Will price testing annoy customers? Pricing strategy testing
- 11 February 2026
A well-optimized pricing strategy doesn’t have to be a minefield, yet the decision to change e-commerce prices still feels like a game of Russian roulette for many companies. Silence falls over the conference room, and the same paralyzing question pops into the minds of managers (perhaps including yours): “What if customers get angry?”
It’s a natural reflex. Business psychology, specifically Daniel Kahneman’s prospect theory, teaches us that the pain of losing a customer is twice as intense as the joy of gaining a new one. This fear of loss (Loss Aversion) means that most pricing strategies in Poland are, in fact, “just don’t mess up” strategies.
The result? Prices are set based on historical habits (“it’s always cost this much”), the sales team’s intuition (“the customer won’t pay more”), or, worse yet, reactive imitation of the competition.
Meanwhile, in 2026, price isn’t a fixed label. Price is the most flexible profit lever in your P&L. McKinsey has been saying for years that a 1% price improvement (with unchanged volume) generates an average increase in operating profit of 11%.
No other action – neither cutting costs nor increasing sales volume – has such an impact on the bottom line.
So, if you’re afraid to move your prices, you’re not protecting your business. In fact, you’re slowly bleeding your margins. In this article, I’ll show you how to stop guessing and start testing your prices in a way that’s invisible to your competitors, safe for your image, and deadly effective for your bottom line.
The Myth of the “Angry Customer” and the Declaration Trap
Before we get to the methodology, we need to address a fundamental cognitive bias. Most concerns about “upsetting the customer” are based on the false assumption that the customer knows and remembers the prices of all products.
Sure, a customer will notice a price change for an iPhone or butter at a discount store (KVIs – Key Value Items). But will they notice a 3% change in the price of an HDMI cable, specialized laminate flooring, or rare dog food? Research shows that the average consumer’s price awareness is significantly lower than category managers, who spend eight hours a day poring over price lists, might think.
The second mistake is asking customers for their opinion. If you conduct a survey asking, “Would you buy this product if it were 10 PLN more expensive?” 99% of respondents will answer “No.” Why? Because it’s a logical question, but shopping is emotional.
A classic example is the story of Philips in the 1990s. They wanted to introduce a youth-oriented line of audio equipment. They held focus groups, showing teenagers boomboxes in bright yellow and classic black. In the surveys, the youth raved about the yellow models. “They’re cool, they’re different, we’d buy them!” After the study concluded, participants were allowed to take one of the devices home as a thank you.
Everyone took black.
This is the difference between declared preference and revealed preference. Your job as an e-commerce leader is not to listen to what customers say that they will do. Your task is to create a test environment in which you will check what they really do when it comes to payment.
Safe Experiment Architecture
Price testing isn’t about randomly changing numbers in the store and watching the customer service desk for phone calls. It’s an engineering process that requires data hygiene and strict discipline.
To do this safely, you need to eliminate reputational risk. The biggest fear is when two customers talk to each other and discover that one paid 100 PLN and the other 120 PLN for the same item at the same time. This is a surefire path to a social media crisis.
How to avoid this? By isolating variables. Here are three methods used by top market players (from Amazon to Zalando) to test price elasticity without provoking a revolt.
- Tests based on geographic segmentation
This is the safest form of A/B testing. If you operate in multiple markets or a large country, you can vary your prices based on region. For example, users in the Masovian region see price A, while users in the Greater Poland region see price B.
Why does this work? Because we naturally accept price differences in the physical world (fuel at a station in Warsaw is more expensive than in Radom). Transferring this online, while maintaining consistency within a single IP/region, minimizes the risk of cross-talk (information exchange between customers).
- New vs. Returning Tests
A returning customer typically has a set reference price in mind (“I bought this a month ago for 50 PLN”). Changing their price is risky. However, a new user who lands on your store through Google Ads doesn’t know your pricing history. For them, the price they see is the only reality.
Advanced pricing systems allow you to display a different margin for “cold” traffic (new customers) and a different one for loyal customers (retention). This allows you to analyze the maximum willingness to pay (Willingness to Pay) of new cohorts without disturbing your permanent customer base.
- Assortment Testing (10% Rule)
If you don’t have the technology to dynamically segment traffic, use the “test basket” method. Don’t change prices across your entire product range. Select 5-10% of products from the “long tail”—i.e., products that are less frequently purchased, less sensitive to comparisons, and unique.
