The Daily: Media measurement in a video first world

On today's podcast episode, we discuss the most important measurement considerations for marketers, how brands are able to stand out on TikTok, and how to best determine the true impact of media buys. Tune in to the discussion with Senior Director of Podcasts and host Marcus Johnson, Vice President and Principal Analyst Jasmine Enberg, and Aashish Takkala, Product Marketing, Media Measurement at TikTok.

Subscribe to the “Behind the Numbers” podcast on Apple Podcasts, Spotify, Pandora, Stitcher, YouTube, Podbean or wherever you listen to podcasts. Follow us on Instagram

Episode Transcript:

Marcus Johnson (00:00):

This episode is brought to you by TikTok for Business. Vibe Check. Are you looking to slay your next campaign on TikTok? Obviously. Get ahead by measuring ahead to improve your campaign's performance on TikTok. Better measurement is the key to an improved ROI. Learn more from TikTok for business at TikTok.com/business.

Aashish Takkala (00:22):

The trend that we're seeing with video is not slowing down anytime soon. But what's also interesting is on the advertiser's side, the advertisers are no exception to this trend with short form video or mobile video. Most advertisers have embraced video, short form video, for storytelling about their brand, their products. And that's what you see in our platform today.

Marcus Johnson (00:49):

Hey, gang. It's Thursday, October 24th. Aashish, Jasmine, and listeners, welcome to Behind Numbers Daily, a eMarketer podcast made possible by TikTok. I'm Marcus. Today I'm joined by two folks. Let's meet them. Why not? We start with our vice president and principal analyst covering all things social media. Based in California, it's Jasmine Enberg.

Jasmine Enberg (01:10):

Hey, Marcus. Hey, everyone.

Marcus Johnson (01:12):

And also joining us is the product marketing manager at TikTok. Based on the other coast, New York City, it's Aashish Takkala.

Aashish Takkala (01:19):

Hi, everyone. Nice to meet you, Marcus, Jasmine.

Marcus Johnson (01:20):

Hey, fella. Welcome to the show. All right, today we're talking about media measurements in a video first world. But before that, we'll start with our speed intro, get to know our guests a little better. 60 seconds on the clock from Victoria, who edits the show. Rapid fire here, let's get to know the folks. Question one. Aashish, you'll start. Jasmine, going to go second for each of these. You are based in New York City, but where are you from?

Aashish Takkala (01:45):

I'm originally from India.

Marcus Johnson (01:47):

Whereabouts?

Aashish Takkala (01:48):

Hyderabad, India. South.

Marcus Johnson (01:49):

Ah, okay. Very nice. And Jasmine, based in California, but from?

Jasmine Enberg (01:54):

I was born just outside of Helsinki, Finland. But I kind of grew up all over, so I don't really know where I'm from. I'm from LA these days.

Marcus Johnson (02:02):

She's from everywhere. But she's from Finland because she makes me bring a Finnish fact every time she's on. I do not have one for you this-

Jasmine Enberg (02:02):

You have done that in a long time.

Marcus Johnson (02:09):

I have not. I ran out. No, I just got lazy. What do you do, in a sentence, Aashish?

Aashish Takkala (02:14):

I am on the product marketing team at TikTok. I lead our good market efforts for our attribution product.

Marcus Johnson (02:20):

And Jasmine, when you're not hassling me for Finnish facts, what are you up to?

Jasmine Enberg (02:25):

I lead our coverage of social media and the creator economy here at eMarketer.

Marcus Johnson (02:29):

Very good. Aashish, your morning drink?

Aashish Takkala (02:32):

Yeah, it's coffee on the weekends. Weekdays, I have a herbal tea that I go with.

Marcus Johnson (02:36):

How are we taking that coffee?

Aashish Takkala (02:39):

Black.

Marcus Johnson (02:40):

Madness. How do you do it? Jasmine, what about you?

Jasmine Enberg (02:44):

Iced coffee with oat milk every day.

