Dynamic pricing ain’t all that
Through a special arrangement, what follows is a summary of an article from Retail Paradox, RSR Research’s weekly analysis on emerging issues facing retailers, presented here for discussion.
Dynamic pricing seems to be a popular buzzword these days.
My take: at least half the hype comes from a misplaced definition of what dynamic pricing actually is. The problem comes from confusing dynamic pricing with “personalized” pricing.
What Amazon does — change prices multiple times during the day (although the number of items receiving this treatment is still pretty small) — is “dynamic” pricing, but it is in no way personalized. It’s just a lot of price changes happening during a short time period – and someone who might have paid the higher price benefits just as much as someone only willing to pay less.
At best, dynamic pricing may evolve to sophisticated approaches that detect whether a competitor has made a significant price change or is facing an understock, and also whether your own conversion rate on the product in question is taking a related hit.
Personalized pricing is different — and requires more complex analytics to pull off. For example, at Jet.com, depending on what you add to your cart, each item gets repriced right in front of you. There is “the” price, and there is “your” price. Repricing might depend on the deals Jet.com has with suppliers, the supply chain efficiencies of the items you’ve selected, or your shopping behavior, whether that is localized behavior, like adding something to your cart and then not buying it; or behavior over time, like offering discounts to high value shoppers to expand the categories they shop.
As long as those offers can be rationalized to avoid accusations of bias, variable prices can be explained by showing how “your” price was calculated versus “the” price.
I think personalized pricing is the future, and that requires a lot more capability than simply changing prices many times over the course of the day. Dynamic pricing, the definition in use today, is just a fad, and will ultimately be out-gamed, whether by competitors, or by consumers themselves.
Discussion Questions
DISCUSSION QUESTIONS:Â Is dynamic pricing here to stay? Do you think personalized pricing will replace it?
Poll
BrainTrust
Mark Heckman
Principal, Mark Heckman Consulting
Ross Ely
President and CEO, ProLogic Retail Services
What ever happened to the price being the price? With dynamic pricing someone is always going to feel as if s/he’s been had. With personal pricing, how does a consumer know that s/he is not being taking advantage of? Unfortunately, both dynamic and personal pricing seem to be here to stay, and there is little consumers can do to insure that they are getting the best price and value.
Perhaps the term “dynamic pricing” should be used to describe variable or changing prices by item. There appears to be two models for dynamic pricing.
Amazon’s approach is a “shotgun” model which changes an item’s price and blasts it out on the site to everyone. It is certainly dynamic in that the price can be changed many times a day, even hourly.
Jet’s model is entirely different. It is more of a rifle shot, targeted approach based upon the customer characteristics and/or the relationship between items in the shopper’s cart.
Nikki Baird is right on … the personalized model of dynamic pricing is the wave of the future for two key reasons. The markdown on the items is specific and does not create across-the-board gross margin declines on an item. More importantly, personalized pricing in the cart focuses on the “market basket” approach of generating more overall gross margin by incenting the customer to attach more things to the initial purchase.
Nikki is also right on about the value and complexity of personalized pricing. It takes much more sophisticated systems and algorithms to implement.
But ultimately personalized pricing could be a huge differentiator! Do you want “your price” … or what Amazon decides to charge that hour?
I think both types of pricing, along with others, are proving to be immensely valuable to those retailers that are truly leveraging their capabilities. Dynamic pricing, if you sort through all the hype, at its core allows retailers to vary pricing strategies and tactics based upon the analysis of multiple external sources of data, both structured and unstructured, including social chatter, news items, weather, sporting events and even local events at a micro level. As with personalized pricing, I believe dynamic pricing requires complex analytics to execute properly. As you can imagine, taking all of these sources of data and analyzing how these data interact with each other requires deep machine learning functionality, which goes way beyond the traditional “gut feel” of dynamic pricing.
Personalized pricing should use many of these sources of data, but also utilize internal data, like call center, customer reviews, etc.
For the sake of clarity, it’s good to see someone distinguishing between these two distinct but often concomitant pricing capabilities.
That said, to call dynamic pricing a “fad” may be overstating the case. Will it become a widely adopted, table-stakes capability? Some might argue that it already has in some industry segments.
Will it go away? I doubt it.
As the article notes, many retailers are already re-pricing at high frequency based on contextual data about demand elasticities, price image, profitability, remaining stock levels, competitive positioning and more. This is integral enough to their pricing strategies (and has been for so long) that I struggle to imagine it being retired, though industry buzz will certainly fade.
