Sign in to follow this  
Followers 0
mki

Digital Advertising Test Budgets Explained

3 posts in this topic

Okay, this frustrates me. So I see a lot of advice regarding ad spend budgets that go something like this : "Start with a 5 dollar a day budget" or "Start with whatever you can afford" (I have no idea what that means to most people.)

This might be okay advice is this is your very first paid advertising campaign but what people don't seem to understand is that every variable in a campaign is a separate test.

So, just because a $5 a day budget campaign performs one way, doesn't mean that it's going to perform the same way with a $50 a day budget. It might, but there are multiple reasons why you could get two completely different results.

Also, realistically speaking, there's a maximum return on investment here. I really doubt you're going to spend 5 dollars and turn it into 50 dollars in profit, even if you could, I'm not sure why you would want to do that in the first place. Realistically, a good campaign will return a 100% ROI, so you make $5 a day. Oh Boy... $5

Better advice:

You really have to sit down and figure out what your goals are and work the math backwards, then build the campaign around the data you have.

As an example: Say I want to earn $200 a day with my campaign. I know from some limited tests that my conversion rate is something like 1.5% and I earn an average of $20 per conversion. I'm willing to spend up to $10 per conversion on advertising to generate it, so to earn $200 a day, I need 20 conversions a day and I am willing to spend up to $200 a day on ads.

So 100/1.5 = 66.7 * 20 = 1,333 clicks per day (With CPM based advertising we will have to run the ad(s) for awhile and estimate the CPC.)

200/1333 = .15 ~ 15 cents per click.

Also, ideally we are generating email leads as well. So we have to calculate our average value per email lead and we are going to have to estimate conservatively.

So lets say, 4% of the time, we get an email lead, but no sale, and we estimate that lead is worth $2 on average.

100/4 = 25, 2 / 25 = 8 cents per click.

So this breaks down into 3 separate scenarios:

1) Ideally, we are either, getting more traffic then we want at less than 15+8= 23 cents per click, or we are getting the volume we want at a lower cost. This is possible with testing and these cases might be worth testing first, if there's no quality score.

2) This could be difficult with something like Adwords, as the traffic is typically not there (we might need to target a lot of keywords and it's relatively competitive.) In this case, we either optimize for click volume, or the return on investment.

3) The math just doesn't work and we can't really get the campaign to work and we opt to invest in the possibility of return customers and long term mailing list profits. This is pretty typical with certain advertising platforms since it's usually not cheap to buy traffic at that volume. For this to work, we need to either be breaking even, or losing a very small amount up front.

If you believe that sooner or later, the tests will work out, you could end up with the campaign that works out pretty close to the original goals. If it's profitable, there might be room to scale it if that's what you want to do. But don't expect a campaign like that to get handed to you, it's a lot of work and a lot of testing.

Another thing:

Drop the attitude of "Spending Money" and start thinking "Investing Money." It's still very valuable information to know what doesn't work.

To me, the "spend" is the money I lose, not the budget.

Don't cut ads super early, try to let campaigns run until you get statistically relevant data. If the above campaign worked regarding the volume and leads, but the conversion rate was only .70%, ($21.43 CTA) personally I would still let that campaign run for an entire week. Then I would start cutting lower value placements, possibly use day and time parting, try different segments, and swap ads. Without a weeks worth of data it's going to be difficult to make those kinds of decisions. So for week one the campaign ran at a financial loss, this to me, isn't a big deal. I could also try redoing the math and lowering the bid price and try to make the loss back.

The only campaigns that should be cut early are the ones that are clearly burning money. So if we buy 800 clicks at 23 cents each ($184) got zero conversions and our lead conversion rate is super low, I would cut that... Clearly something isn't working right.

"Spending" a couple hundred dollars on that campaign is great investment, in my opinion. Realistically, at those numbers ($21.43 CTA+$2 Per Lead) , I'll make it back with the mailing list anyways, it will just take a while.

Note:

If you're like me, you will use spreadsheets to factor in your other costs per CTA, typically over 30 days. Things such as hosting bills, ESPs, etc. I personally like to know, to the penny, exactly what is going on. This also allows you to marginalize the campaign if you desire. Obviously to most people, as long as the funds clear in your account relatively quickly, a 30% profit might be okay if you're spending $500 a day on ads.

This can also get a little bit complex if you are using a retention system (such as retargeting), which I really think you should.

Mike Friedman likes this

Share this post


Link to post
Share on other sites

I hate the "start with $5 a day" line too. Unless you are getting clicks for $.10 or under, you are not going to get enough data to have any clue what is and isn't working. 

You are right. There are so many variables to a PPC campaign, and your daily budget is a big one. At $5 per day, you are competing with everyone and paying premium prices for each click. At $500 per day, you may have competitors in the last 4-5 hours of the day who have hit their max budget. You are no longer competing against them and can get clicks for 20-40% less (or more). Those are just random numbers. Every campaign is going to be different. 

If you don't test with just maxing out the day a few times though, you will never discover that. You can mess around with what hours your ads run at, but I have never had the same results with that. For AdWords, it is almost like Google rewards you for running ads all day and tries to keep from rewarding you for holding off until competitors have drained their budgets.

If you can only afford $5 per day, it is probably a good idea to just admit that you cannot afford pay-per-click advertising at this time.

 

deezn and mki like this

Share this post


Link to post
Share on other sites
1 hour ago, Mike Friedman said:

I hate the "start with $5 a day" line too. Unless you are getting clicks for $.10 or under, you are not going to get enough data to have any clue what is and isn't working. 

You are right. There are so many variables to a PPC campaign, and your daily budget is a big one. At $5 per day, you are competing with everyone and paying premium prices for each click. At $500 per day, you may have competitors in the last 4-5 hours of the day who have hit their max budget. You are no longer competing against them and can get clicks for 20-40% less (or more). Those are just random numbers. Every campaign is going to be different. 

If you don't test with just maxing out the day a few times though, you will never discover that. You can mess around with what hours your ads run at, but I have never had the same results with that. For AdWords, it is almost like Google rewards you for running ads all day and tries to keep from rewarding you for holding off until competitors have drained their budgets.

If you can only afford $5 per day, it is probably a good idea to just admit that you cannot afford pay-per-click advertising at this time.

 

Now that doesn't mean I don't recommend experiments, there are definitely things you need to know.

For example, I setup a campaign to buy placement data, and I noticed something weird about the targeting.

Simply put, I would personally cut the vast majority of the placements.

The ad is only running on low quality gun websites, the volume is way low, and I'm paying about half my bid.

No idea what's going on here, I'm not advertising guns... So ...

A different targeting option performed wildly differently and is much more in line with what I was after.

After looking through the placements, all I'm going to say is that some people have some pretty interesting ideas about what a home defense weapon is.

Spoiler

combat-handguns-august-2016.jpg

I mean, I guess it would make some people feel safer...

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0

  • Recently Browsing   0 members

    No registered users viewing this page.