Going Beyond Facebook and Google Measurement for DTC Brands

In this episode, CEO of Elevar, Brad Redding, and CEO of Rockerbox, Ron Jacobson discuss the ways that brands like Rothys, FIGS, and Rockerbox customers are tackling OTT, CTV, podcasts, and other data-driven problems. Additionally, you will learn what to consider when deciding to move marketing budgets outside of Facebook and Google.

What did we discuss in the episode Measuring Beyond Facebook & Google, In-House Data Teams, and How The Paranoid Survive?

• Rockerbox 101
• What do in-house data teams look like
• What happens beyond measuring Facebook and Google
• Expect data to be taken away
• OTT insights to action
• What can go wrong
• Do UTMs work for just FB and Google
• The future of eCommerce and measurement

If you just want to refer to some important points from the podcast, you may read the transcript below, or listen to the podcast using the widget above.

What is the ideal type of Rockerbox customer?

It's definitely evolving over time. I'd say when we first launched Rockerbox, we were very much going for mid-tier DTC brands. They were spending 5-10$ million in marketing across multiple channels. And what was interesting about that was that those brands scale and they get larger and larger. 

So what we realized was that to keep up with their growth, our product needed to evolve and needed to become good for them. So there's a large portion of our time and energy spent making sure we can go up in the market with those brands things like data warehousing, customizability, and transparency.

Our goal is to be able to have marketing measurement technology that works for brands through their lifecycle as they scale from seed to IPO. Our product has changed and varied what our customers needed during all those different parts of their lifecycle. We're very much B2C focused, very much focused on digital brands, companies like FIGS, Rothy’s, and 1-800-Flowers.

How does it look like the marketing team of DTC brands from Shopify-operated brands to big brands?

When we're working with a brand, like an early stage Shopify brand, primarily on Facebook and Google, maybe they're doing a little bit of influencer, a little bit of affiliate. It's really just a marketing team. You're talking about two or three folks and they're kind of doing a bit of everything. There's a Head of Marketing and maybe somebody focused on paid somebody focused on organic and retention. But it's a small team and they're doing a bit of everything. There's no data science organization, no analyst, and no BI framework. They're spending a lot of their time, frankly, on platforms. actually logging into Facebook and Google Analytics and spending tons of time in Excel.

As they scale, the marketing organization gets more sophisticated. It gets more verticalized inside the marketing work across different channels and kinds of channel managers specific to each different strategy. Then you maybe have one person as an analyst, they're doing a little bit of getting APIs, trying to automate a little bit of what they're doing. That's kind of like the next phase of growth.

And then finally, as they scale you're talking about data science organizations. It's really interesting when you start to expand into finance and operations because they get to a point to understand if the impact of marketing has real business implications for how they operate as an organization. 

As the organization scales, the scope and scale of its marketing analytics team increase as well. So being able to support that is difficult, but it's what we have to do at Rockerbox.

What should be the best time for a DTC brand to hire a technical analyst?

When you start to realize that you are in the growth stage, you can't just have the data in UI, you need it internally, and you need it in a place where it's going to be in a predictable schema. So you can start to build on top of it. That could be dashboards, obviously, it could be scripts, it could be hitting API's buying platforms to make adjustments and things like that. 

You get that first analyst. Sometimes they're in marketing sometimes or outside of marketing, and they're kind of some portion of their time is resourced to marketing. We've seen it happen both ways.

How should a DTC brand start building on their own data warehouses? What are the steps?

This all comes back to resources and time. A company that's raised $20 million and in scale mode is very different from a company that's growing 10 - 20% per year. 

For some companies, it's getting up and running with someone like a Funnel or Fivetran if they're more technical or Supermetrics just to get data flowing from the platforms into Google Sheets. Eventually, they're going to need a sort of a database and drop files that should be maintained manually and get some sort of engineering and process that data put into a database.

It should be sort of synchronized databases in warehouses putting data into their data warehouse, and for the most part, the vendor being reliable to synchronize it and to ensure that it's constantly staying up to date. That's the preferable way so that at the end of the day, the brands don't have to worry about getting the data into their warehouse and what I do with it.

Can you describe the Rockerbox in a more technical way? What does Rockerbox onboarding look like?  

The moment brands kind of go beyond Facebook and Google, things get complicated. That's actually changed in the last year because things can get complicated just within Facebook and Google. But at least stepping back it was, you've scaled for a certain point when Facebook Google, now you're launching OTT, CTV, direct mail, or linear, and all of a sudden, like, you spent a million bucks last month on a TV ad campaign. You log into Google Analytics and you don't see TV.