Will more retailers offer shoppers a higher level of self-determination of price or value through explicit choices like waiving return privileges or loyalty? Probably so.
In that sense, personalized pricing is the future. But it’s a case of “and,” not “or.”
Airlines have been employing dynamic pricing for several decades. If you don’t believe it, price a ticket that departs tomorrow and compare it to the price of the same destination a month from now.
I do believe that as technology improves, especially at the shelf with digital signage, etc., dynamic pricing will continue to evolve and become more widespread and in effect allow savvy retailers to fully enjoy the fruits of their existing price optimization software.
Personalized pricing, which Ms. Baird appropriately points out, is much different. It implies a one-to-one relationship with the shopper, which requires resources, strategy and innovation that simply does not exist in all practicality today. I believe the jury is still out on whether personalized pricing will ever reach critical mass practice, while I believe dynamic pricing’s best days still lie ahead.
Dynamic pricing will be here until, and if, the consuming public starts voicing its discontent with it. We have dynamic pricing in the travel and hotel industries today, and I think it does a lot to frustrate the consumer because there is no predictability and planning a purchase becomes an anxiety-filled endeavor (unless you don’t mind paying top dollar). Retailers don’t want a purchase to become an anxiety-filled endeavor.
Personalized pricing, by contrast, has the implication of always being lower than “the” price. That is an instance of surprise and delight, and every customer enjoys that. “Your” price is always pegged against a measurable “the” price. Does anyone know if there is a “standard” price for an airline ticket in coach, from JFK to LAX? If my assessment is correct, then personalized pricing may well replace dynamic pricing.
Dynamic pricing delivers value to both retailers and consumers and is not a fad as the author suggests. Retailers that are able to deliver spot discounts to consumers will benefit as they pass along supply chain efficiencies and other cost reductions.
A frequently cited example is the soda vending machine which lowers its prices on cold, cloudy days. The operational challenge for the seller is how to dynamically calculate and communicate the dynamic pricing to the consumer.
Personalized pricing is a complementary and more advanced level of dynamic pricing that takes individual shopping behavior and demographics into account. Personalized pricing is already a key element of most shopper rewards programs, and can exist side-by-side with dynamic pricing as retailers optimize pricing based on both shopper information and cost efficiencies.
I’m sorry but this is semantics as dynamic pricing has been around since I was the POS project manager for Stop & Shop Supermarkets from 1979 to 1981. It is and has been here to stay.
Personalized pricing is the future!
It is real-time pricing based upon the who, what, when and where at the customer level and is the future of pricing.
Dynamic pricing has been available at the following levels since 1979:
It can be done on any frequency by item, department, class, etc.
A fad … maybe online but it has been in the store forever.
Doesn’t anyone remember the Blue Light Specials at Kmart? An original version of dynamic pricing!
How does the person who bought an item at one price and then just three hours later sees it at a lower price feel? (Rhetorical question!)
Many places offer a low-price guarantee. If the price is lower in the next 30 days, they will match it. A “machine” like Amazon could automate this. Let the customer know that if they drop the price they will credit the customer’s account. That gives a sense of security for the customer. It could also increase the likelihood of a sale.
A relevant fact needs to be included in this discussion. Amazon does not control pricing on about 40 percent of items it sells and ships. These prices are controlled by third-party sellers who use a myriad of pricing tools. Third-party is much more profit-centric than Amazon. The trend is algorithmic pricing that is tied to velocity and margins.
I’m on a few number one BSRs at 45.87, and when I sell my goal, my software will take me to another price point (up or down) based upon rules I’ve programmed. My algorithm’s main component is competitor logic, however it also includes meaningful margin and velocity logic … it’s not AI, but it is smart.
I agree with Nikki, there is a big difference between dynamic pricing and personalized pricing. I would however offer that dynamic pricing is far more than a pedestrian rules-based approach to pricing. Specifically, Amazon changes prices millions of times a day (exactly where/when the consumer will respond). Their pricing engine is looking at a myriad of elements. Included are concepts of price elasticity (as well as day-part elasticity), competitive price influence on demand, availability, predictive shipping that contemplates price, cannibalization and halo impacts of items in the category and across categories. This approach to pricing has created a competitive environment that most retailers cannot match and presents significant heartburn for multi-channel retailers who make a claim of pricing the same across channels.
