The scoop: We spoke with Jorge Ruiz, global head of marketing science at TikTok, to discuss “the sweet spot” in ad frequency, how TikTok is developing its measurement and attribution tools in an iOS 14.5 world, and the state of connected TV (CTV) measurement.
The interview was conducted in conjunction with today’s release of TikTok and analytics company DIRT’s “US Brand Building Study 2022” on ad engagement and attention, and how to drive brand affinity on TikTok.
Insider Intelligence: Your study showed that creative tailor-made for TikTok was watched for 3 seconds longer and was more emotionally engaging than ads repurposed from other platforms.*
How do you balance that finding with the launch of some of your new ad formats, like catalog listings ads, that are more so-called “traditional” digital advertising?
Jorge Ruiz: I think whether it's brand or performance campaigns, [it’s important for advertisers] to have a process to say, "Have the ads and the messages that you're putting on TikTok been customized in any way to speak or resonate with consumers?" Some of [the customization] could be visual, some of it could be in the copy, and some of it could be a basic context.
We do also see an opportunity to test ads that have the ability to be repurposed but have very clear messages that speak to the TikTok community. We're betting big on our commerce products.
II: Customization is what works for brands on TikTok. Let’s turn that around. What are some of the mistakes that you see brands making when they're planning their first TikTok campaign, and how are you working with them to improve those initial efforts?
JR: I'll give you a very good, obvious one: Effective frequency. The sweet spot that we found is around two to three exposures per week, but, of course, that's going to vary by campaign and by vertical.
One of the first data science projects I asked my team to do, once we had enough scale with brand lift, was to go through every single one of those brand lift studies that we had. What we saw is that the campaigns that had the most brand lift were the ones that had around two to three exposures per week. Meanwhile, one of the most common attributes among underperformers in the study was low frequency.
II: But one reason that advertisers may be posting less frequently is that getting the creative right on TikTok can be really hard. How do you balance the need to create tailor-made ad content with the need to post ads more frequently?
JR: It depends on the complexity of the message. If you're trying something brand new that consumers don't understand much about, you need to overinvest in frequency to make sure that your message comes across. If it's more minor messaging or something to help you build a more favorable selling environment, you can probably play around with the frequency.
II: So which would you say is more important, frequency or creative?
JR: For some brands, creative is going to be most important; for others it’s going to be frequency, or having multiple placements. We’re doing this foundational research to help provide advertisers with best practices, but those that are going to perform really well are the ones that build a test and learn discipline. That's also where our agency partners are critical, because they can guide advertisers.
II: TikTok is working to build measurement tools against a very different digital ad backdrop than what existed when many of the other social platforms developed their ad businesses. From a measurement and attribution perspective, how are you navigating the privacy challenges as you develop your toolset?
JR: If you’re in my shoes, you have two fundamental things to worry about: One, you have to meet marketers where they are today. Two, you also have to build for the future. We grew up during the pandemic and also as many technological changes happened, so we've been lucky to, in some ways, start from scratch. But we're also having to race to very quickly build solutions that work today and that are privacy-first or privacy-enhancing to be compatible in the long term.
One of the things that we've put probably the most effort in lately from the measurement side is showing that TikTok can drive offline sales, through our partnership with Nielsen NCS, for example. Each new solution will take its time, but we want to make sure that as we mature and as the solutions come online, we work toward driving real and reproducible business outcomes.
II: Speaking of long-term goals and the future for TikTok, you are currently exploring other ad surfaces like CTV. From a measurement perspective, does some content perform better than others on these nonmobile surfaces?
JR: It may be too early to tell. From where I sit, I'm looking at our mobile experiences, because that's where we've been able to monetize with ads, and that's where we have that scale. So it's early days for us on anything connected to TV on those new surfaces, but I know that our ad teams are pretty excited and bullish on growing that practice.
What I look at more broadly is that most of the research on cross media says that more screens and the more that you can diversify as a marketer is a good best practice. Even on the [media] mix modeling work that we've been doing, we find that multiple placements are better together. We find longer campaigns are better together. Those are best practices for brands and things that we're set on corroborating.
I'm hopeful that maybe in a year's time, we can check in and say, "How's connected TV doing? How are we measuring that?"
* TikTok and DIRT’s study was based on an in-lab test, which measured responses to a mock feed during and after exposure to ads on five social and mobile video platforms from 300 US consumers between January 24 and February 1, 2022. At-home testing and surveys were conducted with 5,208 US consumers during the four weeks after initial exposure, from February 9 to March 16, 2022.