TikTok Ads Audience Insights

It’s no secret that the advertising platforms have a lot of information about the users on those platforms and that they use that to help advertisers reach certain audiences of people. Now, we don’t get access to their personal information, which is great. I probably couldn’t steal your identity even if I wanted to, genuinely don’t know-how.

But I do want to know how large groups of people perform within my campaigns. Whether younger or older audiences, men or women, may be different device categories. How are all those impacting the performance on my campaigns?

We’ve talked about some of these different breakdowns, and some of them are disappearing on other platforms like Facebook with privacy issues. But one of the platforms that is still giving a lot of insight is actually Tik-Tok, one of the newer ones. So today, I want to run through the Tik-Tok ads audience insights section of the interface, show you the data you can get out of it, and talk about some ways that you might use it to optimize your campaigns or even change strategy moving forward.

When you head into the Tik-Tok ads manager, the first place that it always lands you, or at least that it always lands me, is on the dashboard. This is actually going to be the first place that has some of this audience insights information. You’ll see up here in the upper left we are on dashboard because it has this little blue line, but later we’ll hop into some of the other areas of the account.

At the very top, you can see some of the stats for today, the fact that we have one campaign active, and then performance trends. But if we scroll down the page a bit, we’ll be able to see some of the first sets of the audience information.

The first grid here on the upper left is going to just show the one campaign that I mentioned, but over on the right, we’ll start to see the gender information. Whether the users in the campaign so far are male or female, and you can see right now for impressions, we have about 63-ish percent are female, and the remainder are male. Then we can use this drop-down up here to choose any different set of stats that we want to see.

So we can see whether the breakdown for clicks is about the same, which looks like more females have clicked on the ad, or we can start getting into other metrics like the cost per click, where we can see that the chart has changed, and you can see that things are actually a little bit cheaper only by about one cent cost per click for females. Then you can even come up and see the difference in conversions for the campaign.

Here, we have 21% are coming from females, a little over 9% are coming from males, but 63% are coming from unknown. That’s because the conversions that we’re recording for this campaign are based on website actions, and these are going to be people who have opted out of being tracked.

So while you’re going through these different reports, just know that there are going to be gaps in the performance stats for anything that relies on a third-party pixel on your website, because of all the different privacy issues that are out there right now. The next box down here on the bottom left is going to show the placement performance, and you can see for cost, it’s going to be all of it is on Tik-Tok, 100% is.

Long story short, for this one, everything is going to be on Tik-Tok if you’re in the U.S., you can’t target any of those additional placements. You could do the breakdown for any of these additional stats. So if you’re in a country that allows you to target all the different placements that are available on Tik-Tok, this could be really helpful to see if Tik-Tok or any of the additional placements or platforms is performing better than the other.

Think about this as the difference between Facebook and Instagram in Facebook ads manager; you would want to know if Facebook or Instagram is performing drastically better or worse than the other to be able to optimize a little bit better.

Then the next chart over here in the bottom right is going to tell us about the operating system and type of device that users are using when they engage with our campaign. So anything in blue is going to be on the iOS platform, anything in purple is iPad, and then anything in this green is going to be android.

As you can see, we have basically no android users so far, either impression, conversion, or click-through rate-wise, and each of these will operate just in the same way that the drop-downs did for the other two. You’ll just be able to compare three of them at a time, and the charts will look a little bit different depending on the metrics.

So obviously, impressions and conversions both have this bar chart, but then click-through rate is going to have these different tables right here, and it’s just going to show you the number because they’re not going to make a vertical bar chart alongside the horizontal bar chart.

But as I mentioned, there are a lot of different metrics, so you can use impressions, clicks, click-through rate, cost per click, cost per conversion rate, and CPM, and as that scrolled down on the screen, it took us down to the bottom portion of the dashboard where we can see the same performance stats by hour of day and day of week, and it’s going to use this heat map to show us where the performance lies.

So here you can see that for impressions for week 41 of 2021, we had most of the impressions down here between Sunday to Monday, kind of in the afternoon-evening hours. But if we wanted to see where the cost per clicks were, and how expensive they were, you can see here that the most expensive costs per click were actually kind of toward the middle of the week.

