A/B testing pricing
Is it okay to A/B test the price of your product. What I mean by that for example is sell it at $7 to some people and $5 to others during the same period of time.
Have any of you tried this?
Is it okay to A/B test the price of your product. What I mean by that for example is sell it at $7 to some people and $5 to others during the same period of time.
Have any of you tried this?
Please please don't A/B test anything until you have millions of users!
All the talk of A/B testing at tech giants like Facebook has somehow persuaded the rest of us that we can benefit from A/B testing too. Unfortunately, the laws of statistical significance tend to prevent tech minnows from getting any meaningful findings from A/B tests, even if we have hundreds of thousands of users.
When it comes to A/B testing pricing, even the tech giants can fall foul of the extraordinary aversion we all have to being treated unfairly. We just about accept paying more for an airplane ticket than the people sitting next to us, because we know that they booked earlier. We absolutely do not accept paying more on the basis of a more opaque algorithm, much less randomness. Famously, when Amazon customers found out that they were being charged different prices from other Amazon customers, they were furious, and Amazon quickly ended the test.
For sure, experiment with pricing. Raise your prices, and see if you make more money. If you make less, put your prices back down. But don't do an A/B test. You're unlikely to learn anything, and you'll risk antagonizing your customers.
I'm not a huge fan of A/B testing for pricing. You need a lot of data to really trust an A/B test, and in this case, that means a lot of sales. If a company is doing an A/B test for pricing, probably it's on the early stage, and early-stage companies don't usually have such sells amount.
There are better ways to find the correct pricing. At the end of this Hotjar post, you'll find a 1h30' video with some strategies (and maybe you can see me on the second row haha). https://www.hotjar.com/grow-your-saas-startup/pricing-strategy/
I would not recommend A/B testing different prices at the same time.
My thought would be either do it at different times or separate the price on any one vector.
E.g.,
1. Separate by channel: There could be a discounted price for IH folks, PH folks etc.
2. Separate by product features: Most common vector - Basic, Advanced, Pro etc.
3. Separate by lock-in: Another common vector - 1 month vs 12 month
4. Separate by position in the marketing funnel: New visitors might get a discounted price vs regular visitors pay full price. If you can collect email - send an expiring offer to that email to new members.
Hope this helps.
Seems necessary and shouldnt be a deal breaker when relative small. I'm planning to A/B test both price and also .99 / .00.