How do I test new ads in my client’s Facebook ad account? That’s what we’re going to be talking about today in this installment of Online Confidential, where I take you behind the scenes of Secret Ad Manager Business.

So this was a question asked by one of my Elite Ad Managers students. This is a question, not just for somebody new, but for any ad manager, who’s been running ads for a long time. Sometimes we just don’t know, we feel like we’re winging it, or we find ourselves asking…should I be doing it this way, or should I be doing it that way?

This is what I recommended, and it’s a golden rule here that any ad manager will tell you, that if you have an ad set that is working, don’t muck with it! As Steve-O the crocodile hunter would say, don’t muck with it. We don’t want to upset an ad campaign that is working. Leave it and run! Don’t mess with it. So what I like to do is use a sandbox campaign, and this is where we’ll do testing. Typically, you will want to have around 10% of your ad budget put aside for ad testing.

Hats off to you if you’re running ads for someone and you only have ten or twenty dollars a day to spend; you have my biggest regards. It is so hard to run ads for clients who only have ten or twenty a day, or even fifty a day. It really doesn’t give you much wiggle room for lead generation, re-targeting, and a testing budget.

When you can test, ideally we want to be testing something all the time. We want to test every week with various ad copy audiences and all the other bits and pieces, so that when our ads are ticking along nicely and they’re performing well, we’ve already got ads that have been proving themselves.

They’ve got social proof, and we’ve got audiences that are working so that we can continue to horizontally scale as well as vertically scale. Having a sandbox campaign is where you can do that testing. Depending on your audience sizes and how much you’re spending per day on those ad sets,, it is going to determine how much you can put aside to spend in your sandbox.

So if you’ve got an audience size that’s a million people, and you’re spending $100 dollars a day on that ad set, or $200, maybe even $500, and you’ve got a sandbox and you’re wanting to test a new ad or new creative, you could just put $20 a day over into a sandbox campaign that is targeting that same audience just to test and see if these new ad copy or creative is working.

Are people clicking on it? Are people engaging with it? Are they going over? Are they registering? If it’s proven and if it’s working, then great. It may be in the sandbox, but I’m not going to muck with it. What I am going to do is gently scale it up. If you need to do things faster, you might go, “Okay, it’s working well at $20 a day. I want to see if I can jump this up really quick and put a hundred dollars a day behind it and see if it still performs at a hundred dollars a day.”

So you have $500 a day going on, and then in the same audience in your sandbox campaign going at a hundred dollars a day. You’re just testing over here. Now, if it’s a million audience size, then you’ve still got plenty of wiggle room. It it’s just something that you’re going to have to keep an eye on.

You might launch it at a $100 dollars a day. It might not work as well as it did at $20 a day. So you’ll just stop it, go back to your $20 a day and then just start gently scaling it up. And what does gently scaling look like? That will be like 20% every couple of days. So if you’re at $20 a day, you’ll add $2 and let it go for a couple of days, then add another $2. Meanwhile, with your $500-a-day campaign that’s working really well, you can be scaling that one up as well.

So the key for our sandbox or the key for testing new ads is to not try and put them into things that are working, but to put them over into a sandbox where you can test with these audiences.

Another great way to test our ad copy and creative is with warm audiences. For example, I like to do a warm audience to people who may have purchased. And I’ll serve that ad to them for seven days. It may be a registration for the webinar that they’ve recently joined, but if they have purchased, then they loved the webinar. They’re enjoying ideally what they have just purchased. They’ll see the ad and hopefully give it some love. They may even talk to a friend and say, “Hey, this is what I purchased.” Give us some great social proof. So that’s a good way to actually test an ad. Get some love on it initially so that if it does have wings, then you can try putting it in other ad sets. Just remember to not muck with things already working. Duplicate it, relaunch it, and scale it right. If you’ve got other assets that haven’t been performing so well, then you can try incorporating this new ad with social proof into those ad sets, turning off some of the ads that are not performing there, and seeing if the new ad in there gives a life.

If it doesn’t, then I duplicate, turn off the original of those ad’s ad sets, and relaunch their ad set. Then see if it relaunches with the new ad copy that resurrects that ad set for you. That’s how I go about launching new ads into campaigns or into any of the Facebook advertising that we’re doing for clients.

Golden rule: you do not want to mess with ad sets that are performing well. You want to be testing over in a little sandbox. When you’re in that sandbox, if you find things that work, run with it, and scale them up.

The name of the game is to get things working and to get our clients results.

If you would like to know more about being an elite ad manager, then I invite you to head over to the eliteadmanager.com and look at my certification program. It contains six modules where we go into various key elements to help make you an elite ad manager.