One of the most common questions we hear at Digonex is “how does dynamic pricing work?” Even for people familiar with the basic concept—that is, regular analysis of data leading to frequent price updates—the mechanics of how price recommendations are generated remain mysterious.

The actual answer to the question “how does dynamic pricing work” involves mathematical and statistical methods beyond what the average person has experience with, which is why we employ Ph.D. economists. To explain in detail how dynamic pricing works, we’d first have to cover things like differential equations! But if you’re asking the question “how does dynamic pricing work,” you probably aren’t looking for post-graduate math lessons; you just want some level of assurance, before investing in a dynamic pricing program, that what you’re buying into is sound.

It’s hard to do your due diligence, though, on something that is both highly proprietary AND involves Ph.D.-level math. Therefore, a better question than “how does dynamic pricing work” is “WHY does dynamic pricing work?” The latter question is a reasonable proxy for the former, and also easier to explain and understand.

**WHY dynamic pricing works**

To understand WHY dynamic pricing works, it’s helpful to contrast it with other methods of pricing:

**Fixed Pricing**

Imagine running an attraction where the demand varies widely from day to day over the course of a year. Now imagine that this attraction has single, standard price that applies year-round. When we graph “demand” vs. price, the inefficiency of fixed pricing is apparent. (For illustration purposes, I’m using sales as a proxy for “demand,” which is a gross oversimplification. As my colleague Justin points out, proper analysis of demand is much more complex than this, which is why dynamic pricing is best left to professional economists). There are some days when admission is likely worth more to the market—but the cost of a ticket on these days is the same as any other day.

There are also lower-demand days where the value of admission maybe isn’t quite as high. Perhaps your attraction can boost sales on these days by offering various discount programs (“$5 off Tuesdays!”), but using that approach alone does nothing to capture the potential upside of the highest-demand days.

**Variable pricing**

Once you recognize the gap between what you’re charging, and what you could be charging based on market demand, the easiest response is to institute variable pricing. In variable pricing, the price of admission is still fixed for any one particular day, but different groups of days have different prices.

In variable pricing, it’s common for there to be 2-5 different tiers of pricing, and each day in the season gets assigned to a tier. Prices for each tier are fixed and don’t change throughout the season. Here is illustrated a potential two-tiered (“normal/off-peak”) variable pricing plan for your attraction:

Here is what a three-tiered “peak/normal/off-peak” variable pricing setup might look like (this is the type of pricing, incidentally, that Disney Parks recently instituted):

The more tiers there are, the more flexibility you have to try and match price to demand. Therefore, the most flexible approach to variable pricing would be to make each day its own price tier—for a total of 365 tiers!—to account for the fact that demand isn’t constant over a year, or even over a portion of a season, but instead varies from day to day.

However, it’s challenging enough to project at the beginning of your season which calendar days should fall into which of 2-5 price tiers, and then to come up with pricing for each tier. You *definitely* don’t have time to figure out the best price separately for each of 365 days. (Imagine doing this just once, let alone revisiting prices throughout the season!) That’s where the computers and higher math come in.

**Dynamic Pricing**

The two main limitations of variable pricing are:

- The practical limit on how many separate pricing tiers a person or team can administer on their own.
- The price for each tier is fixed at the beginning of the season; predictions about the season are locked in, with no flexibility to adjust based on changing conditions.

Dynamic pricing with automation overcomes both of these problems.

With dynamic pricing, data is continually captured and evaluated in order to adjust prices for each day in light of the most up-to-date forecast about demand for that day. (Again, I’m oversimplifying, because changing price for one day can also affect demand on other days, so dynamic pricing has to take that into account).

Fixed, variable, and dynamic pricing all require analysis of the available data to make informed pricing decisions. So why does dynamic pricing work the best of any of these methods?

- Dynamic pricing algorithms can process more data than individuals or teams.
- Dynamic pricing algorithms apply advanced statistical techniques and econometric theory that that aren’t accessible to the layperson.
- Dynamic pricing algorithms never stop fine-tuning prices, but people have other jobs to do.

(Of course, what dynamic pricing algorithms don’t have is access to your experience and intuition concerning your business and your customers; that knowledge can’t easily be quantified. That’s why it’s important that humans at least have the opportunity to review the pricing algorithm’s recommendations before they are implemented).

**Conclusion**

It’s not necessary to know the detailed answer to “how does dynamic pricing work” if you understand the more straightforward answer to “why does dynamic pricing work.” It works because it takes the same approach you would to pricing—only it does so more often, much faster, and using tools you don’t currently have available.