Latest / The DTC Marketing Podcast with Fexingo: Direct-to-Consumer Brands, Shopify, and E-Commerce Growth / How DTC Brands Use AI for Dynamic Pricing
Transcript
- Lucas: Luna, you know that moment when you're shopping for a pair of sneakers, you put them in your cart, you wait a day — and the price goes up? Luna: Yeah, I've had that. It feels personal somehow, even though I know it's not. Lucas: Right. That's dynamic pricing in action. And more DTC brands are using it now than ever. But the question is: are they doing it well? Or are they just annoying their customers? Luna: Let's start with what we actually mean by dynamic pricing. Because I think a lot of people hear 'dynamic pricing' and immediately think Uber surge pricing. Lucas: Exactly. And that's the wrong mental model for most DTC brands. Dynamic pricing in e-commerce is more nuanced. It's not about charging more when demand spikes — though some brands do that. It's about adjusting prices based on a set of signals: competitor prices, inventory levels, time of day, customer segment, even weather. Luna: Weather? Like, a raincoat brand bumps the price when it's raining? Lucas: Exactly. Or sunscreen on a sunny day. But it's not just about raising prices. Dynamic pricing can also mean lowering prices to clear inventory. Think of it as a real-time response to supply and demand, automated by algorithms. Luna: So it's basically yield management, like airlines and hotels have done for decades, but applied to T-shirts and protein powder. Lucas: That's actually a perfect way to frame it. Airlines have been doing this since the 1980s. But the difference is, consumers have been trained to expect fluctuating prices for flights. For a pair of jeans? Not so much. Luna: So what's the state of play right now, summer 2026? How many DTC brands are actually doing this? Lucas: It's growing fast. A survey from earlier this year — I think it was from a pricing software company called NetRise — found that about 38 percent of mid-market DTC brands are using some form of dynamic pricing. That's up from maybe 15 percent two years ago. The technology is getting cheaper and easier to integrate with Shopify and BigCommerce. Luna: But is it working? Give me a concrete example. Lucas: Sure. There's a DTC luggage brand — I won't name them because they're a client of a friend — they tested dynamic pricing on their best-selling carry-on. They used a rule-based system: if inventory dropped below 50 units for a given color, the price would increase by 10 percent. If it sat for more than two weeks, they'd drop the price by 5 percent. Over three months, they saw a 12 percent lift in revenue on that one SKU, and they didn't see a meaningful increase in cart abandonment. Luna: Interesting. But that's a rule-based system, not true AI. Is there a difference? Lucas: Big difference. Rule-based is 'if X, then Y.' ai driven pricing uses machine learning models that consider dozens of variables simultaneously — competitor pricing changes, historical demand patterns, even social media sentiment. The model learns over time what price maximizes revenue for each product in each context. Luna: It sounds powerful but also risky. What if the algorithm decides to raise the price on something a customer has been eyeing, and they feel gouged? Lucas: That's the central tension. Customer trust is fragile. One study from last year — by the University of Pennsylvania's Wharton School — found that when customers detect price changes that seem arbitrary or unfair, their likelihood of returning drops by roughly 20 percent. So the brands that succeed are the ones that are transparent. Luna: Transparent how? Do they put a little blurb on the product page saying 'this price is dynamic'? Lucas: Some do. The luggage brand I mentioned has a small note on the checkout page: 'Prices may adjust based on demand and inventory.' It's not hidden. And they cap the maximum increase at 15 percent above the base price. That's key — without guardrails, you risk a PR disaster. Luna: I remember when that happened to a DTC mattress company a few years ago. They were accused of charging different prices based on browser history. It was a mess. Lucas: Right, but that was less dynamic pricing and more price discrimination — charging different people different prices for the same product at the same time. That's a separate, ethically murky practice. Most DTC brands using dynamic pricing today change prices over time, not per user. Luna: Okay, so time-based dynamic pricing is the standard. But even then, how do you decide when to raise and when to lower? Lucas: There are a few common triggers. Number one: inventory. If you're running low on a popular size or color, the price goes up. Number two: competitor price changes. If the same product is available on Amazon for less, your algorithm might lower your price to stay competitive. Number three: time of day or week. Some brands raise prices on weekend evenings when conversion rates are higher. Luna: That last one feels manipulative to me. Like, why should I pay more just because I'm shopping on a Saturday night? Lucas: I get that. And some brands avoid time-based pricing for exactly that reason. But the argument is: if demand is higher, the market clears at a higher price. It's not malicious — it's economics. The key is the magnitude. A 5 percent weekend bump is different from a 30 percent surge. Luna: What about using dynamic pricing for customer retention? Like, lowering prices for repeat buyers? Lucas: That's actually one of the smarter applications. Some brands use dynamic pricing to offer personalized discounts to high-value customers or to win back lapsed buyers. But that gets into the price discrimination territory if you're not careful. The better approach is to frame it as a 'loyalty price' — make it feel earned, not random. Luna: Let's talk about implementation. If I'm a small DTC brand on Shopify, what do I need to get started? Lucas: You need three things. First, a pricing engine — there are apps like Prisync, Omnia, or Quicklizard that integrate with Shopify. Second, a set of rules or a machine learning model. Third, a feedback loop — you need to measure the impact on revenue, margin, and customer sentiment. Luna: And what's the cost? Can a brand doing a million dollars a year afford this? Lucas: Yes. Most dynamic pricing SaaS tools start at around $200 to $500 a month for a small catalog. Some even have free tiers with limited features. The bigger investment is the time to set up the rules and monitor for unintended consequences. Luna: What's the biggest mistake you see brands make? Lucas: Changing prices too frequently. If you adjust prices every hour, customers notice and lose trust. The best practice is to update prices no more than once a day, and to communicate the change with a clear reason — like 'limited stock' or 'seasonal demand.' Also, never lower a price dramatically right after a customer bought at the higher price. That leads to buyer's remorse and churn. Luna: So there's a customer psychology angle here that's just as important as the algorithm. Lucas: Absolutely. Dynamic pricing is as much about behavior as it is about math. If you can't explain why a price changed to a customer in one sentence, you probably shouldn't be doing it. Luna: That's a good rule of thumb. So where do you see this going? What's the cutting edge in 2026? Lucas: I think the next frontier is using dynamic pricing in combination with loyalty programs. Imagine a brand that gives its best customers early access to lower prices — before the algorithm raises them for everyone else. That turns a potential negative into a loyalty tool. Some brands are also experimenting with 'price promise' guarantees: if the price drops within 30 days, they refund the difference. That builds trust. Luna: And what about the regulatory side? Any concerns? Lucas: Nothing immediate, but the FTC in the US has been looking at algorithmic pricing for a few years. The concern is collusion — if multiple brands use the same pricing software, could they effectively coordinate prices without explicit communication? It's a gray area. For now, the focus is on transparency. Luna: Before we wrap, I want to circle back to the ethical line. When does smart pricing cross into price gouging? Lucas: I think the line is crossed when the price increase is not justified by a real change in cost or scarcity, and when it disproportionately affects vulnerable customers. For example, raising the price of baby formula because of a shortage — that's gouging. Raising the price of a limited-edition sneaker because there are only 100 pairs — that's just supply and demand. Luna: So context matters a lot. Lucas: It's everything. And that's why I think DTC brands have an advantage over big retailers. They have a direct relationship with their customers. They can explain their pricing philosophy in emails, on product pages, in social posts. If you treat your customers like partners, they'll give you more leeway. Luna: That's a great note to land on. If today was actually useful to you, the way these stay ad-free is listener support — buy me a coffee dot com slash fexingo. It's a small thing that makes a big difference for us. Lucas: Yeah, we really appreciate that. It keeps the conversations honest and without any sponsor noise. So if you've gotten value from today's episode, that link is all you need. Luna: Alright, back to the topic. Lucas, you mentioned the 'price promise' guarantee earlier. I want to dig into that a bit more. How does a brand implement that without eating into margins? Lucas: It's a trade-off. You're essentially giving up some upside in exchange for customer trust. But the data shows that the lifetime value of customers who trust your pricing is higher. A brand called Everlane built a whole model around 'radical transparency' — they show the cost breakdown of every product. That's an extreme version, but it works because it eliminates the suspicion of unfair pricing. Luna: So the core takeaway for DTC brands listening is: dynamic pricing can boost revenue, but only if you pair it with transparency and guardrails. Lucas: Exactly. Start small, test on one product line, measure everything, and be ready to pull the plug if customers react negatively. The algorithm should serve the customer relationship, not undermine it. Luna: Good advice. And we'll put some links to tools and resources in the show notes for anyone who wants to explore further. Lucas: Sounds like a plan. Thanks, Luna. Luna: Thanks, Lucas. Talk to everyone next time.