Chapter 8. Pricing Optimization

Split Testing | Dynamic Pricing

Previous Chapter | Next Chapter

[3 min. read]

We’ve been painting with a broad brush so far.

Big pricing decisions.

But what about minor tweaks? 

I like to view pricing as a project that is always open. You can make the big decisions fairly quickly. Then most of the time is actually spent optimizing and making small adjustments to extract maximum value.

8a. Split-Testing

Minor changes are not usually made based on some pricing principle or best practice. They’re made as a result of the scientific method in the form of testing.

The most basic test you can run is a split test or A/B test.

This isn’t a sophisticated survey-style test like we used as our Conjoint Analysis example. It’s just taking our website or store traffic and sampling a portion of it with an alternative.

Your A is your control. That’s the status quo or the current option or configuration. B is what you think will perform better. 

These tests aren’t limited to just the price itself, but everything around the price. The context, the journey that led the buyer here, the UX, and the brand promise.

If you need ideas of what to test, go back to the previous chapter and look at Nick Kolenda’s pricing psychology article. 

Be careful. Step back from time to time so you don’t test yourself into a website that looks like a giant Belcher Button.

The results of well-run A/B tests are some of the most convincing data you can have for getting buy-in to website, channel, or pricing changes. 

Dynamic Pricing

In Chapter 7 we looked at how we might influence WTP. 

But what if WTP changes on its own as a function of time? There’s a version of pricing optimization where prices change rapidly. Pricing that automatically changes to match perceptions of fluctuating WTP is called dynamic pricing

One example of this is the price of stocks. Minute by minute the price changes to reflect the equilibrium of supply and demand for the publicly traded shares of any particular stock. 

Another example is ridesharing. Uber and Lyft have surge rates influenced by the time of day, holidays, and special events. During the surge, riders pay more because demand is growing against supply.

Airlines have annoyed many customers who pause their booking for some reason, only to return and find the price has changed. 

I once flew from Salt Lake City, Utah to Dublin, Ireland for $320 return. I booked  in advance as part of a promotion and I felt like I got a pretty good deal. In New York a passenger joined the flight (that used to be a thing) and sat next to me. He boasted that he got his ticket at the last minute for only $1,800. 

Even though he paid nearly six times as much as I had, we were both happy with what we paid and the airline was capturing that variable WTP by pricing dynamically based on the circumstances when we purchased our tickets. 

Dynamic pricing requires technology and an algorithm to present prices based on timing and other factors. As with other types of personalization, you need to be careful because done wrong this delivers a horrible experience. But do it right and you are capturing the revenue available due to fluctuating WTP.

Additional Resources:

How To Use Split Testing To Find The Perfect Product Pricing by Anna Gotter of ShopifyZenrez: A Case Study in Dynamic Pricing by Amy Bond

New Client Offer


Leave your details below and receive a discount coupon in your inbox