Skip links

Why & How You Should A/B Test Your Pricing

For any business, the way we price our products & services is critical to our success. To create sustainable, valuable businesses, we must generate profits & to generate profits, we must price appropriately.

Despite the centrality of pricing in business success, a recent experience with a prospective client has opened our eyes to the fact that there are many perceived barriers to A/B testing in this area:

The Challenge

In our first face-to-face meeting with a prospective client, they expressed an interest in experimenting with their pricing structure. As a service business with a tiered product offering, increases in average order value (AOV) were impossible to achieve by selling multiple products to the same visitor, and therefore the individual package prices were their principal means of driving additional value from any individual visitor.

As this discussion continued, they mentioned that they had previously been advised to steer clear of anything pricing-related within their CRO programme and that that was a disappointment as they too felt that there was great opportunity in that area.

So this got us to thinking about why they had been advised to avoid testing one of the most fundamental levers of business success & the inevitable question arose around how technically it could be achieved. And having worked on commercial tests like this with a variety of clients over the last few years, we were able to provide a series of options, each with their pros & cons:

The Solutions

Voucher code entry

The vast majority of ecommerce sites now provide the functionality to submit voucher codes to give money off certain products or services, and this can represent one of the most obvious opportunities to test pricing. At it’s most basic level, your “B” experiment will include the automatic submission of a voucher code to alter the cost of the product or service in question.

There are a number of challenges that you may face with this method however:

  1. Your voucher code system will not accept vouchers of negative value – in essence, this means you can only reduce, and not increase, your prices
  2. To truly test the impact of the pricing change alone, you will need to hide the fact that a voucher code has been applied – this usually means that a series of front-end changes will need to be made to hide the default functionality confirming the voucher code has been added
  3. Some voucher code systems will only allow one code to be used at a time and therefore if you are hiding the fact that a code has been applied as in #2, you may confuse some visitors who then attempt to add another code – depending on how widely you use voucher codes, this may represent a minimal risk, or it could be a showstopper for this method

Back-end triggered switch

If your website already operates a dynamic pricing model, then this may be the easiest solution for you. This will of course require additional development effort outside of just the test build code, but it does represent a more robust method than the front-end only voucher code option.

The basics of this approach involve creating a back-end setting to switch the price of a product or service to a new price, when a particular variable is present. That variable, often a cookie, is then deployed by the CRO platform that you are using as your “B” experiment to then be read by the back-end code to trigger the price change.

It is noteworthy however that when setting that cookie in the “B” experiment, you are likely to need to force a refresh of the page to ensure that the back-end script picks up the presence of the cookie & triggers the pricing change accordingly. The alternative to this is to deploy a standard front-end change in your “B” experiment that alters the visible appearance of the price you’re updating on the landing page, knowing that the back-end trigger will kick in on the next pageload. And depending on how many landing pages show that price(s) you are changing, this may be a better approach than forcing the refresh which would delay time-to-first-render a little.

Alternative product creation

If duplicating & editing an existing product on your site is a simple task in your CMS, this may be the easiest means of delivering a pricing A/B test for you. All of the details of the “new” product that you have created will match the existing product you want to test the pricing of, with the exception of the price itself.

In your CRO platform, you can then perform the front-end changes to change the visible price as well as to alter the action of your Add to Basket CTA to add the product with the new price to the basket. Once added, that new product will display its (altered) price in the basket as any other product would, and your visitors will be none the wiser.

This option can be very low effort if duplicating a product is easy for you, so should definitely be explored if you’re looking at testing your pricing.

Why bother?

If you ran a pricing test and discover you could have been charging 10% more with no negative impact on conversion, would you increase your prices?

If the answer is yes, then hopefully one of the three methods above will help you to achieve that.

If the answer is no, then you really are missing out on a key business lever – every business has to change their pricing at some point, and doing so without testing could land you in hot water.

Join the Discussion