Most retailers are not prepared for personalized pricing and need to move from static pricing practices to dynamic pricing practices before they could possibly consider personalized pricing. The road to personalized pricing is paved with real-time visibility to inventory, supply chain, demand, competitive position, financial plans, customer value, customer retention, loyalty and the opportunities to improve these conditions through the lever of price.
The gaming industry as well as the travel industry have shown a path from rules to dynamic pricing to personalized pricing that can build strong loyalty and improve enterprise financials. Consider for a moment just what Disney is doing with their magic bands. They understand every movement and purchase of every guest, individually and by household. This type of information opens the door to a very personal dialog with the guest. Personalized offers can be crafted that elevate the experience and value of the guest to Disney.
I can see retailers working to the same end. There is a path and that path demonstrates tremendous value. However, skipping stones on the path will not lead to success. Companies who attempt to leap to personalized pricing will learn that they need to have a well-developed discipline of dynamic (or as others might say “optimized”) pricing combined with real-time visibility to the enterprise as well as the consumer before they begin the next level of pricing maturity.
Finally, personalized pricing is not the end of the path. I believe there will be further advances beyond personalized pricing that will take loyalty and enterprise financials further than the current promise of personalized pricing. Therein rests the bigger issue. Retailers must invest in technologies that can grow with their needs. Current legacy on-premise providers and software can not do this. My advice; identify technology providers that understand this is a journey and offer configurable software that can grow with your organization’s maturity and the demands of the consumer.
In 1999, “Coca-Cola abandoned the [dynamic pricing] experiment because many customers felt they were being taken advantage of,” according to a blog.
Personalized pricing does make sense as a replacement. But in order to do so a retailer will need to have high degree of trust in their customer data and shopping history profiles.
Another question for the group: Will personalized pricing information be delivered via mobile or wearable only? Or will the promise of those dynamic e-ink shelf tags finally get called off the bench?
I agree that personalized pricing has a bigger and broader future. Mostly because it will sit well with the consumer. Consumers hate feeling like they are being ripped off or that someone else might be getting a better price for no apparent reason. The more the retailer can explain as to the pricing changes the more the consumer will accept it and join in.
On the other hand, if a retailer has a hot product on sale for a specified time and chooses to up the price as the end of the sale approaches, I think they will take advantage of that too.
and that’s my 2 cents.
” … [personalized pricing] requires a lot more capability … “
Indeed! Mostly through increased data collection, and the real questions are whether or not the forces that might defeat this — privacy concerns, increased “opting out” — will be stronger, and whether or not consumers will accept the idea (just because something can be “rationalized” doesn’t mean it will be accepted). I think it will become more common — less rare might be a better term — but never to the extent that it could be in theory. As far as personalized prices in physical stores, no I don’t see that happening (beyond, of course, the obvious fact that we already have it in the form of “club cards” and other “membership” benefits). It would be a legal quagmire over issues of transparency.
Dynamic pricing may be good for business-to-business and durable consumer products. However, consumers compare prices too much on non-durables, creating potential backlash issues, both from people who got the lower price and from people who got the higher price. Both groups will question the fairness.
To me, the distinction is Dynamic Pricing is business-centric yield management pricing that maximizes revenue for brands based on a set of business rules to optimize cost of goods sold at the time of sale. Personalized pricing, on the other hand, is customer-centric that provides the optimal pricing to motivate a consumer to purchase based on the value they are putting on the product at the moment they desire to buy. Both are valuable tools that brands have to maximize profit based on supply and demand and individual motivations that drive consumer to buy during a shopping experience. Both can and should be used, it shouldn’t be one replacing the other.
The key to both is real-time data and customer analytics. Sophisticated programs and databases can figure out the best pricing to present during the shopping experience, but knowing the what, when and who of data to set the optimal price is the challenge. Most retailers can’t get this actionable data out of their system, let alone use it to impact real-time pricing.
This isn’t the future of pricing as it has been around for years, but it will become more commonplace going forward, and both are here to stay.
While dynamic pricing and zone based online pricing strategies are likely here to stay, personalized pricing strategies should be carefully examined and balanced within the context of the right thing to do. Can retailers begin to understand and apply pricing strategies down to the consumer level, sure, but the genuine question is “should they”? Retailers and brands need to examine their relationship with their target shopper and evaluate if personalized pricing will further build or erode that relationship over the long term. While dynamic pricing relies on the principle of supply and demand to justify, personalized pricing is overtly favoring one shopper over another.
