##### Preferences
• Metric: Average Revenue Per Visitor
• Statistics: Z-Test
• Tails: 2
• Confidence: 95%
• Power: 80%
• SRM Confidence: 99%
##### Data Input
To compute ARPV, the CSV file should have 1 variant per column, 1 order value per cell including non-converting visitors as orders of value 0, and no headers. If you need an example, you can download demo-arpv.csv. The 1st column always corresponds to the control variant, and the only column needed for test planning. In that latter case, you need to include one-week worth of values!

Pre-Test
Values per week
###### Variants
No variants present, pre-test mode

# Average Revenue Per Visitor Calculator

The average revenue per visitor, a.k.a. ARPV or RPV is actually composed of both conversion rate CR, and average order value AOV.

$ARPV = CR \times AOV$

$ARPV = \dfrac{Conversions}{Visitors} \times \dfrac{Revenue}{Conversions}$

$ARPV = \dfrac{Revenue}{Visitors}$

It’s recognized as one of the best metrics to optimize for, as it’s randomization unit is the visitor like for conversion rate, yet it allows optimization for revenue directly. It also contains the information of both CR and AOV.

To compute pre-test for it, you need a CSV file with one column of a week’s worth of order values with non-converting visitors having orders of value 0.

So the total number of values (rows) should be your total number of visitors. To run a post hoc analysis (post-test mode) you need the same for all variants (order values with 0 for visitors without purchase), and the first column being the control always, and for the duration of the test (if the test is still running, then the values should be up to the present moment).

RPV is the most important metric because it contains the information of both conversion rate and average order value combined. It allows optimizing for revenue while keeping the visitor as the randomization unit, which is extremely practical, unlike the average order value metric that has orders (i.e. conversions) as its randomization unit, which is a little more tricky to test properly.
Because $ARPV = CR \times AOV$, you can increase either the conversion rate or the average order value. So standard CRO techniques will definitely work well for this.

Before talking about Ecommerce metrics, let’s look at the important distinction between a metric and a KPI.

A metric measures how a particular workflow or process is executed on your site. It turns something abstract like - purchase intent - into something quantifiable that can be measured and improved.

A KPI is a Key Performance Indicator. It is a metric! But it is one that tells you how effective the workflows on your site are.

A KPI is in most cases a lagging metric because it is OUTCOME focused. You tweak leading metrics that you have control over, but your success is defined by your KPI improvements.

Here are some important eCommerce metrics are broken down by stages in the Buyer’s Journey:

• Awareness or Discovery:
• Total Traffic
• Impressions per page
• Visitors to Category pages
• Social shares of product
• Consideration or Desire: