How do we price goods and services? As business leaders, we have asked ourselves this question since the history of trading.
Today, many companies employ people to analyse a multitude of data to determine the best pricing approaches to take. But how successful are these approaches really, and how do we know whether we are leaving money on the table?
The problem with econometric pricing
Econometric pricing models work by looking at fluctuations in historical data in terms of sales and pricing of products and services, to determine their price elasticity. This model takes into account both changes in products’ pricing and competitor’s changes in price. So far, so good.
However, digging a little further into the data used to provide this analysis shows us why this model is flawed in specific situations.
Imagine a scenario where all prices rise by the same relative level. The econometric model will say that if the price of products within a category all rise by the same degree, then sales volumes will remain constant for those products. However, this is clearly not the case. If spending power decreases, or remains constant, and prices increase, this can create a situation where lower priced products win more disproportionately than what these models show, which can have implications on future price changes.
We also know when we look at how volume reacts to price changes, it isn’t a constant. The elasticity can and will change at different price points. A $0.05 increase will not have the same impact at $1, $2 or even $10.
How do we then determine a price’s reaction at different price levels we may never have seen in a market before? Why should we assume consumers will react as they have done in the past?
Defining the impact of promotions on sales volumes
Often, we are dealing not only with price changes, but also promotional changes, and in many cases, businesses will refer to a base price and a promotional price. The price of an item may be $10, but this week with 20 per cent off it may be $8. This creates complexity when it comes to understanding the price impact versus the promotional impact on sales volumes.
Many companies undertake price and promotion modelling to pull apart these two mechanics. In most cases, they will see a large variation from their promotions and therefore good data to determine sensitivity. The same amount of variation is unusual with the base price, however. The base price changes infrequently and therefore it is difficult to understand the impact of a change, as we don’t have a lot of data points to determine sales volumes as it changes.
One key issue with both the econometric and price versus promotion models, is that neither can look beyond what has been done before - they are constrained by the past. If, for instance, a business decides to promote a product that has never been promoted before, these models aren’t going to be very accurate. Further, if a business decides to undertake an EDLP strategy for a product, there is no existing data on its likely impact.
With Coles and Woolworths announcing they are moving many brands from Hi-Lo promotional pricing to EDLP, we have no historical data for comparison, so the results will be interesting to observe.
Some companies are using ‘dynamic pricing’ – basically, test-and-learn - to determine pricing in the digital environment. Under this approach, the price a consumer sees changes with demand. Travel is one of the industries constantly adjusting prices to demand.
Unfortunately, a reactive pricing model still leaves money on the table, either by discouraging consumers to buy when the price reaches a certain level, or closing sales at a certain price where consumers may have been willing to pay more.
Putting the consumer at the heart of pricing
These models are reliant on looking at historical data at a product/SKU level and determining outcomes. To improve on this, we have to put consumers at the heart of our models and be able to create data about the future that is not predetermined on historical constraints.
How do people determine what they are willing to pay for a good or service? Behavioural economics tell us different variables can change how consumers view price and essentially choose which products to buy. Reframing the category, different purchase occasions, increasing the number of items available for purchase, or introducing products at lower or higher price points, can all affect how consumers make purchasing decisions.
Models that determine how people make choices and how price, products and range influence their choice, result in a more informed way of driving revenue. More importantly, they use past, current and possible future pricing states through designed experiments that can be easily administered to provide proven accurate forecasting of sales, revenue and profit.
These experiments allow us to look at millions of possible pricing scenarios. Through sophisticated analysis, we can then determine at what prices we can maximise financial outcomes for commercial success. By putting consumers and customers at the heart of pricing, we can determine the amount of support and distribution that is required.
Rather than looking at the market as a whole, we can analyse the market as the sum of its parts: Different groups of people, geographies or channels.
By using this approach, businesses can implement more targeted and informed pricing strategies, generating hundreds of millions of dollars in extra profit. Understanding the value that customers put on the product, and not an arbitrary number, shows businesses the benefit of changes in price in terms of their customers’ preferences and what they will actually accept.
If you want to drive the pricing agenda rather than react to the market, then put the consumer truly at the heart of your pricing strategy.
This article was originally published in CMO Magazine https://www.cmo.com.au/blog/customer-credentials/2019/08/13/how-to-create-profitable-pricing/