Marcus Johnson (02:47):

Oh, fancy. And then Aashish, final question. Best movie that you recently watched from your childhood. Something you watched during your childhood, which you love, and you recently rewatched it.

Aashish Takkala (02:59):

Yeah, I couldn't think of a childhood movie, but I watched this movie called Migration over the weekend, with my toddler. It's a story about birds. It's a fabulous movie. You should check it out if you haven't watched it yet.

Marcus Johnson (03:08):

Okay. Good recommendation. Jasmine?

Jasmine Enberg (03:11):

So I didn't watch the full movie, but I played the opening song of Ghostbusters to my three-year-old nephew, and he loved it. And it reminded me how much I love that movie.

Marcus Johnson (03:22):

How did this come about?

Jasmine Enberg (03:24):

Well, it's almost Halloween and we were talking about zombies and monsters and ghosts, and it just popped into my head.

Marcus Johnson (03:32):

Okay, fair enough.

Jasmine Enberg (03:32):

As you do. Yeah.

Marcus Johnson (03:34):

Good choices. All right, folks. They are our two guests. Let's move to our fact of the day before we get into the episode. Today's fact: the Moon is slowly moving away from Earth. Did anyone else know this? I feel like I should have been warned by someone. If you guys knew, I'm going to be furious.

Jasmine Enberg (03:54):

I did not know this. But something you did not know about me is that I am obsessed with the moon

Marcus Johnson (03:54):

Astronaut? Oh.

Jasmine Enberg (03:54):

No.

Marcus Johnson (03:54):

Oh, obsessed is a strong word. Okay.

Jasmine Enberg (04:04):

It is one of the main topics of conversation-

Marcus Johnson (04:06):

Let's unpack that.

Jasmine Enberg (04:06):

... between my friends and I. And we call-

Marcus Johnson (04:06):

Main topics?

Jasmine Enberg (04:10):

We call ourselves moonologists, which is not a thing, but it should be. But this is a fact I did not know, so I'm going to be very happy to bring this to the group later.

Marcus Johnson (04:18):

As a moonologist, you should have known this. Aashish, did you know this was happening?

Aashish Takkala (04:23):

I had no idea. And I did not know about moonologists either, so two facts for me.

Marcus Johnson (04:27):

That's definitely not a thing, so don't worry about it. So it's getting about an inch further away every year according to NASA. That means 30 Earth sized planets could fit in between the Earth and the Moon. The Moon is 10 times further away today than when it was formed billions of years ago. So it used to be about 15 to 20,000 miles away, and today it's closer to 240,000 miles away. So 10 times further away.

(05:00):

You're probably wondering what happens when the Moon moves away from the Earth. It's not great. Science website Live Science explains that it means much smaller tides in the oceans, so less good surfing, I assume, due to reduced gravitational pull, impacting coastal ecosystems and potentially causing climate disruptions. Great. Additionally, Earth's rotation would slow down leading to longer days. And the stability of Earth's tilt could be affected, potentially causing more extreme seasonal variations. All good things clearly. Anyway, let's get on with today's episode before it gets even darker.

(05:38):

Today, as I mentioned, we're talking media measurements in a video first world. All right folks, so let's get into it. First of all, Aashish, we wanted to start the conversation by looking back over the last 10 or so years, media habits and how they've changed, TV being a very dominant form of media in terms of consumer attention and ad dollars as well.

(06:04):

I went back and looked at some of our numbers to set the table. In terms of attention, TV, 2014, 94% of Americans watched traditional television. Today, it's 68%. It's nearly 30 points less. In terms of time spent watching TV, that's fallen from close to 4.5 hours to 3.5 hours. If you look at the ad dollars, TV ad spend has gone from 40%, 4-0% of advertisers' budgets, to 15% in the last 10 years.

(06:35):

And so TV has become less significant, and this isn't a clear, dominant medium anymore. It's become very fragmented, as folks know, thanks, in part, to the rise of platforms like the one you work at, TikTok. So Aashish, what's interested you the most about how time spent has changed over the last 10 or so years?

