Attribution Strategy for Multi-Channel Success

When to Consider Attribution

Attribution is a big conversation, if you’re if you’re brand, youre a DTC brand, you’re growing fast. When should you start to really seriously consider proper attribution?

Its really interesting. I think attribution can add a huge amount of value in terms of improving the efficiency of a marketing mix by knowing, all right. Based on the experiments we’ve run, we can start to see that brand awareness activity is adding a huge amount of value to us Maybe a tube campaign or whatever is adding a huge amount of value to us in incremental growth.

But the problem with attribution is that a lot of people try and implement some kind of attribution methodology too soon in the growth journey and maybe you’ve only got a couple of marketing channels live, typically Meta, Google, maybe TikTok is like a classic smaller e-com business set up and trying to get too granular and really unpick the value of each channel when you’re not particularly high scale, actually can unnecessarily constrain your growth because you can’t implement an attribution methodology. Ideally, media mix modeling based or MMM based, which will truly help you understand the actual incremental revenue driven by each.

And people often get trapped by click based models, which are essentially off the shelf tools, which will really favor lower funnel channels. So going too early and trying to solve attribution, which is still essentially unsolvable, if you want a really perfect answer, it doesn’t exist. You can get close and get in on the indicative results of what might be driving most value, but I’m never going to get a clear black and white answer in marketing.

Using Attribution for Multi-Channel Success

I guess a good time for a brand to introduce attribution really is when you’ve scaled pretty aggressively on Meta. You’ve got a great Google setup, perhaps you have TikTok as well, and you have a slightly more complex channel mix. You want to assess the impact of a change you decide to make. For example, let’s say you decide to scale TikTok. You want to see what the impact of that is on your revenue.

You can actually just look at the impact on your revenue and ask, “Does scaling that channel add more growth to us as a brand?” Another potential way to approach attribution—or when to introduce it—is when you’re doing direct mail or trying out out-of-home advertising. It’s really when you have this more complex channel mix that attribution can become really valuable. You have all these different things going on simultaneously that you’re trying to unpick, and the efficiency gained can massively outweigh the cost of doing the analysis properly.

Until you have a really good size across multiple channels, spending a significant amount of money on advertising, it’s a bit of a rabbit hole that you don’t really need to go down.

I think if you’re below $10 million a year in revenue, you can learn a lot by changing one thing at a time. Typically, you’ll reach that scale with one or two channels that really work, which are usually Meta, Google, or TikTok for an e-commerce brand, for example. By changing one thing at a time and making significant changes, you’ll learn what drives value. That’s essentially what you’re building out when creating a proper media mix model.

You can ask, “When we change this, do we outperform or underperform compared to what we’d expect for that given day or month?”