So Tuesday to Thursday and to Friday, and kind of in the late morning hours. So here, just remember that if you see the darker color on the heat map, that means that the number is higher, doesn’t necessarily mean it’s better.

So for cost per click, you can see that it’s 44 cents, 46 cents, but where we were seeing more impressions, it is actually closer to 29 cents. But for things like conversions, the darker spot is going to be a good thing, because that’s where you’re seeing most of your conversions come from.

Overall, the information on the dashboard is really helpful at a very high level to try and give you some direction on the performance that you’re seeing in the account, but it definitely is not the end of the story when it comes to reporting. For that, I’m actually going to scroll back up to the previous grid where we had the campaign; I’m going to click on the campaign itself and hop into that platform.

So now you can see it took us into the campaign portion of the interface, and we have all the campaigns listed here. Again, this account only has one active currently. This is going to be the best area of audience insights in the entire platform. To find it, you just need to hover over your campaign name and click on view data. This will bring up something very similar to the inspect tab on Facebook, if you’re interested in that, you can check out the video that we have put together right here.

The main page of this tab lets you see overall trends and performance, but there is a small subset here that lets you see placements, and again, for this account, this isn’t going to be extremely helpful. But you’ll be able to see the different placements by all of the different stats that you want to see right within the campaign itself.

So again, the dashboard is going to show you the snapshot for the entire account, but this tab will start to show you individual campaign performance, and how that is broken out by the different placements we have here. Again, ours are going to be all purple because that’s the only place that we can target, but if you have any of the other placements in the account, it will start to show you the performance there too.

So the takeaways for this placement section are going to be relatively similar to what we saw in the dashboard tab, just at an individual campaign basis. But where things start to get really good, in my opinion, are if you come up here back into the main chart and you click on audience.

This is the section of the interface where we’re going to get the most information about the audience that we’re targeting; you’ll see here there is a bit of a caveat, that audience data will take eight hours to process on average, and the information will be updated once it is released. So some values are estimated, so just keep that in mind, but they are pretty good, in my opinion.

We start off with an audience breakdown section here, and we can start to see performance based on age, gender, and age and gender. So it’s going to default to age, and in this account we are only targeting users between 13 and 24, so this is accurate. But if your campaigns have all the other age ranges in there, it will have different pieces of the pie chart.

We can then also break things out by gender, just in the same way that we saw on the main tab. If you want to, you can even break it out by age and gender, and here, we’ll start to get into a bar chart rather than a pie chart.

Just like with all the other metrics, we’ve been able to see we can use either cost, impressions, clicks, conversions, results, click-through rate, conversion rate, results rate, cost per click, cost per thousand impressions, cost per conversion or cost per result. So there’s a lot that we can do here. So let’s say I want to look at cost per conversion by age and gender.

Now we can see here that in the 13 to 17 range, the cost per conversion is about the same, $15.18 for males, $14.96 for females. But as we get older, males get a lot more expensive on these conversions than females do, we’re up to almost 23 dollars for males, but females have gone down, now we’re down to just below eight dollars which is even quite a bit less than what we saw in the 13 to 17 age range.

So depending on what we’re doing in this campaign, what creative we’re using, all this sort of thing, it might make sense for us to start to think about ways that we can target and interact with these users, and really lean into the fact that older females convert at a much lower cost. Or we can focus on it from a different angle and think about how we can get the older males to convert a bit better and be a little bit more cost-efficient.

Down below, every one of these charts that we’re going to go through, there will be a detailed analysis. So here you can see the age and the breakdowns for the cost, cost per click, impressions, click-through rate, clicks, and then if you scroll to the right, you’ll get to see a number of the other stats that are available here as well. So this is a really helpful way to see at a high level without needing to use the tables and charts up above.

As you can see, we have age, but then we also have gender, age, and gender, and then a number of the other breakdowns that we’re getting ready to get into. So we’ll have the country and region, sub-region, network, interest, behavior, device model, and operating system.

