When Your Digital Advertising Hits a Ceiling — And What to Do About It
by Andrew Krebs-Smith | February 17
Paid online advertising is one of the most scalable channels that marketers have at our disposal, but the methods and tools that get you to a certain point can’t always get you to the next level. Just about anyone can figure out how to effectively spend $200 a day on an ad channel as long as they understand how the channel works. But if you want to grow to $1,000 a day, then you have work differently. You have to develop systems to track a much greater amount of data, generate creative much more quickly, and so forth. After that, the next tier might be $2,500 to $5,000 a day, and that requires strong deep-funnel analytics capabilities in order to break through that digital advertising ceiling.
So, to scale successfully, consider managing your budget as if it were a portfolio of individual investments. There is a big difference between the way a $20 billion hedge fund portfolio is managed versus someone on eTrade trying to optimize returns in a retirement account. If you don’t know the recipe for building the $20 billion portfolio, you’re never going to get there using eTrade strategies, even if you’re making brilliant moves in the market. The tool—as good as it is for what it does—simply does not give you the control you need for that next level.
What, then, is the recipe for successful, profitable, skillful spending?
- First, you need the right data infrastructure—meaning you have to collect the right data in the right way—in a way that gives you actionable information.
- Next, you need to know something about experiment design and how to design tests that will matter. Anyone can test anything, but what is really going to save you time and money is testing the right thing at the right time.
- Then you have to actually implement the experiments you’ve designed.
- In addition to those three capabilities, you need to be able to analyze what the ongoing data you’re collecting from the experiments is saying, which requires data science experience and capabilities within the organization.
- Finally, you need to be able to pivot in an agile way — which means you have to know what to do when the data looks a certain way and have the organizational will to do it.
Conceptually, that’s the recipe: data infrastructure, experiment design and execution, ongoing analysis, and very fast and agile re-implementation based on that analysis. It may sound simple, but it’s not. We’ve learned this managing campaigns across platforms for a whole host of companies—spending far more in aggregate that most companies would even aspire to. If you’re interested in knowing what industries and verticals we have experience with, reach out!