How Analytics Can Inform Product Pricing


How analytics can inform product pricing

With long-term closures and a decrease in consumer spending, retailers accumulated inventory resulting in tied up revenue and additional expenses. In many cases, the stock consists of seasonal goods that need to be discounted and liquidated within a short time frame, while still generating enough cash flow to enable the business to invest in future deliveries.

Identifying a successful pricing strategy has always been considered both a science and an art. However, the need to discount heavily to keep the business going after COVID-19 has become a necessity for numerous retailers. The complexity of creating strategic pricing comes into play when a retailer decides to mine historical data to see whether there is an opportunity to implement a fine-tuned promotion, rather than a site-wide final sale.

In a traditional approach to pricing, a product goes through progressively higher discounting toward the end of its life cycle to generate maximum revenue before the liquidation. Brand strategy determines the overall level and structure of discounting, which is, in turn, tied to seasonality, product life cycle, competitive pressure, and consumer response.

The impact of the pandemic has tightly compressed the typical product cycle and elevated discount levels far above the previously accepted norms, whether in absolute level or percent of total store sales.

Look for opportunities to lower discounts

Reviewing past promotions can prove useful in understanding customer response and estimating the profit of future discounts. However, this will inevitably raise the question of whether there is an opportunity to decrease the discount while still garnering a strong response to generate the same or even higher margins.

Data science can help identify the optimal level of discounting to reach desired targets, especially if these levels have not been tested previously. When deciding how to price your inventory to move it quickly and generate the most margin, consider the following: the deepest discount is not always the best approach.

In order to identify an optimal discount rate, consider the following:

1. Can discounting any particular product or category have an impact on the entire store?

Before tackling basket analysis, check the available data to see whether a specific product category has an undeniably significant effect on the sales performance of the entire website or store. If yes, identify the optimal discount level for the product that resulted in the most significant ROI for the whole store. The identified optimal discount can frequently end up being less than the previously deployed final sale.

2. Perform basket analysis

Next, utilize basket analysis to identify product categories that sell best together. At times, it’s possible to determine which of the attached products drives the sale of the entire basket and even at what discount rate. Historical sales analysis can help to fine-tune discount levels needed for each of the products that have a proven attachment to each other. Identify the best discount rate intersection between the two. Consider the possibility that some products may be more price-sensitive than others and use them to your advantage.

The challenge in identifying product combinations that can be leveraged while creating a promotion lies in separating causation vs. correlation. Are customers adding both products to the cart because they are merchandised next to each other in the store or on the website? Are they always discounted together during past promotions?

3. Analyze historical sales to gather consumer response to previous promotions

Once proposed product groupings have been identified, the next step is to determine the discounting level for each product that will cause the best consumer response. Analyzing previous sales data can help identify the best level of discounting by measuring the highest unit and dollar sales, as well as the margin generated in each promotion.

Strategic promotions utilize historical data along with data science driven forecasts

Based on the above, create a promotion that combines both the science of the historical analysis with the art of consumer engagement. Once a promotion with the identified product groupings and discount levels is created, identify the best timing and consumer segment to implement it.

Customer segmentation can provide useful insight not only about who would respond best to the promotion, but also who is best suited to be a trial group (i.e., You would not want to spend ad dollars on showing sale merchandise to full price clients).

Finally, it’s essential to consider the impact of deep discounting (no matter how strategic) on the brand’s future. At times, liquidation can be a less profitable, but more sustainable in the long run, approach to preserve the brand image.

To formulate a successful discount strategy, utilize the following levers: identify store driving products, isolate symbiotic products using basket analysis, and use historical data to determine the optimal discount strategy. Historical sales data can reveal a plethora of insights that can help retailers create a strategic approach to discounting their overstocked inventory. Further, these insights can be applied to future promotions to help streamline discounting and improve margins.

About the Author

Yana Averbukh is the founder of Green Retail Consulting where she develops and implements retail strategy solutions driven by analytics. She has an MBA from Baruch College and holds a Microsoft Data Science Certification. See Yana’s work on her website, or follow her on LinkedIn.

Categories
Search By Tags