Aashish Takkala (06:53):

Yeah. Thanks, Marcus. I think what's been really interesting over the past few years, I would say moreso the last five years, is how much video content is created online. So on one hand, the demand for video is on the rise, and that's clear from the rise of platforms like TikTok.

(07:12):

But on the other side, we also have the effort it takes to create a video coming down every year. You have an iPhone in your hand and you can create really high quality video content today. And with generative AI tools coming up, I think that ease is only going to get simpler and simpler for people to create more video content. So I think the trend that we're seeing with video is not slowing down anytime soon.

(07:39):

But what's also interesting is on the advertiser's side, the advertisers are no exception to this trend with short form video or mobile video. Most advertisers have embraced video, short form video, for storytelling about their brand, their products. And that's what you see in our platform today.

Marcus Johnson (07:58):

Yeah, the video one's really interesting because we have numbers that show, if you look five years ago, video accounted for 1/3 of social media content that was being consumed. Today, it's 2/3. And those last five years coincide quite nicely with the rise of TikTok and how much of a significant player it's become in recent history. Jasmine, when you think about time spent and how it's changed, what jumps out to you?

Jasmine Enberg (08:26):

Well, I wanted to second Aashish's points. I think one of the biggest things, especially when you're looking at social media, is the shift toward video. And you, Marcus, already pulled some of our stats on that. But another one that is really shocking, although it shouldn't be altogether that surprising, is what's happening with ad spending as well. So next year, we are expecting social video ad spending to surpass linear TV ad spending here in the US. And so that is a huge milestone, and just goes to show not only that consumer attention has shifted toward video and social platforms, but advertiser attention and spending has now as well.

Marcus Johnson (09:06):

So consumer behavior changing, advertiser spending habits changing, measurement not quite kept up. Aashish, can you walk us through where we are in terms of measurement today?

Aashish Takkala (09:18):

Yeah. When you think of measurement, it's a bit of a spectrum, with some advertisers being sophisticated and using the tools that we would like to see them using. More scientific, more sophisticated. But on the other end of that spectrum is advertisers who are still using traditional measurement. And what I really mean by traditional, again, it sounds like these methods have been around for decades, but that's not really the case. Click-based measurement has only been around for about 15, 20 years now. And that is what a lot of advertisers still use today, despite video establishing itself as a dominant channel.

(09:55):

And if you look back at measurement before the internet age, people use media mix models. You look at Nielsen, the TRPs that they generate and then put it into a model and try to understand how each channel is driving revenue for their business. Now with internet, that shifted slightly to click-based measurement. And now we are at this inflection point again, where video is the dominant channel. So marketers have to think about how do they change or evolve from click-based measurement to using a better way to measure their video.

Marcus Johnson (10:28):

And you mentioned there's quite a range of people from one end of the spectrum to the other in terms of how advanced their measurement is or measurement strategies are. What to you are the main ways that you can get people moving from the less advanced end of that spectrum to the more advanced end when it comes to measurement?

Aashish Takkala (10:50):

So talking about products that we have to offer advertisers, if you run performance ads on TikTok today, you have the ability to check what your performance reporting looks like through a tool called Attribution Analytics. What it helps you see is what is the average time it takes for people to convert?

(11:07):

With click-based measurement, one of the hypotheses is that people see the ad, they click on it, and they end up buying the product. But with the insights that we're showing on our platform today, you can actually see how long it takes for someone to convert after they've seen an ad.

(11:21):

And aggregated data shows us that in the US, for e-commerce, the time to conversion is about 10 days, so that's how long it takes for people to convert. Which means everything that click-based measurement is telling you, it is missing out on a big chunk of conversions that we are seeing on our end. And this data is available for advertisers to see for your own ad accounts as well. So that's one way to get started on that journey.

(11:45):

And once you understand the magnitude of the conversions that you're missing out on, you can start thinking about Conversion Lift Studies, which is generally available as of last week, we announced this at Ad Week. It's the most scientific way, we call it the gold standard of measurement. So it's a test group, a control group. You show ads to the test group, don't show it to the control group. You see the difference between the two. So there are different ways for you to do it. Conversion Lift is one of the ways.

