Google Ads Household Income Targeting

Demographic targeting can be a really valuable tool to make sure that you’re reaching the target audience across all different platforms. Pretty much every channel has the basic demographic information around age, gender, most of them have parental status that sort of thing.

But one that I get asked about more often than not is Google Ads household income targeting. I personally don’t think it’s quite the magic strategy that I think a number of other people do, but that doesn’t mean that it can’t be useful in some specific accounts.

So in this video, I want to walk through what the household income targeting is on Google Ads, talk about where you can apply it, and maybe some considerations before integrating it into your strategies.

I want to start off by talking about how Google comes up with its household income targeting, so that you can determine for yourself, how accurate you think it is, because it’s not perfect. Historically, Google used to use the us census data and grouped zip codes together to effectively make income categories based on the data in the census.

So essentially, the household income levels, the way they used to be calculated were just an extension of zip code targeting. But for a number of reasons, Google changed the way that they come up with these household income levels. There are a number of signals that Google now uses to put together household income targets that are a little bit more specific than just zip codes.

First is based on demographic information, people provided directly to Google for their Google account. There are ways that people can provide their income for the account, and Google will use that. Google also tries to estimate your income level based on activity while you are on a Google-owned property like YouTube.

They can also use data shared from websites where a user has given income information. Google says that sites like social media sites are one of those types of things where they get that information from, and they also use other publicly available data to come up with some estimation of your current income status, to help group you into one of those buckets.

One of the benefits of Google shifting the model of how they come up with household income targeting away from purely based on U.S census data, means that they can now offer this in a number of different countries outside of the U.S. When this option came out originally, it was only a United States targeting option, but now there are a number of different countries.

I have two different tables here, because display campaigns have a limited number of countries that you can use household income targeting for, search campaigns have a much longer list. The other takeaway from this list is that you can see the different categories that Google uses for income targeting. It’s going to be based on a percentage of the country. So it’s top ten percent, 11 to 20%, 21 to 30, 31 to 40, 41 to 50, lower 50 percent, and unknown.

The percentages seem to make sense, you can kind of follow that you can basically target within 10% increments of the top 50%, and then the users that are designated in the lower 50% of household income for that country are grouped altogether. I think the unknown portion scares people a little bit, I see people opting out. So let’s go to the Google help section and see how they define the unknown group.

Essentially, Google Ads says that they can’t know or infer the demographics information for all people on the network. So unknown just refers to people whose age, gender, parental status, or household income hasn’t been identified.

Additionally, specifically for display campaigns, there are some websites that opt out of demographic targeting, and if you don’t have the unknown category selected for your targeting, you will not show up on that website. So it’s not even that Google hasn’t inferred it about the user, it’s that the website itself on the display network has opted out.

So although it might seem appealing to opt-out of an unknown group, I would encourage you to leave it in and always just let performance speak for itself. If it doesn’t generate the conversions or performance that you want, then it might make sense to make some adjustments.

But don’t opt-out of the unknown category just because you think it’s something that’s outside of the income targets that you want. There easily could be people in the top 10 that are still in that unknown category, Google just hasn’t bucketed them that way just yet.

So now I want to hop into an account and show you how to apply household income targeting, and some of the different ways that you can customize it for your performance. To get started, I’m just going to hop into one of our search campaigns. To apply household income targeting, you need to head over to the audience’s tab within Google Ads.

For those of you who’ve been advertising on the Google Ads platform for a while, like I have, you’ll remember that the household income target used to be part of the audience segment, and I’m going to show you where that is. So anybody who’s new to Google Ads, bear with me for just a little bit while I show us old folks where it used to be, and then I’ll show you where it is now.

So if we head into audience segments, I quickly switched over to browse, and you’ll see the first one at the top is who they are or detailed demographics. If we click on that, you can see that parental status, marital status, education, homeownership, and employment are all listed here.

Household income used to be one of these targets that was in here, and this is how you would apply it. But Google has shifted away from treating household income like an audience, and now it is a core detailed demographic that lives in the demographic section where age and gender live.

So I’m going to cancel out of this, and I’m sure you saw it earlier while I was going on my rant about new folks, young folks all that stuff. But the household income is now down here in the demographics tab. You can see over here that household income is one of the three categories for this search campaign, and now you can see the bar chart for the performance, for each of the different income levels that Google has broken folks out into.

We’ll come back to the performance charts in just a little bit, but first I want to show you the table which even though I don’t like this user experience, you can scroll down after expanding it. Now you can see the household income targets, all the one that we talked about, top 10 percent, 31 to 40 percent, and the performance for each of those off to the right.

