Skip to contentSkip to footer
The apparent simplicity of a single product price for potential customers to consider belies the science behind the digits on display.
Consumer and business budgets have been severely tested in relation to recent inflation and interest rates rises – and supply chain challenges – bringing the construct of price elasticity into sharper focus.
So, when these macro, social, and industrial economic components collide, have organisations really understood how much stretch they have in what they are selling?
Pricing in retail businesses can be difficult. Usually, it entails setting price points for hundreds or thousands of products across many sites and different channels. Growing investment in Data Science capabilities has given greater hope to senior managers that they can master price elasticity to unlock the benefits of better pricing as a key driver of profitability. However, we have seen that the reality on the ground often falls short of expectations.
A major TV shopping channel has invested millions of dollars to recruit a team of skilled data scientists, equip them with the necessary tools and apply advanced machine learning algorithms to find the optimal prices for flagship products: however, only 20% of recommended prices are implemented. Elsewhere, at a large casual dining chain, prices proposed by analytical algorithms still go through a very manual validation process and management have remained limited in their insights regarding the price elasticity of the business.
In part, this gap between expectations and results is because the effective application of data science capabilities is not a ‘quick win’; it requires time to learn and scale, from data integration to engineering, from analysis to testing – and this is largely a structural lag that can be managed but not eliminated.
The gap is also the consequence of choices around what insights to prioritise, how to generate them, and which use cases to assign to data science teams. Many businesses seem focused mainly on hard measurements, rather than adopting a balanced and holistic approach that blends quantitative analytics with qualitative insights to paint a comprehensive picture of customer behaviour. However, these choices can be redefined and reframed based on first principles to achieve better results more quickly, with three key concepts to consider.
1. Price elasticity is best leveraged with broader customer insights
Price is an important component of the overall customer proposition and one that many retail businesses are so focused on that they may over-estimate its importance. For instance, branch staff of a major distributor of building products believed price was the most importance factor driving their customers’ choice of where to shop. In truth, contractors and small builders viewed the ability to find everything in one place and in stock as more important than the cost to them.
Understanding the role that price plays in the overall offer and how customers trade off price with other attributes is critical to effective pricing strategy and optimisation. The ultimate objective should be to grow customer lifetime value, rather than the profitability of an individual product category. This means knowing how far pricing can be leveraged and the point at which it will become ineffective because other offer levers will assume greater influence on customer choices.
2. Price elasticity plays out across many dimensions
Price elasticity is usually looked at within a product range, where it varies significantly both by category and across the different levels of the product taxonomy (down to individual product lines or SKUs).
At the same time, elasticity differs by channel and by site. At an individual location level, it reflects the local competitive environment and demand profile. At a more aggregate level, it changes by format and the related valuer proposition (e.g., the breadth of choice of large supermarkets vs the convenience of corner shops).
Finally, elasticity varies by customer too: both at a segment level (e.g. by socio-demographic, attitudinal, or need-based cluster) and by shopping situation (the same customer may choose to go to an expensive restaurant for a special occasion while visiting a cheap café to grab a sandwich and soft drink on the move).
The implication is that price elasticity readings should be best thought of as a grid or a cube rather than as a single number. Understanding how elasticity changes as you move across that space is a key insight to tune pricing decisions across different parts of the business.
3. Price elasticity is better measured broad than narrow
Efforts to measure price elasticity often focus on individual product lines. In part this reflects the importance of certain SKUs (so called Key Value Items or KVIs) in shaping customers’ perception of a retailer’s prices. In addition, it is driven by the operational requirement to set a price point for every single product, and automating this task can be a valuable productivity gain.
However, analysis of elasticity at product level can be challenging – the level of noise is usually very high and controlling for the large number of factors influencing sales of specific SKUs is tricky at best. These measurements also reflect cross-elasticities among products in the range as they also capture the impact of mix shifts within the pricing architecture. Therefore, they can over-estimate the overall change in volume and revenue for the business.
Price elasticity principles
Managers should ask themselves “what do we really need to know about elasticity to define better pricing strategies that grow customer value and improve financial performance?”
In our experience, businesses are best equipped when they have insights on elasticity that show how the impact of price:
(a) varies across the different parts of the business at a level of aggregation consistent with decision-making for commercial strategy, plans, and target setting; and
(b) maps to different customer behaviours, to understand how customers are responding to price changes and how to mitigate adverse reactions.
In the Hospitality example above, the Seat-in format had high overall elasticity mostly driven by mix shift suggesting that customers adapt their product choices to higher prices, so the selection and cost engineering of menu items was critical to protect margins; the Grab & Go format had low overall elasticity with a high share of customers walking away and not shopping in response to higher prices, so careful use of entry lines helped reduce volume loss.
These principles allow retailers to prioritise the insights that matter most on price elasticity and harness them effectively to grow customer lifetime value, improve financial performance, and realise ROI on Data Science investments.
How far is your business on this pricing path?