevant to the UK giftware market is getting more difficult. Although retailers succeed in attracting customers by slashing prices, they still do not get what they want — to earn more. A growing number of retailers acknowledge that traditional pricing approaches are not as efficient as they used to be. As a result, businesses are embracing AI to optimise their pricing and boost financial performance.

*The article cites a study by YouGov and numbers from Endole.

Popular Articles

Latest Articles

Related Articles

This content is copyrighted and cannot be reproduced without permission.