As you can see in this ad group, we are opted into all of the different income categories from the top percent all the way down to unknown. If for any reason you wanted to opt-out categorically of any of these different segments, there are a couple of different ways you can do it.

First, let’s say you want to categorically exclude any of these different segments. Let’s say for whatever reason, you just really are not liking that lower 50% group, and you want to opt-out completely. To do that, you would come up to the edit demographics section up here, and then over on the right, you have a check box of each of the different groups. To opt-out of any of them, all you have to do is uncheck it.

For this campaign, I don’t actually want to do this, so I’m going to opt back into it. But let’s say you had it unchecked for a while, you let performance run and you decided you wanted to opt back into the lower 50 percent, it’s just as easy to come back in here, recheck the box, and then click save demographics.

Maybe you don’t want to completely exclude any of these income targets, but maybe you just want to lean into or lean away from different groups based on performance, and that’s where you can utilize the bid adjustments, in the way that you would for any other bid modification layers within your account.

You’ll have to be on the ad group view to utilize the bid adjustments, you cannot do this at the campaign level, which I personally find frustrating and seems a little bit silly. But at least, we get the option to modify them.

So based on performance here I might say that on the top 10%, I want to decrease my bid by 10% because the cost per conversion is a bit higher than some of the others. So I’ll just click in here, come down to decrease and add my bid modifier layer, then click save.

The unknown group, and the 21 to 30 percent group are doing pretty well, so I’ll actually give them a bid increase of 10% each to try and lean into this group a little bit more and get more performance out of it.

So we have the table of all of the different performance data down here, but I did mention the charts that we could see up at the top, and there are a few different ways that you can model this data to see things a little bit more visually. In this bar chart, you can choose from a number of different performance metrics, you can use anything in this performance group, conversions, or some of the competitive metrics to see maybe what your impression share was for each of the different income targets, which could be really interesting.

All of the data in here is going to be for the given date range that you have chosen to review, so far, I’m reviewing it for just this month. But you can change the chart type, and you can start to see a time series, or a stacked time series. So let’s just look at regular time series, and now we can see the click values for each of the different income layers over the course of time for this month.

Right now it looks a little bit goofy because I only have 20 days’ worth of data and it’s set to weekly, so you can change the groupings that you have here. Now we can start to see a trend of which income targets had higher days, lower days if there are any specific trends.

Kind of interesting that the top 10 seems to be getting a lot more aggressive in the past few days, not totally sure why, but it might be worth looking into. We can also utilize a stacked time series chart, which basically shows the same kind of thing. But just a little different view of the data.

Again, we can see a number of different metrics for this, so we could shift into cost, impressions, conversions, even some of those impression share metrics that I mentioned earlier.

It’s a little bit less compelling when you look at it this way, but there’s a lot of ways that you can review the data and make sure that you’re adjusting your bid modifiers accordingly. Looking at these time series charts would be great to try and identify what impact on performance you had by adjusting bid modifiers.

Let’s say that on November 10th, I decreased the top 10 percent bid modifier. I might want to come in here look at some of these time series for a number of different metrics, and see how the 10% segment trended over time, and if my bed modifier did what I wanted it to do. That’s pretty much it for household income targeting.

We talked about how Google gathers the data, how you can apply it to your campaigns, and how you can review performance to make sure that you’re optimizing the income targets based on how your campaigns are performing.

Hopefully, through all this, you’ve kind of seen my philosophy on using the household income targets. I personally don’t think it’s something where you should come in and target only the top 10 percent of users, because one, there could easily be top 10 percent people filtered into the unknown group, simply because the data is being blocked.

But additionally, the way that Google gathers the data and buckets, people, into these groups is not perfect, it’s not a science, it’s absolutely some art going into it. I think you’d be limiting yourself if you targeted only a specific group of people.

Rather than doing that, I would treat household income in the exact same way you do any other demographic information, audience, location targets, anything where you can have a bid modifier layer. Start by opting into every category, review data on a regular basis, and adjust the bids accordingly based on the performance that you’re seeing.

If over a long period of time you don’t see any positive trends out of it, that might be a good time to opt-out. But until the data speaks for it, there’s no reason to limit the number of people you’re targeting, based on inferred household income.

I know that household income targeting is absolutely appealing to many advertisers, and I’m sure that many of you or your clients have been using it or are intrigued by it, maybe you’re having a number of conversations about it.

I would love to hear any additional questions you have for me about household income, or any strategies that you’ve seen work really well in your accounts. So feel free to share those questions or your additional strategies in the comments.

Written by Michelle Morgan