That's disconcerting for the brand. You don't really know what's happening. Maybe your organic search went up, maybe your direct traffic went up, and maybe your paid search went up. And it's hard to actually attribute what the correlation was between that TV spend that you just finished and what's actually happening. What's the business impact?

That's essentially where Rockerbox comes in. We provide you with the underlying data that can help answer all these questions. We provide an approach to measure channels that are difficult to measure or channels that don't have an explicit right way to measure a channel like TV. You literally can't click an ad on TV, right? You can't know for a fact that a person who saw a TV ad came to your site or your app and converted there.

So as a random brand, you can choose ‘Hey, I'm going to spend the next quarter hiring an engineer and a scientist and then put them on the task of figuring out how to measure TV’. 

And they're going to start that process by figuring out how do I even get the bid? How do I get the data where the TV ads are served? How do I get the session data from my site? How do I get that in one location? How do I start to think about marrying that data? How do I connect the dots? What's the approach that I want to take? That is really where why brands come to Rockerbox, rather than having to try to develop the expertise, and more importantly to maintain that in perpetuity making that pipeline, we rely on Rockerbox to do all the work for us to construct those underlying data sets that are needed to be able to run a TV analysis. 

What is the difference between Linear TV and OTT? How does the measurement actually work for Linear TV and OTT?

You have OTT (over the top) and CTV (connected TV), which are two terms that actually mean the same thing and that is your Hulu and the Roku. When I say linear TV that's like classic TV or like logging your TV or changing the channel classic cable.

Essentially, this is a really good example where these are comparable places to spend your ad dollars, with very different data sets available, and very different ways to measure it.

So for example, on the TV side, the most basic way to think about TV is just getting exposure data where and when a TV ad served - at 12.30 and Woodstock, New York, and the ad was served on the BBC -  getting that data set is actually a bit of a challenge because there's often a delay in that. Once you actually figure out how to get that data on a recurring basis, you have to figure out how to process it and connect that to what's happening on a client's website. And that's where you need full session data to start to understand who's arriving on the site.

That's where you need full session data to start to understand who's arriving on the site. This also steps back to that idea where there are different ways to measure that right. There's one way to measure that TV ad spot drives more people to my site. Was there incrementality in terms of the visitors to my website? It doesn't mean that people actually converted so that's where one sort of measurement isn't the kind of helping my top performer get to my website. There's a much more bottom of funnel question of is actually leading people to convert? And also, there's a world where people also engage with other channels in between those two different areas. 

There are more deterministic datasets available in the OTT space where you can get some form of impression-level data which are IP addresses, user agents, and device types. If you have relationships with data vendors like Rockerbox does, you can get those datasets and do a better job of measuring that channel. We've done 150 + integrations in the past four years since launching Rockerbox.

Stepping back your DTC brand and thinking of exploring OTT, CTV! You've never bought it before, you have no idea what data sets are available. You have no idea how to get that data set, you have no idea how to connect an IP address of a TV with an IP address and a desktop or a mobile phone. Like these are all questions that you just don't have expertise in and frankly, that's okay. That's not your job. But if you rely on a measurement provider, they can do that for you so that you can focus your time and skill versus focusing your time on how I actually think about getting to the point where I have data to answer my questions.

How is Rockerbox helping marketers measure performance post iOS 14.5 updates?

There's no doubt that over the past couple of years, there have been a lot of changes from Apple's iOS changes to ITP to the deprecation of third-party cookies.

We can only choose to fight it or we have to choose to just accept it to figure out how to deal with it. And fortunately for us in a weird way, we've been at this for so long that Safari got rid of third-party cookies two and a half years ago. It's been some time, since we realized that we need to expect data to be taken away from us. When we first launched Rockerbox I was always concerned about how we measured the impact of Facebook investment back in the day Facebook used to give feeds to attribution providers. And they wouldn't give me the time of day they didn't want to talk to me at all. They just couldn't care less. 

We realized internally that we need to build systems to take aggregate data from platforms like Facebook and leverage that to measure the impact of things that we might not be able to get deterministically, particularly view-based data. So how can we build models and leverage machine learning to better understand the impact of a channel even if we don't have user-level data? So not wanting Rockerbox and not having access to data that we would have wanted, forced us to become really good at leveraging aggregate datasets to measure impact on the market.

I think that what's going to have to happen is you go into per-channel basis, you figure out what the best data is that you can get for it, you figure out the best way to measure it, recognize it's imperfect, and kind of going back to that example. You have a channel like linear TV I just mentioned, that's literally not usable at all. It's just something happened in location, we can figure out measuring. All the way down to like Pinterest, which is very bottom-funnel click oriented like you can connect those dots really well. 

I mean identity resolution and figuring out how to connect different dots with different datasets with different levels of aggregation, and different levels of data is a really difficult technical challenge that Rockerbox has become really good at over the past couple of years.

