In my next life, I want come back as an economist.
“Assume a demand curve of 50000 units – Price * 5000 units”
If only it were that easy.
The reality is we don’t have neat little price-volume relationships handed to us by the market place. At least not in the form of pretty little equations we can feed into a routine to determine the optimal price.
We have data points, specific measurements of how the market responded to a particular offer. Usually acquired at high cost, especially if we’re selling in retail or another channel where we need to make the same offer to everyone who walks up the point of sale. Data points which are affected by other factors:
- Audience Differences
- Product Mix Featured
- Display / Placement
- Competitor Pricing
- Substitute Products
- Recent Promotions
- Channel Support
Unfortunately, each of these factors also affects demand. In many cases, at the same level of effect as changes in pricing. This makes measuring price elasticity in the real world extremely tricky.
There is also the question of how the incremental demand in response to a lower price is generated:
- Was our product sold to new or existing customers?
- Did existing customer load up, potentially reducing our future sales? (or did we prevent them from buying from a competitor when a similar product gets featured next week….)
- If we generated new customers, was this from broader exposure or better conversion?
This last point is easy to measure in digital but much harder to quantify in retail and wholesale channels. There are generally multiple interactions between price level and channel partner support. You put your best deals at the front of the store. The lift from a better price is a combination of:
- More people seeing the offer (impressions)
- More people accepting the offer (conversion rate)
It is even harder to interpret business-to-business results or service purchases. Business buyers tend to be process driven, working within a standard procurement approach to qualify and select items. There is often a significant lag in reactions to market events due to review cycles and the time required to execute significant changes. You can also lose business for factors unrelated to that specific product and price, such as competitive pressure on other items in the bundle or broader challenges with customer service. Customers will frequently cite multiple points when you debrief them about a decision. Price is one of many factors: “you raised your price AND you were late on a shipment AND your rep never visits me”. These dynamics make it difficult to properly attribute business gains and losses to pricing actions.
This doesn’t mean you can’t measure price elasticity. If we can construct a controlled experiment – where we run the same promotion, with similar advertising and channel partner execution , at different prices – we can generally infer that most of the remaining difference is due to price. Particularly if we can repeat the experiment a few times. Those measurements of price elasticity should be valid.
Another element to consider in real world pricing analytics is the emergence of “winner take most” dynamics in many channels. We see this in digital, with Google search and product selection engines. The first result gets most of the customer attention and the audience drops off quickly as you move down the page. The same effect can occur in offline channels. A wholesale rep has a list of deals and the time to present only a handful of them. Imagine the value of being first out of their selling bag vs. fifth. There are handful of prime display placements in a retail store – yet plenty of room on a quiet retail shelf, seen by only a few. Location matters.
The impact of “winner take most” dynamics can be huge: the first result in Google can get over 50% of the clicks. The fifth result usually generates around 5%. The 10th result (still front page!) can generate as little as 1%. Anything past that is fighting for scraps: only a tiny share of searchers even see page two.
Ranking systems exist for the benefit of the customer: in a world of many choices, simplify their view so they can review a manageable number of offers. And both Amazon’s search algorithm and your friendly local wholesale salesperson have the same objective: answer the customer’s request in as few “clicks” as possible, to conserve time and keep the customer satisfied. The degree to which your offer “converts” business helps you improve your overall position, visibility, and potential audience within a channel.
Which introduces a fascinating opportunity, if your management team is thinking about price elasticity: what happens if we can use price as part of a larger effort to move up in a “winner-take-most” ranking?