Блог Datawiz.io

Price Elasticity in Retail

Antonina KelmanMarch 10, 2016, 4:09 p.m.

When retailer lowers product price, consumer demand is growing, but there is no profit. If retailer increases product price - demand falls.
 

So as you can see there is a strong correlation between price, demand, and profit.

Elasticity model gives you the opportunity to find the optimal ratio of cost and number of sales, which allows to increase profit. In this case, the price stands as a balancing factor between expected demand and revenue.

We propose a model of price elasticity which works for correlation of product prices in different shops. It helps our clients to increase everyday profit with clever price reduction that affects customer demand.

Many existing models of price elasticity claim to "world domination". But it is impossible to consider all factors that affect on price formation in one ideal model of price elasticity.

Our approach for price elasticity in retail has such features:

  • product price correlation for each shop, not for whole retail chain, it helps to make more accurate recommendations, analyze customer demand for each shop type, assortment, placement;
  • historical sales data for half-year and more than 4 changes of the product price for adaptive price formation and exclusion of risks of inflation or another external risks of retail market;
  • correct predictions for right products with the system of exclusion of controversial data;
  • data visualization of main steps for best price definition which helps certificate that the decisions for price change are right.

How model works?


For example, lets choose sales of one product for the last half-year and find recommendations of best price for it with the help of BI Datawiz.io.

 


On a graph, we see how changes of the price (red line) influenced on product sales (blue line) for six months.

First lets track fluctuations in demand, depending on price. We noticed that the changes in demand located near each price change range. So we can group this demand ranges for each price meaning. We exclude from the analysis of all single sales as invalid.

In food retail, where the prices do not change dramatically, it would be logical to group sales price changes with increments of 0.5 $. Then determines the average number of sales in each group.

 


We arrange the data points of average sales for each product price group on the chart.

Next graph clearly shows that there is a relationship of the data and we can define a linear regression (trend).

Experimentally, we have determined that the model can be used, if the price of a product has changed at least 4 times, otherwise the result will be inaccurate.

 


Second, we construct a model of the dependence of profit from product price. To do this, we need to determine the dependence of the price margin and everyday profit.

Profit is based on everyday sales, multiplied by the margin of the price.

On next chart, we build regression line to find optimal price for each product.
Vertical - day profit.
Horizontal - the price.
The line begins construction from a point of product self-price where profit = 0.

 


So we can find the point on the graph where the profit is maximum and define the value of the optimum price.

The model shows recommendations for lowering the price from the current to the optimum, which will bring additional revenue by increasing the number of sales per day.

Our model excludes recommendations of increasing product prices because it is necessary to consider many external factors: competitors prices, pricing policy of chains, government guidelines and restrictions.

Also, service BI Datawiz.io build recommendations of price change for each shop with the list of products. The example you can see on the image below.

 


As you can see our approach of determination the optimal price is close to the practice and the real situation in retail business.

Our model use the time factor, frequency, number and stability of sales. It provides an opportunity for self-analysis through visualization of historical data changes in prices and demand.

Service BI Datawiz.io treats all sales data and displays the results in a list of products, where the price should be changed. We increase everyday profit of our clients for more than 10% in each shop.

Stay Connected