I'm split testing two different offer pages and so far I've got about 400 clicks to each. One generating $180 in revenue and the other $137 in revenue. It's been running for a few days now and I just don't know when to decide its conclusive. Anyone have a rule of thumb on this? lol:food-smiley-010:
How much split testing is enough?
- Thread starter affluent
- Start date
google "split test calculator."
the top 3 results will give you seemingly similar calculators,
but for some reason they produce different results.
Try your numbers out on all 3 and see if you find a trend between them.
Rules of thumb are
90% Confidence Interval means there's a 1 in 10 chance you're going to pick the wrong thing based on your data.
95% -> 1 in 20
97.5 -> 1 in 25
99% -> 1 in 100
good luck bros.
the top 3 results will give you seemingly similar calculators,
but for some reason they produce different results.
Try your numbers out on all 3 and see if you find a trend between them.
Rules of thumb are
90% Confidence Interval means there's a 1 in 10 chance you're going to pick the wrong thing based on your data.
95% -> 1 in 20
97.5 -> 1 in 25
99% -> 1 in 100
good luck bros.
I like to test at least 200 clicks but I've concluded with way less and more. Depends really on the feel I get.
This may help
"For example, after 5000 visitors to your page page A generated 12 sales and letter be generated 18 sales. so here, total sales is 30. Square root of 30 is 5.4. The difference in sales in two page is 6. Now, the difference is greater than the square root. That means page B is performing better than page A."
Split Testing For Dummies|Deelip Khanal
"For example, after 5000 visitors to your page page A generated 12 sales and letter be generated 18 sales. so here, total sales is 30. Square root of 30 is 5.4. The difference in sales in two page is 6. Now, the difference is greater than the square root. That means page B is performing better than page A."
Split Testing For Dummies|Deelip Khanal
well seems logical but
you concider it a rule of thumb? because after 5000 visits the fact that
18>12 means page B is better,without the square root.
or if the differnce is lower then the square root it mean i should take another look at whats working better .
you concider it a rule of thumb? because after 5000 visits the fact that
18>12 means page B is better,without the square root.
or if the differnce is lower then the square root it mean i should take another look at whats working better .
I think that's true to a point but this is a case of diminishing returns. At some point your time and resources will be better spent setting up a new campaign and focusing on split testing that. Rather than squeezing another 0.05% out of a highly optimized campaign, you could squeeze out 50% increase in another campaign.
to the OP, sorry, but I don't think anyone can definitively say where that imaginary line is..."It depends".