(12:13):

You can also lean on third-party partners. We have a lot of partnerships that we've announced in the past few months now. So third-party partners, Conversion Lift Studies, Attribution Analytics. We have a few different ways for you to get started on that journey.

Jasmine Enberg (12:26):

So that was something that I have been talking a lot about recently as well, about how the path to purchase, especially from social media, is not always linear. And it can take somebody a long time to actually go from discovering an item to eventually purchasing it. And that purchase doesn't always happen on the exact same place where they discovered it. And so it's really fascinating to hear the evolution and what's still needed in order to get advertisers to catch up with that.

Marcus Johnson (12:57):

Jasmine, you've mentioned, before the recording, one of the big misconceptions about platforms like TikTok is that it is a place for a big, home run brand awareness players. Aashish, how are you trying to change that perception?

Aashish Takkala (13:15):

Yeah, great question. So I would start by saying that a big reason for that misconception is because of how people measure TikTok as a media channel. So if you're running performance campaigns today, and you're using video as your go-to format, and you're looking at click-based measurement, I don't expect a lot of people to see these ads and click on the videos right away to go and purchase a product. A click-based measurement would always downplay the impact of a channel like TikTok.

(13:43):

With that in mind, what we have invested in is a whole suite of measurement products and partnerships. So on the measurement product side, we have our first party tools. So Attribution Analytics and conversion lift are two solutions.

(13:58):

And then on the third party side, we've announced an exciting partnership with Northbeam last week as well. It's a closed beta where we are sharing our impression data with Northbeam, which means their MTA model not only takes into account the clicks that happen on our platform, but also includes the impressions or the views that are happening on our platform, and feeding it into their MTA model.

(14:21):

So it's exciting that third parties are leaning into view-based measurement, realizing the value of video as a format, and including that into their overall modeling methodology.

Marcus Johnson (14:34):

Could you give us an example of a brand, of someone who's been able to stand out on TikTok and show some very measurable results?

Aashish Takkala (14:44):

With the holiday season coming up, I have two examples that come to mind. The first is a large e-commerce player. They have done very well on our platform. They have leaned into native creative, so creative that lands very well on TikTok. They've continuously optimized their ad account, meaning checking with strategies that are working and then scaling those strategies.

(15:07):

But the one lingering question that they always had was is TikTok actually incremental? So is TikTok actually driving conversions for me or are they showing impressions to everyone hoping that those would lead to conversions as well?

(15:20):

So what they leaned into to test that out is Conversion Lift Studies again. And what they saw was TikTok was not only incremental, it actually beat their expectations in terms of how much additional incremental revenue TikTok was driving. So based off of those results, they increased their media spend during holiday time, during holiday season last year, and they saw a 2.5 times increased ROAS on their ad account. So the return on ad spend was 2.5 times higher than what they had seen the year before. So a big jump year-over-year during their holiday season.

Marcus Johnson (15:55):

Yeah. And the second example?

Aashish Takkala (15:58):

Yeah, the second is a DTC brand. Now, if you are a smaller advertiser, a DTC company that has a Shopify store for example, this example might be closer to a pure business model. This is a jewelry brand that is strictly DTC. They're predominantly in the US. And what they have leaned into was post-purchase surveys, which is another third party solution that we offer. It's simple integration. It helps you serve a survey to the customer at the end of their purchase. So after they hit purchase, a survey loads and asks them a few questions.

(16:36):

So the questions that they structure were, where did you hear about us? And TikTok was one of the options. Why did you buy this product? Is it for yourself or for someone else? And what they learned from this information was that TikTok drives about 1/3 of their new customers. So with a smaller brand, especially in the DTC space, you really care about new customer acquisition. And in this example, they saw 1/3 of their total revenue for new customers coming from TikTok.