So lots of control that we can have down in this detailed analysis section. But the biggest takeaway is that as I go through these other breakdowns that we see up here at the top, just remember that there is a detailed analysis correlation table that you’ll be able to see all the performance for as well.

So the next tease is going to be the location breakdown, so here we have everything ranked by the performance that we’re seeing here, and again, this account is relatively limited, we are targeting only Tennessee in the United States. So this table looks to be accurate, makes all the sense in the world to me. This top countries and regions piece, there’s nothing that we can really do any breakdown-wise other than choosing a different metric. So this is always going to start off at the country level, but the top sub-regions are going to be a little bit more dynamic based on where you’re targeting.

So again, in this account, we’re targeting just Tennessee, but you do have the option to change the breakdown here by clicking in, and you can have all countries and regions, which is going to be the default, but then depending on which country you’re targeting, all of these currently are just static lines, you can see that it’s just Canada and there’s nothing.

But for the United States specifically, there’s this little arrow over here off to the right, and now I can choose to look at performance either by state or by DMA, which is going to be a lot more valuable for this account specifically.

So since I’m targeting the entire state of Tennessee, this DMA breakdown is a lot more useful because now I can see that Knoxville actually has the most expensive cost per conversion, but then Nashville, Memphis, Chattanooga, Paducah after that, there are lots of different ways that you can break down performance within the state that we’re targeting, and try to make performance a little bit better based on the insights you can find on this tab.

The next section is probably the one that I think is the coolest; if I’m being honest, it’s around the interest and behavior performance within this campaign. Now we can start to see the performance and these different performance metrics, so for right now, I’m going to shift this back to impressions just to do it.

We can start to see this based on the interests and behaviors for the account. But remember that interests and behaviors are created a little bit differently on Tik-Tok than some of the other platforms. So I’m going to hop into the builder for this campaign to just give a quick overview of that. So here we have the interest and behavior sections; interests are based on the users and how they’ve engaged with certain videos on the platform, and the videos that they’re interested in.

These are considered long-term trends, so over a long period of time, these are the types of videos they have interacted with, so they are interested in them. Behaviors, on the other hand, are shorter-term, and it’s how your users have specifically interacted with certain videos or, in this instance, certain creators on the Tik-Tok platform. So again, think of interests as long-term segments that people are put in, and behaviors are more short-term. We did do a video that talks about all the targeting options on Tik-Tok, so if you’re interested in that, you can check that out at the top of your screen right now.

But back on the audience insights portion, we can see that there are a number of different interests that people are interested in, so apparel and accessories, games, food and beverage, news and entertainment, and all sorts of other things that are in here.

You can change the breakdown, because right now, it’s at tier one, but if you want to do something that’s as low as tier four, now we can start to see the very specific breakdowns of what these are. Again, I’m going to hop back into the builder to talk about the difference between the tiers, because that is a little bit confusing.

So on the last tab, the example that I showed was in interest, and one of them was games. So let’s head into interest, and it’s asking what kind of videos are they interested in. So if I click here, it’s going to bring up this list, and I’m going to scroll down a little bit, because games is going to show up here.

It is a tier 1 option because it is showing up right at the very beginning. But if we want to get into tier 2 options, it’s going to be games by type; the tier 3 options are going to be shooting, sports, puzzle games, and then the tier 4 option that I chose is going to be first-person verse 3rd person versus air combat games.

So this is how we talk about tiers when it comes to the breakdowns. It’s going to be tier one as the first layer, and then if there is a drop-down, you’ll have a tier two. Again, if there’s a drop-down, you’ll have a tier three, if there’s another drop-down, you’ll have tier four, but it will not go below that.

So jumping back into the other tab, you can see that since we’re on tier four, we’re seeing the third person versus first-person shooter games that correspond to the tiers that we saw before. But if we wanted to see something based on tier 3, if there’s not a tier 4 available, we would have to choose that, and now we’ll see puzzle games that is in the list that we just saw, but it did not have a drop-down.

So to see puzzle games, you have to have chosen tier three. So this isn’t the best user experience in my opinion, but there’s still quite a lot that you can see here. So now you can start to break down performance by these top interests that different people have based on their long-term experience and engagement on the Tik-Tok platform.

