In our tests I looked exclusively at B2C webshops
Posted: Sun Jan 19, 2025 8:10 am
In the example you see seven orders with an average order value of 102.86 euros. That is the bottom of CLV. You also see that customer value is on average 144 euros. That is 1.4 times the average order value. You can also calculate this ratio by dividing the number of orders by the number of unique email addresses.
From data from B2C retailers I saw that an average customer made a purchase from the webshop between 1.18 and 1.27 times, in a period of the past two years. Incidentally, the tested webshops did not do e-mail marketing, the number is higher there. I also did not include webshops that sell consumables, because the number is also higher there. Comments:
Don't forget to filter out the rejected orders or failed orders that have been redone. This will prevent double counting.
Phone numbers can be fake (for example: 1234567890 ). In our data this was rarely the case.
It is possible that the same customer places an order more often under a different e-mail address. Then the CLV is also higher. In these tests I did buy georgia whatsapp database not take that into account. I think that it will not deviate more than 0.5% from the actual value, provided that there are enough orders. In our tests it was about thousands of orders per webshop.
Customers can also enter your webshop twice via paid channels. Then you have also made marketing expenditures twice. If that is often the case, then the CLV is lower.
I tested this method with several fields to define unique customers, including phone number, email address, and mailing address. I found around 0.5% difference in unique counts between phone number, email address, and mailing address. That means they perform about equally well.
It is plausible that for B2B retailers a postal address or VAT number may be a better indicator of a unique customer.
If you are going to use phone numbers or addresses, it is useful to clean them up. For phone numbers, you should remove all characters except the numbers.
From data from B2C retailers I saw that an average customer made a purchase from the webshop between 1.18 and 1.27 times, in a period of the past two years. Incidentally, the tested webshops did not do e-mail marketing, the number is higher there. I also did not include webshops that sell consumables, because the number is also higher there. Comments:
Don't forget to filter out the rejected orders or failed orders that have been redone. This will prevent double counting.
Phone numbers can be fake (for example: 1234567890 ). In our data this was rarely the case.
It is possible that the same customer places an order more often under a different e-mail address. Then the CLV is also higher. In these tests I did buy georgia whatsapp database not take that into account. I think that it will not deviate more than 0.5% from the actual value, provided that there are enough orders. In our tests it was about thousands of orders per webshop.
Customers can also enter your webshop twice via paid channels. Then you have also made marketing expenditures twice. If that is often the case, then the CLV is lower.
I tested this method with several fields to define unique customers, including phone number, email address, and mailing address. I found around 0.5% difference in unique counts between phone number, email address, and mailing address. That means they perform about equally well.
It is plausible that for B2B retailers a postal address or VAT number may be a better indicator of a unique customer.
If you are going to use phone numbers or addresses, it is useful to clean them up. For phone numbers, you should remove all characters except the numbers.