- Joined
- May 9, 2018
- Messages
- 4,259
In the last week, I decided to do a very "dumb test" where I run a campaign from Brazil and split test my VPS and CDN to see if there's any difference in terms of CTR/CR.
South America, Africa, and the Middle East are always tough places to run because it's difficult to have a server in those regions as they're not as widely available as North America, Europe, and APAC. Because of that, it makes sense to use a CDN if you know there's a pop located near that region.
My tests consist of:
Target geo: Brazil
VPS location: Vultr USA
CDN: Pop from Brazil
I checked and confirmed most of my visits on the CDN were hitting the local pop and my stats showed 99.5% of the visits were hitting the cached lander on the CDN.
The distance between the USA (Chicago) and Brazil is quite considerable. Almost like going from one continent to another around the globe. So I was expecting my CDN lander to outperform the VPS my a noticeable margin.
Here are the results:
After sending 250,000 visits to each server, I realized for this test they are performing almost exactly the same.
CTR
South America, Africa, and the Middle East are always tough places to run because it's difficult to have a server in those regions as they're not as widely available as North America, Europe, and APAC. Because of that, it makes sense to use a CDN if you know there's a pop located near that region.
My tests consist of:
Target geo: Brazil
VPS location: Vultr USA
CDN: Pop from Brazil
I checked and confirmed most of my visits on the CDN were hitting the local pop and my stats showed 99.5% of the visits were hitting the cached lander on the CDN.
The distance between the USA (Chicago) and Brazil is quite considerable. Almost like going from one continent to another around the globe. So I was expecting my CDN lander to outperform the VPS my a noticeable margin.
Here are the results:
After sending 250,000 visits to each server, I realized for this test they are performing almost exactly the same.
CTR