Then the behavior chart down below is going to work very similarly, but again, this will be based on short time periods. So here you’ll see anything that is based on performance, 15 days entertainment, 15 days and it will always have a date range behind it. But you can then choose different tier levels to get performance stats for, if there’s enough data available to pull.

So here, we start to go down to tier two, and now we can start to see comedy, whatever lip means, daily life, games, so on and so forth. So again, you can continue to break things down further, and you can utilize all the different metrics available in the same drop-down that we had before. But again, as you start to get more and more granular, you’re going to miss out on some of the data performance.

Then lastly, in this audience breakdown, we can see device, and this is going to look pretty similar to what we saw before; we will have the metric performance breakdowns that we have up here at the top, and then you’ll see the operating system down below for those metrics, as well as the device model or the brand more likely, based on what we’re seeing here.

So most of that is going to be coming from Apple, but there are some L.G., Samsung, Motorola. But then the one that I also like up here at the top is it does show you the network that people are using. More often than not for this account, 90% of people are on wi-fi, but there are some on 4g, 3g, 2g, and then kind of the unknown last little group that’s there. There’s not an option for 5g showing up here yet, which I find a little bit interesting, but it is what it is.

But since network targeting is one of the options that you can choose within your campaign, it might make sense for you to start to look at the performance differently here. Maybe you notice over time that wi-fi has a far lower cost per conversion than the other networks, or maybe vice versa, so you might want to adjust your campaigns differently based on how those different networks and connections are performing.

So I’ve given you an overview of how this entire audience tab works. There are only two more things that I want to show you before we end this video, and the first thing is going to be very short, because the biggest takeaway I have for you here is that all of this data that we can see on this chart, this data breakdown, whenever you go in underneath your campaign name and click view data is also available at the ad group and creative level. So let me show you that.

This account, for simplicity’s sake, only has one ad group. So if I show you the view data tab, it is going to look very similar to what it did for the campaign level, and we will still have all the different breakdowns the way that we wanted to for audience as well. So we can do all the same metrics at the ad group level.

So if you have lots of different campaigns, you can see performance differently by that, and you can see different performance for the different ad groups based on the targeting and settings that you have in the account. Then lastly, you can also do the same thing at the ad level as well. Each of these different ads has its own view data tab, and all of them are going to be basically the exact same as what we saw at the campaign level.

You’ll be able to come in and see the overall trends in performance, you’ll be able to see placement performance, and you’ll be able to do the breakdowns for age and gender, location, interest and behavior, and devices for all your different creative. So again, all of this means that whether it’s at the campaign, ad group, or creative level, you have all the insights into how your audience is engaging with your campaigns, and be able to find areas to optimize within your account.

The last thing I want to show you is that whenever you find a table that you like, let’s say this detailed analysis tab, you’re just really into it, and you want to get this information out of the platform in a little bit more usable way, all you have to do is come over here and there’s this little icon over here that lets you export data.

If I click on that, and it will then download into an excel sheet to where you can utilize the data; however, you need to put it in a pivot table, share it with somebody any different way that you need to do that. All of these different data tables are downloadable in whatever fashion you want.

Overall, there’s quite a lot of performance and insight information you can get out of the Tik-Tok ads platform, so I encourage you to go in and, whether it’s at the campaign, ad group, ad creative level, see how the different demographic groups, interests and behaviors, device groups are performing and start to leverage those insights and how you optimize your campaigns, and how you build your campaigns moving forward.

You might find that it makes sense for your account to always have wi-fi as a separate campaign, compared to the other platforms, because of how performance is. Or you might find that it makes the most sense to group more things together than you thought because you get economies of scale a little bit better.

No matter what it is, it’s always better to have more data when it comes to online advertising, take advantage of everything that we have while we still have it, because as we can tell from other platforms, we never know when it’s going to go away. I would love to hear about any ways that you have used these data insights to optimize your accounts or something that made you rethink your account strategy. So if you have anything like that or any additional questions, as always, share them with us in the comments.

Written by Michelle Morgan