Not Just a Tool, A Mission: Shape Your Pricing Edge with Iterative Strategies from Amazon, Uber, Airbnb

In the fiercely competitive airline industry, mastering pricing through iterative improvements is crucial for staying ahead. Giants like Amazon and Uber showcase the power of dynamic pricing strategies: they focus on understanding customer reactions, sophisticated forecasting, and global optimization in pricing. But no matter how smart the techniques are, there's no way around actually testing the real-world outcomes. It's an approach that requires constant evaluation to ensure each strategic move aligns with goals and secures a long-term competitive advantage in a market where pricing strategies can make or break success.

Table of Contents

How to Stay Ahead in a Highly Competitive Field

Nothing is static in a competition. All players are eager to get ahead. They train hard, refine their tactics, and adapt to their opponent’s strategies. Corporate competition isn’t any different; companies are constantly looking for small improvements to get ahead—especially in the airline industry with highly price-sensitive customers.

We want to think there’s a big move to win: implement that new technology, and we’ll be ahead. However, the reality looks different. Take Formula 1, for example. Having the best car at the beginning of the season only gets you so far. Without continuous improvements, the other teams are quick to catch up.

Leading tech companies like Amazon and Uber have understood that. They have large pricing teams continuously working on their pricing strategies for +10 years and are still improving. It’s only with daily dedication that you can stay ahead.

They’ve mastered three essential components:

  • Elasticities: Understanding how your customers react to prices is key
  • Forecasting & Optimization: You don’t just need the right prices for tomorrow but need the best prices across the load-factor curve, alternative routes, and ancillary options
  • Evaluation: For iterative progress to work, it’s essential to constantly evaluate whether you’re moving in the right direction 


For any pricing strategy, we must answer the age-old question: How will my customers react to my pricing changes? Of course, any pricing solution has implicit assumptions about how customers will behave. However, few show these curves explicitly, and even fewer rigorously test whether they’re correct.

Having explicit elasticities has two huge advantages: 1) they can be verified by intuition and experiments, and 2) they can be used for broader strategic considerations. 1) is especially interesting: instead of having to judge the outcomes, aka ticket prices, we can now understand why these prices are suggested. That allows us to judge whether they make sense or seem unreasonable. It’s always good to start with a gut-check: Do these make sense from our market understanding? Have we seen any evidence of these predicted reactions in the past? From there, we can go deeper and run fast experiments to ensure our models predict real-world outcomes.

With elasticities, we almost always think about customers. But the same holds for the competition: How do they react to our pricing changes? Combining explicit elasticity models with experimentation allows us to understand the two essential factors in our pricing decisions: customers and competitors.

Explicit and validated elasticities for customers and competitors elevate our understanding of the market, inform large-scale strategic decisions, better inform analysts of daily choices, and are the core ingredient for any algorithm-driven pricing solution like dynamic pricing.

Forecasting & Optimization

Elasticities tell us how customers and competitors will react to our pricing changes. However, there are still endless options and combinations of options. Consider, for example, the state of the load factor/anticipation: is it optimal to sell more tickets early or later in the process? What’s the trade-off of not filling the flight compared to increasing the prices? And, of course, there are many more.

To make these decisions, we need a strong forecasting engine that can evaluate the outcomes of tens of thousands of configurations. From there, we need to find the optimal configuration over the lifetime of a flight and its alternatives. If we only optimized for today’s or tomorrow’s revenue, we might increase revenue today but sacrifice an even larger increase that would have been realized after that.


We live in a rapidly evolving world—especially in competitive spaces like air travel. There’s no perfect model of everything, particularly if we aim to move fast and make progress in the right direction. Circling back to our Formula 1 metaphor, even designs tested in endless wind tunnel and computer simulations fail to perform in reality. That’s why we need to constantly evaluate via real-world experiments whether our ideas and insights actually translate into the desired outcomes.

When competing with the world’s most sophisticated airlines, there’s no single move to stay ahead. We must continuously invest and improve our pricing strategies and tactics daily. This involves deeply understanding our customers’ and competitors’ reactions to our actions, taking a broad perspective on optimization that prioritizes long-term objectives over short-term gains, and continuously evaluating our progress.

Picture of Dr. Christian Nauerz
Dr. Christian Nauerz
At acmetric, I work with C-level executives and startup founders to identify their largest opportunities for data/ML/AI. Over the last years, I've developed a battle-tested approach to generating value. I support the whole process from quick validation via an MVP to global rollouts. I have a deep passion for translating business challenges into scientific problems and quickly testing their impact. There's nothing more satisfying than solving an abstract problem with data and quickly seeing the impact on the real world