I think dynamic pricing and personalized pricing can work together to boost conversions. Think about it — differential pricing has always existed in retail. Decades ago (and definitely still today), if you had a relationship with a store clerk they were likely to cut you a deal. They knew that you were a loyal customer who supported the store often, so they gave a little discount or threw in something extra to say thank you and keep you coming back.
Personalized pricing online is the next level of this and it’s a bit more tricky. You aren’t in one store having a conversation with one clerk. You probably have several tabs open with the same item at different retailers to compare prices, or are using a CSE. I’d argue that dynamic pricing has become a necessity in online retail. If your price is uncompetitive, shoppers are going to ignore you and click over to a competitor (unless given a really good reason to pay a premium). It’s easier to have static prices in-store when shoppers are willing to pay more to be instantly gratified. Maybe that’s why dynamic price changes are much more common online than in-store.
I don’t think it’s a matter of personalized pricing taking over dynamic pricing, I think they should work together, especially online. Personalized pricing is like being friends with the store clerk, except online it might come in the form of a “we miss you” email with a discount code or free gift offer with your next purchase. Overall, price changes are necessary to move retail forward, whether they are personalized or available to all shoppers. It’s worth testing different prices to find the sweet spot.
The economics of pricing a product/service with limited quantity, a shelf-life, and a low marginal cost to the seller (like an airline seat) are very different than for traditional merchandise sitting on a shelf in a store or a warehouse.
The goal of dynamic pricing is to price discriminate … to allow those who are willing to pay more to do so. For retailers in competition with each other however, the price they post on the “tag” is capped by their desire to avoid looking expensive while their willingness to accept less “dynamically” is limited by the hard cost of supplying that thing. This is often a very narrow window.
As Ms. Baird points out, the ability to dynamically price discriminate hinges on the availability of real-time information indicating each customer’s willingness to pay.
To succeed in practice, retailers must not only create this real-time pricing engine, but implement it in a way that avoids negative customer reaction. I’d argue that getting the psychology right is more important and challenging than getting the technology right.
As Nikki notes, dynamic pricing can have a lot of different meanings to different individuals. If dynamic pricing is mainly about responding quickly to competitor price changes then retailers have already been employing dynamic pricing for years. If it means changing prices on a very frequent basis then that has been done by the e-commerce retailers for a number of years. If it means using pricing to shape demand then this type of dynamic pricing has also been in use for a number of years by leveraging price elasticity and what-if or optimization engines. If it means personalized pricing then that is something relatively new and still has some runway.
When it comes to execution of dynamic pricing, each of the above definitions requires a different set of capabilities. With competitor price matching all that is needed is a way to collect and incorporate competitor prices and then some simple rules for how the retailer should respond.
With frequent price changes there is usually a requirement to have some automated way of making the frequent changes depending on which factor is driving the change and what metrics are being used (e.g. inventory levels or sales velocity are often used).
If the retailer is attempting to use dynamic pricing to shape demand then they could also rely on simple metrics like inventory levels and sales velocities but they will typically also need to leverage elasticity data so they know the magnitude of the volume impact from a range of potential price changes.
With personalized pricing the retailer needs a way to identify the individual shopper, a way to analyze the individual shopper data, a way to quantify how the individual shopper would respond to price changes (i.e., price elasticity at the customer level) and then a way to determine the personalized prices for each individual shopper (i.e., a price determination engine using some combination of rules and optimization that can set the personalized prices). These personalized prices then need to be available for use with individual customers either through traditional paper vehicles or new digital or mobile vehicles.
Dynamic pricing and personalized pricing in whatever definitions are used are here to stay — all that remains to be seen is how retailers embrace them and how shoppers respond to them.
Dynamic pricing has always existed and isn’t going anywhere. Consumers experience it everywhere from plane tickets and sporting event tickets to taxi services and ski trips. Weather and snow conditions can change consumers’ perceived value of a day on the slopes, just like it matters to sports fans which opponent their team will face in a game they’re looking at buying tickets for.
What’s different now is all the newly available data associated with the digital “footprints” that consumers leave behind as they interact with retailers. Retailers analyzing this data to anticipate consumer behavior and use price as a lever to increase margins is inevitable, but it will take some time to get right.