Promotions with Artificial Intelligence
Retail Promotions with Artificial Intelligence
Possibilities and potential in today’s environment.
Promotions in retail have always played a key role in driving traffic, price image, volume and ultimately revenue/ profit. While the process of funding and executing promotions has evolved over the years, the current system is cumbersome and slow. On an average, a full-cycle from planning a promotion to execution takes a minimum of 6 weeks to implement in a brick and mortar retail store. The online retailers are able to execute promotions much faster while the consumer demand is changing ever more rapidly. There is huge inefficiency in planning, executing and having to deal with an unsold product that results in expensive markdowns that go unsold.
Promotions can be made intelligent. The components needed to make a live, intelligent, promotion system are as follows:
The retail store can provide a wealth of information, especially when relevant data is made available quickly for consumption. Data sources can include point of sales, loyalty data, IoT devices, beacons, smart shelves, smart carts, cameras and traffic sensors.
The deep learning algorithms can discover a wealth of information underlying in the data especially when multiple data sources are processed in the same model. When live triggers are encountered inside a retail store – such as a change in weather or fresh produce condition, the algorithms can trigger meaningful automatic actions that can be implemented or sent for quick approval to the decision makers.
Electronic Shelf Labels/ Beacon/ Website:
The quick data gathering and instant output action that is computed by the neural networks ultimately needs to be implemented in-store or online. Retailers will need a system of Electronic Shelf Labels, Display Units, or Store employee devices to be able to execute the outputs. It is relatively easier to implement the output when it comes to online pricing.
Trade Fund Integration (Ideal):
Traditional retail promotions have been funded by Consumer Product Goods (CPG) companies. In most cases, retailers have an integrated trade fund solution that allows the buyers or category managers to exchange data on the Electronic Data Interchange (EDI). This system can be leveraged to present and get approval on instant promotions. The projected volume for different price points, including the optimal price point recommendations for maximum profit makes negotiation an efficient process. Further, this system can be leveraged to institute a process of “bidding” and automation of trade fund approval with a more mature and integrated system.
Once the connections are made in the Artificial Intelligence-based system. A wide array of possibilities emerges from the connected system.
Some of the use cases are listed below.
Weather Based Promotions:
Weather plays a significant role in the demand for certain products. Umbrellas or Gloves are some products where a significant relationship can be observed. In other products such as Beer and Ice cream, a less pronounced but still significant relationship exists. Based on the measurement of these relationships, combined with other factors such as inventory levels, predicted demand and product life cycle, promotions can be efficiently leveraged to move more goods and increased profitability based on the live weather conditions measured at the store level.
Demographics Based Promotions:
Everybody has a preference for their favorite products. Demographics as measured through census data has always played a significant role in governments and organizations changing the services provided in each geography. Now demographics can be measured inside a retail environment and this can be done in real-time. Using this information, promotions relevant to the demographics inside each store can be changed to increase promotion effectiveness and profitability.
Inventory Based Promotions:
Excess retail inventory can lock up precious resources of money and shelf space. While ordering systems have gotten very efficient at ordering the right amount for everyday sales, ordering for promotions is still a guessing process. Many expensive products get overstocked when the demand is less than predicted and the retailers do not have the bandwidth to resolve this problem at the store level. Automated promotions based on excess inventory levels can automatically put money back into the retailer’s hands while driving up revenue and clearing up store shelf spaces.
Local Event-Based Promotions:
Retail promotions can be planned for big events such as the World cup or Olympics. However, planning for local events such as a Downtown Concert or a local Football match oftentimes leads to “unpredicted” demand and loss of sales due to irrelevant promotions and stock-outs. A local calendar for each store location can be configured to connect with the main Artificial Intelligence promotions engine.
Shelf Life Based Promotions:
Many retail products come with a limited shelf life. This applies to products not only with expiry dates but also to products that are season dependent or to products that need to be trendy. Unsold products result in huge losses to margin, revenue and might contribute to food waste. Other unsold products create the need for expensive and loss-making markdown sales which are not desirable both from a margin and image perspective. Retail promotions can be triggered automatically at the store – product level based on the number of days left, inventory level and predicted demand to sell through at the maximum possible margins.
Holiday/ Season Based Promotions:
Holidays and Seasons have a significant impact on many categories of retail sales. While certain categories could see more than 2/3rd of their annual sales in a certain season, certain products actually see a decline in sales during a holiday season. To make things more complex, this change in demand is evolving continuously and reacts to competitive pricing at a different intensity at different points of time. These factors can be leveraged in an Artificial Intelligence-based promotions system to trigger promotions at the right time at the right level.
Competitive Promotions Based Promotions:
Competition can often get aggressive without business benefits. Retail promotions need not be the “lowest price” in all categories in all locations. Using deep learning algorithms can allow retailers to discover a healthy margin without losing sales. This leads to huge improvements in margins when executed at the store-product level. For example, a headset (or a banana) at an airport store should not have the same discount level as a suburban store.
A deep learning-based promotion system can be further trained to trigger promotions based on factors that can be triggered by one or a combination of factors above. Products and prices for promotions are selected based on historical learnings and current conditions resulting in increased promotion effectiveness and increased profitability.
To add the icing on the cake a fully connected vendor delivery system can be leveraged to negotiate with the suppliers based on volume and discount offerings on the fly. The category manager’s job turns from managing and monitoring planned promotion cycles to reacting to instant alerts and promotions that can be automated to run independently at each product store location.
Guided by the company objective which could be profits, price image or market share, the artificial intelligence backbone has the capability to increase gross margins in each promotion by 5% or more.
RapidPricer helps retailers to increase margin and reduce waste by automating pricing and promotion in real time unlike traditional consulting solutions.