How are your customers actually applying data to take action?

There's a big need for help with testing. Just generally, you've been kind of steady on a couple of channels and you're dipping your toe into something new that you've never done before. That's a big challenge, right? Because the worst thing you could possibly do is spend that money and come back internally the next week and show reporting and have nothing to show. Dipping their toes into things like there are offline channels as big an area where clients find Rockerbox helpful because all of a sudden they can actually say ”I spent money in this area. I have data I can use to show the impact of it and we can make decisions”. So that's definitely happened with CTV with linear TV, direct mail, and even podcast advertising. We've had clients that leverage Rockerbox for that. 

Another use case I'd say very much is around budgeting. How do we think of where to spend our marketing budget for the year? How do we think of setting that at the beginning of the year and changing that on a daily, weekly, monthly, or quarterly? Our datasets become highly involved in that budget process.

We also have clients that actually build their own models to gauge the impact of their marketing, but they do it based on Rockerbox data. They take Rockerbox to get raw session data, and impression-level data and they use us to figure out the joining of a direct mail and a conversion. More recently, actually, we launched a product a couple of months back that's more focused on getting those core even platform numbers into a client's data warehouse as well. So they can really start to build on top of Rockerbox and we have clients that build their own models on top of our datasets, have built their own scripts to automatically adjust bid prices and budgeting and things like that. 

So it really just depends on the use case. Lately, a huge one is obviously Facebook and iOS. How do you act you know, day to day, week-to-week basis, get the most out of channel and Facebook! So that's been a huge thing at Rockerbox. And one of our first things was again, I mentioned a couple of years rolled out synthetic events - the way that we measure Facebook for clients, and that's been hugely important, especially in the past couple of years given an eyelash interest.

What does 'testing' mean for Rockerbox customers? How do they run testing with Rockerbox?

You have your AB testing, interchannel tests, a new set of creatives tests inside Facebook, or let’s say geolocation tests - I'm going to serve in a certain area, not another area, but I want to understand what the cross channel impact is of that. And I'm not going to just rely on what the platform reports out because Facebook is doing their own data, their own conversion, their own impressions. These companies are already running tests. This is something that all brands we find are doing. They need help though, evaluating it. So Rockerbox can say “Hey, give us the input of the tests that are running and we'll help tell you the result of that test.”

The brands can be good at being creative. They're good at coming up with tests.

Do you still think that brands and marketers should be logging and separately checking the numbers in Facebook, Google ads, and any other channel TikTok, etc?

We definitely get to a point with our customers where Rockerbox numbers are their kind of guiding light, they use Rockerbox to make decisions.  Even if our numbers are actually guiding their decisions on a weekly monthly quarterly basis and they give to their CMO the Rockerbox CPA, there are still logins to platforms.

I think as a brand, you need all the data inputs, so you can triangulate and make the right decisions. If you see anything in Rockerbox that might be right, then you might want to dive in and take a look at the platform (Facebook, Google,...). And I think that's fair. And I'm really big on skepticism. I'm very skeptical, and I think any marketer should be skeptical as well. As a brand, you need to be paranoid and sense if something feels off look into it.

Should DTC marketers start thinking to pull out data from hard-to-track channels like podcasts, TV, and offline rather than just relying on channel attribution?

I would say, since iOS changes, even if you are really large on Facebook, you're gonna need measurement help. We have a lot of customers that come to us for the challenges in management budgets on Facebook. We have a lot of clients who have low success out there. 

And like there comes a point where just the amount you're spending on a given channel, it doesn't make sense to not have a third party there to help you measure it. The cost of not having it is not worth the cost of losing out on the incremental revenue that you could be getting if you've had the right signal to actually make decisions.

What are you seeing in the future of e-commerce and measurement and marketing?

I think things are only getting harder. I think measurements are gonna get more difficult. I think privacy changes are not going away. Nothing I'm saying here is overly smart, but like, these changes are here to come. I think good brands are going to have to diversify their mix more and more than ever, and they're gonna have to spend more and more money on trying to capture that first-party data so they can re-engage the customers in pseudo free channels, email, SMS, and things like that also, to connect dots.

Marketers are becoming more data-oriented. I think all good brands are going to need underlying datasets that enable them to be data-driven and I think they will be working with technologies that enable them to better do that, like conversions flowing properly between different platforms, more consistent datasets, it's just gonna be critical. So like any brand, the sooner you can start to invest in getting your marketing data infrastructure in place so that you can actually make decisions that are based on data but better and it's going to become more and more important in years to come.

On the other hand, I think companies just need to be really adaptable. The methods that they use to measure today might not be the same metric that the method used to measure in two years.

The blog is updated on April, 22 2022