(17:07):

And the second question was actually very interesting as well. They found out that about 40% of their purchases were gifts that people were buying for someone else. So it was really cool for them to take that insight and then use that into their future campaign planning. So their copy, their creative, everything could use inspiration from a stat like that as they think about holiday season.

Marcus Johnson (17:30):

All right. Some great examples there, and examples of what does work. We'll flip that on its head. Can you tell us what doesn't work on the platform?

Aashish Takkala (17:40):

Yeah. So in terms of what doesn't work-

Marcus Johnson (17:41):

"Nothing," he says. Moving on.

Aashish Takkala (17:44):

Yeah. So the biggest one is when you're thinking measurement, you need to align your measurement strategy with how users are behaving online. And if you're not aligning the two, that's when you wouldn't see results that really make sense to your business. Now, you might think that they are making sense because that's what you've seen for the longest time, but that's where you're missing out on a lot of value that platforms like TikTok, and the video format, is creating for your business. So from a measurement perspective, I would say looking away from last click is where I would start.

Jasmine Enberg (18:16):

And from my perspective, because I also get asked this question a lot by advertisers and marketers, and one of the things that I always say, on TikTok and other platforms, it's really crucial to match the kind of content and the language that is being spoken on those platforms, and to really make the content feel native to them. And I think, Aashish, you brought that up already, talking about the native creative that was working. And so where I see some brands not perform as well is when they're trying to or they're not quite matching the medium that they are on in that moment.

Marcus Johnson (18:54):

Last question folks. Aashish, I'll start with you. If you could give marketers one piece of advice for how to determine the true impacts of their media buys, what would you tell them?

Aashish Takkala (19:04):

Measurement is a journey. So it's not a simple recommendation, I would say. You have to get started on that journey for you to start seeing results. So the first thing I would do is check your data within the Attribution Analytics tool. So this is available within TikTok Ads Manager. You can pull any campaign that you've run and see what this reporting looks like. So I've talked about time to conversion. You can also see touch points to conversion. So what is the path that led to purchase and how many impressions were involved.

(19:32):

So that data is really powerful for you to start convincing maybe folks on your end, if you have leadership that is leading to last click, you can use some of this data as proof points for them to start thinking about evolving to better measurement. And from that point on, it is picking a measurement solution that works. Like I said, Conversion Lift Studies are great. Third party partnerships are also available for you to tap into.

(19:56):

And then the last step is calibrating your old world with the new world. Once you start using a new source of measurement, you have your old data from a click-based model, you have your new data from a post-purchase survey or a Conversion Lift Study, so how do you bridge the two worlds? So that's where I would spend some time thinking about the strategy there.

(20:18):

To give it a couple more... Maybe I'll spend 30 seconds on this. Let's say the Conversion Lift Study said there is 50 conversions in the last month on TikTok, and your click-based model said there is 25 conversions. So what you're seeing there is that ratio between the incremental conversions, which is 50, versus click-based conversions, 25. So that 2:1 ratio is what I would use for calibrating my internal attribution on an ongoing basis. So for this month's conversion, I would just apply that multiplier to say, "Okay, this is what we observed initially, but based off of our learnings from better measurement, this is what we're bumping it up to." Because that is a more scientific way of measuring.

Marcus Johnson (20:59):

Terrific. Great way to end the episode. Thank you so, so much for that. Thank you so much to both of my guests for hanging out with me today. Thank you first to Aashish for joining us.

Aashish Takkala (21:09):

Glad to be here. I thank you for having me.

Marcus Johnson (21:10):

Of course. Thank you, of course, to Jasmine.

Jasmine Enberg (21:13):

Thank you, Marcus. Bring a Finland fact next time.

Marcus Johnson (21:15):

Okay, fine. I'll look. Thank you to Victoria, who edits the show, Stuart runs the team, and Sophie who does our social media. And thanks to everyone for listening in to today's episode of Behind Numbers Daily, an eMarketer podcast made possible by TikTok. We hope to see you tomorrow for our Weekly Listen show, an eMarketer video podcast.