Safe testing of pricing strategies

Price testing is definitely one of the most difficult projects you can undertake. Those who have tried partisan ways on the principle of “I’ll change the price and see what will happen” quickly found out that it is 100% effective but only as far as “seeing what will happen” goes. Increasing the revenues or building a better brand position, it’s a whole different story.

So how can you test new prices safely and effectively?

Step 1. Preparation → market research.

It is best to divide the preparation stage into two parts. First, in the focus group, try to initially set customer price thresholds. You can do this by asking questions like:

  • In your opinion, how much is this worth?
  • Does such a price seem justified?
  • What is the highest price you would be willing to pay?
  • Should this product be cheaper or more expensive than that product?
  • E.t.c.

If you want me to write in detail about principles of conducting focus tests in the process of determining prices, let me know in the comments.

Once you have dealt with focus groups, go to the mass testing during which, using questionnaires, surveys, and interviews (e.g. by phone) you will determine the price level acceptable by your clients. Comparing these findings with historical data (how customers used to react to changes in prices of products in this category) should give you quite precise information on what range of modification you can afford.

– It seems simple, what could possibly go wrong here? – you might ask.

Customer declarations can differ from their actual choices. A spoken declaration of “I would buy it”, and taking out cash from your wallet to buy are two different things. When Phillips wanted to bring new musical equipment to market, with fantastic colors and designs, the customers from the control groups were delighted. As a reward for participation, they could choose one such device for themselves. They all chose the black ones.

How to protect against this? Theoretically, there are many solutions, but basically, everything boils down to providing a sufficiently large number of respondents. The old rule says that in a sufficiently large test group, all abnormalities will be evened out and in this case, it is no different. Not all interviews will have a full set of answers. For others, you will have to remove some yourself, because, for example, the respondent will mark A top to bottom. Finally, you will exclude some respondents to maintain the proportions guaranteeing the representativeness of the group.

In the end, if the product is a mass-market product (not a niche one, for example, only for veterinarians), then you will want to have at least 1000 – 2000 respondents to begin with.

Ok, but now that you’ve done your research you can just change prices and enjoy the growing income … almost. I mean, if you like a little thrill of “will my company/brand survive until tomorrow?” This is a good solution. All you have to do is change the price of all products and the matter is settled. But, on the off chance that you are one of those “boring” people who do not play the business equivalent of Russian Roulette, you may want to know how to run a pilot.

Step 2. Implementation → pilot.

Let’s start with the most basic, but unfortunately still for many unobvious, thing – never offer exactly the same product to the same customers at different prices. Why? Because your clients will know and … they will not be happy. It is not worth it.

The customer must understand why the price is different. Cinema tickets are a great example here. It is clear to everyone that we have to pay a different price for the same film on Monday at noon, and another, higher price, on Friday evening. The product is the same but not completely. Such price differentiation is the basic route to profits.

You can also differentiate prices for the same products in different locations. The price in a 7/11 in one city may be different than in another city – the same customer rarely goes shopping every day in two different cities. It is similar online where no one is surprised that different stores have different prices for the same products.

In general, the idea is to change the prices of only selected items in the entire category (about 10%) and on this basis to test the assumptions and results of price changes. Those can be then translated into the entire product line and/or category.

You can read more about the categorization in my article in the June – July issue of the E-commerce Magazine (#shameless plug), but if you are interested in this topic, please let me know in the commentary.


  • Check price preferences of your clients before making any changes.
  • Test the conclusions using pilot group.
  • Change prices and enjoy higher margins.

Just… Ok, there is one more thing to watch out for – COMPETITION. During testing, you will be particularly vulnerable to its actions. A large promotion or a price war conducted during consumer research or a pilot can affect the attitude of the client and thus the conclusions you draw. Personally, I think that the prices of the competition need to be monitored at all times, but if there is one moment when it is particularly important then this is the price test period.


And you, how do you think? Does price change have to involve risk? Maybe you have a simpler way to do it safely? If so, please share it with me in the commentary.