Jorge Ruiz, TikTok’s Global Head of Marketing Science, explains the mobile app’s approach to measuring brand campaigns.
What were the key marketing science priorities that you wanted to establish at TikTok?
We’ve been trying to help to answer two basic and difficult questions: do the ads work, and how do they work? Two-and-a-half years ago, the business had just started. We had to be really diligent and listen to our clients, but also start from a place of fundamentals. It’s very important because, to me, measurement and research are about building trust, and also showing reproducibility.
Early on, TikTok started with organic experiences [like] hashtag challenges. That evolved with TopView premium placements. All these things told us that we have both an organic and a paid advertising story, so where better to start in the measurement space than with media mix models (MMM)? For years, MMM has been a well-known methodology to understand paid, owned and earned media, so that’s what we’ve been working on with Nielsen across the world.
We’ve also been working with partners such as NCS in North America to look at offline sales at a campaign level. The latest number is that 34 out of the 37 campaigns with tests that are statistically significant showed an average return on ad spend (ROAS) of 2.3.
Some might consider it surprising that TikTok is focusing on MMM over forms of attribution, for example. Why did you choose this approach?
We are going to be focused on as many different measurement approaches that meet where clients are. We have a wide spectrum. The real way to frame the question, I think, is why MMM is coming back. The shared common interest that all of us in the industry have – and it’s particularly amplified for TikTok – is that in the land of privacy-first measurement where there are fewer signals and where we have clients using experiences across paid and organic, MMM, particularly for packaged goods, is a time-tested, well-proven approach. It’s a place where we want to show certain clients that we have that muscle and that discipline, and that we’re building a world class, marketing science organisation.
Beyond the campaign measurement side of things, what else is your team focused on?
The other thing that we’ve been very passionate about investing in is our own learnings. When it comes to the ad experience, what is the consumer appeal, what is the relevance, also what is the ad quality?
For example, the topic of attention has come up a lot. We spent the better part of the year doing a plethora of research. A year and a half ago, when my team heard from marketing about their ‘Don’t make ads, make TikToks’ campaign, we asked, ‘What does that mean?’ Early on, we saw in a lot of research that consumers really gravitated towards the authenticity of the ad experiences, because of the relevance they were seeing in the content. They saw that brands were really participating in that tone. The ones that broke through – the ones that raised that amount of memorability and attention – were the ones that were very much in that tone of being authentic. We’ve been doing a lot of research with leading market research firms on this question: what constitutes different ways of understanding attention? Then we can get into the more interesting questions with clients around what kind of creative they should be building.
As you say, there is a lot of talk about attention. From a TikTok perspective, where do you feel the measurement debate is headed?
It’s still very early days. It’s less about the standards or about solidifying some final thing. We’re going continue to iterate, continue to build. At the end of the day, it’s less about which is the best methodology versus what happened. We help clients have that test-and-learn mindset to drive behaviour change. I’ve met clients where they want us to work with [their own] measurement systems, where they have a phenomenal process. They’ll take something as simple as their own attention approach, but they have such a great way of getting it across to brand teams to activate their tests. They will build an organisational process behind it. That, to me, is the fascinating part. Measurement can’t exist on its own and it can’t always be about the most superior methodology. It has to be about how a business adopts it and makes decisions off of it.
Brands are using TikTok at different stages of the purchase journey, from discovery to conversion. What impact does that have how you advise marketers to measure their campaigns?
We want to make sure that the client is learning the basics first. If you build with the TikTok style, it has the best chance to perform. It’s still super important that clients understand how they get on board and master the basics before they start advancing. Great measurement without great planning will not guarantee a great campaign. If you spend the time to understand the basics, then let’s get to the next stage, which is what’s going to be the business outcome that you want to achieve. Is it brand awareness? Fine, then let’s go do it with a brand lift study. Do they care to optimise against effective frequency, or is there a brand metric that they want to hit? It’s less about the tool and the mechanic, versus making sure that we have the right outcomes that we’re building towards.
Some agencies are finding it tough to work out how best to manage frequency on TikTok. What best practice are you seeing?
We’ve been doing a lot of work around effective frequency. Ever since I first read a paper from the 1970s, it’s something in advertising you always want to investigate. We’ve done some quantitative research around that. For example, what do you do when you have many, many brand lift studies? You go into a meta-analysis and you try to understand some empirical evidence. What do we get from those effective frequency curves? From some of our brand lift campaigns we see that one guideline is an effective frequency of two to three per week. Again, it’s going be case by case. If you have amazing creative, it may differ. But we do see that, generally speaking, frequency matters [on TikTok], as it should.
How about wearout? People watch content on TikTok to be entertained, so does wearout happen faster than, for instance, on TV?
I don’t know yet. What I do know is that, in a world where all ads are skippable, we are very happy to know that all attention is earned. As long as there’s authentic content that’s built to the platform, I’m confident that brands will do well.
What are your key priorities for the next year?
It’s the same fundamental questions – we will continue to keep trying to answer the question of whether the ads work. The second question is how do the ads work? To get there, the third thing that’s super important to me is talent development. I want this team to lead in the future.