Increase their price by 5-10% and watch the conversion for two weeks.
- If sales drop dramatically, the product is price-elastic (customers are price-sensitive). Revert to the old price.
- If sales remain at a similar level, you’ve just found “free money.” Your margin has increased, and customers have accepted the new value.
Experimental Mathematics: When is a Test Successful?
This is where many managers make the mistake of looking only at Revenue or Conversion Rate (CR).
Example: You sell a product for 100 PLN. Your margin is 20 PLN (20%). You raise the price to 110 PLN (10% increase). Your new margin is 30 PLN (a 50% margin increase!). As a result of the price increase, sales (unit volume) drop by 20%. Most managers will say, “Failure! Sales are down 20%! We’re going back to the old price!”
Are you sure? Let’s do the math:
- Scenario A (100 PLN): You sell 100 units. Profit = 100 * 20 PLN = 2,000 PLN.
- Scenario B (110 PLN): You sell 80 units (20% drop). Profit = 80 * 30 PLN = 2,400 PLN.
Despite the decline in sales and conversions, your operating profit increased by 400 PLN (or 20%). Furthermore, you processed fewer orders, so your logistics and customer service costs decreased.
This is exactly what Margin Engineering is. The goal of price testing isn’t to maximize revenue at all costs, but to find the sweet spot where volume and margin are optimized. Without A/B testing, you’ll never find that sweet spot. You’ll be stuck at a local peak, fearing any movement.
Market Context: The Variable You Forget
A/B testing in e-commerce has one drawback that lab testing does not: it takes place in a live, dynamic environment.
Imagine you run a price increase test on product X. After a week, you see that sales in the test group haven’t dropped. Your conclusion: “Great, we can raise prices!” But what didn’t you notice? That same week, your main competitor ran out of stock for this product. Customers were buying from you at a higher price because they had no choice, not because they are price-insensitive. When the competitor returns with the product, your strategy will crumble.
This is why it is impossible to test prices in isolation from competitor data. Every pricing experiment must be timed to the market. You need to know whether, during your test:
- Competition has changed prices.
- Product availability on the market has changed.
- The manufacturer launched an image campaign to increase demand.
Only by superimposing internal data (your test) on external data (competitors’ price monitoring) does a true picture emerge. Without this, conclusions drawn from tests are subject to significant error.
How to implement a testing culture in your company?
Changing the approach from “we set prices quarterly” to “we test prices continuously” is an organizational challenge. It requires breaking down the silos between marketing (which fears a decline in conversions) and finance (which wants margins).
Here’s your action plan for the next week:
- Data Audit: Check if you have the technical capability to differentiate prices (product feed, store engine). If not, this is your number one investment priority.
- Designate a “Safe Sandbox”: Don’t start with bestsellers. Choose a category with medium turnover.
- Establish a Hypothesis: For example, “Increasing the price of brand X by 5% will not result in a decrease in conversions greater than 3%.”
- Start Monitoring: Before you start your test, set up competitor price tracking for these products (tools like Dealavo are essential for a baseline).
- Measure Net Profit, not Revenue: This is a key change in the team’s mentality.
Summary: The Cost of Inaction
Finally, let’s return to psychology. You’re afraid that testing prices carries risks. That customers will leave. That the market will punish you.
But think about the Cost of Inaction. If your competition is already using dynamic pricing algorithms and testing demand elasticity, they’re optimizing their margins right now. In products where you’re too cheap, you lose money customers would gladly pay. In products where you’re too expensive, you lose volume to someone who has calculated the equilibrium point more accurately.
Standing still isn’t safe. You’re just an easy target.
Price testing isn’t gambling. In a world of rising advertising costs (CAC) and logistics, it’s the only way to deliver the results management expects. The question isn’t “should you test?” but “why aren’t you already?”
There is only one conclusion: Technology is only half the battle. Simply having a repricing tool doesn’t guarantee growth, just as having a scalpel doesn’t make one a surgeon. The key is strategy – the ability to formulate hypotheses and interpret test results. Those who can transform raw data into business decisions the fastest win.
That’s why at Dealavo, we not only provide technology but also prioritize in-depth education. We regularly host webinars where we break down real-world market scenarios. We teach how to safely implement automation, avoid price wars, and build an inflation-resistant pricing policy.
If you want to stop guessing and join the group of managers who consciously control the margin – check our schedule and sign up for the next meeting.
