I usually recommend that you have a few experiments active at once.
- testing out a new channel to acquire customers
- testing out an idea in one of your existing channels
- testing out optimizations in your area of focus for the month/quarter/year
All three are new ideas or new things. You're trying to see if they are better than what you have.
What's unspoken though is that the majority of your store stays mostly the same. You're not releasing a brand new design, while throwing away your entire product catalog, and shifting your target market every week.
Besides being insane, that would make it impossible to run actual experiments. You'd have no idea what caused any improvement and you'd be relying on luck.
No, what you do is keep almost everything the same while you change a few bits. Then later on ones those are working, you change a few more things.
This is the standard optimization process and you already know it, even if you haven't thought through it this way.
What I want to point out is this, the new changes get the most attention but it's predominantly "everything else" that gets the most resources and contributes the most to your store's success.
This is what creates stability in your store. This is why it doesn't crash and burn every single day. This is why customers buy and come back to buy again. This is why the historic metrics matter so much in Repeat Customer Insights.
I just want to shine a small light on that aspect that rarely gets mentioned, even by me.
Eric Davis
See how the month a customer orders will change their behavior
Repeat Customer Insights will automatically group your customers into cohorts based on when they first purchased. This will let you see how the date customers bought would impact their behavior.