Unlike a weak phone signal causing only a grainy sound, in growth marketing this can mean the difference between a successful program or massive cash flow bleeding. As we move towards an increasingly privacy-centric world, it is even more necessary for businesses to set the signal early on.
So what exactly is the “signal” in growth marketing? It can have many different meanings, but from a holistic perspective, it’s the event data in our arsenal that helps guide decisions. When it comes to paid acquisition, it is essential to optimize and deliver the correct event data to the paid channels. This is how targeting and auction algorithms have the richest data to exploit.
I’ve seen startups spend thousands of dollars inefficiently due to the lack of optimal signal in their paid acquisition campaigns. I have also spent millions in companies like Postmates to fine tune our signal to the best possible condition. I wish every startup could avoid the painful mistake of not having it set up correctly, instead of making the most of every big advertising dollar.
At first, it may seem obvious to optimize towards a north star metric such as a buy. If the expenses are very minimal, it may mean that the conversion volume will be low from campaign to campaign. On the other hand, if the optimization event is set to a top of the funnel event such as a landing page view, the signal strength may be very low. The reason the strength may be low is because of returning a low intention event as successful to pay channels. By marking a landing page as successful, paid channels like Facebook will continue to find similar users to those low-converting users.
Take the example of a health and wellness application whose objective is to generate memberships to their coaching program. They are just starting to explore paid acquisition and are spending $ 5,000 per week on Facebook. Below is an overview of their events in the funnel, weekly volume, and cost per event:
In the example above, we can see that there is a significant volume of landing page views. As we descend the simplified flow, there is less volume when users exit the funnel. Almost everyone’s instinct would be to optimize either the landing page view, because there is so much data, or the subscription event, because it’s the strongest. I would say (after extensive testing on multiple ad accounts) that none of these events would be the right choice. With landing page views as an optimization event, users have extremely low propensity since the conversion rate from landing page view to subscription is 0.61%.
The correct event to optimize here would be either registering or starting the trial, as they have sufficient volume and are strong signals of a user converting to the north star metric (subscription) . Looking at the conversion rate between signup and subscribe, this is a much healthier 10.21%, compared to 0.61% from the landing page.
I am always a big supporter of testing all events because there can certainly be some big surprises in what can work best for you. When testing events, make sure there is a stat-sig baseline followed for making decisions. Also, I think it’s a good practice to test events regularly early on, as conversion rates can change as other channel variables are adjusted.
In some cases, the news put in place is not optimal for paid acquisition campaigns. I’ve seen this happen frequently with startups that have long windows of time between conversion events. Take a startup like Thumbtack, which provides a marketplace of vendors that can help with home repairs. Once a person signs up for their app, the user can apply but not hire anyone until a few weeks later. In this case, making flow adjustments could potentially improve the signal and the data you collect from users.
One solution Thumbtack could implement to garner a stronger signal would be to add another step between requesting and hiring someone. This could potentially be a survey with propensity control questions that could ask when the user needs help or how big their project is between 1 and 10.
After accumulating the data, if there is a strong correlation between the survey responses and someone starting their project, we can begin to explore the optimization of that event.
In the example above, we see that users who answered “9” have a conversion probability of 7.66%. Therefore, this should be the event that we are optimizing for. Artificially adding steps that qualify users in a longer flow can help steer optimization targeting in the right direction.
Say you have the most ideal stream that captures large volumes of event signal without too much delay for your optimization event. It is still far from perfect. There are a myriad of solutions that can be implemented to further improve the signal.
For Facebook in particular, there are connections like CAPI which can be integrated to transmit data more accurately. CAPI is a method of returning web events from server to server rather than relying on cookies and the Facebook pixel. This helps mitigate browsers that block cookies or users who can delete their web history. This is just one example. I’m not going to go through all the channels, but each has its own solution to help improve the event signal returned to it.
Signal iOS 14
It wouldn’t be a column written in 2021 without mention of iOS 14 and strategies that can be leveraged for this growing user segment. I wrote another post on iOS-14 specific tactics, but I’ll cover it here in general. If the North Star metric event (i.e. purchase) can be triggered within 24 hours of the app’s initial launch, then great.
This would bring in large volumes of high-intention data that would not be at the mercy of the SKAD 24-hour event timer. For most businesses this may seem like a high goal, so the goal should be to have an event triggered within 24 hours which is an indicator of a high probability that a person will fulfill your north star metric. Think about the events that occur in the flow that ultimately lead someone to buy. Maybe someone adding a payment method happens within 24 hours and historically has a 90% conversion rate compared to a buyer. An “add payment information” event would be a great conversion event to use in this case. The landscape of iOS 14 is constantly changing, but this should apply in the immediate future.
Incrementing and staying one step ahead
As a general rule, incrementality checks should be constantly carried out in growth marketing. This gives important reading on whether ad dollars are attracting users who wouldn’t have converted if they hadn’t seen an ad.
When comparing optimization events, this rule still applies. Make sure cost per action isn’t the only metric used as a measure of success, but rather use the incremental increase in each conversion event as the ultimate KPI. In this article, I detail how to run Lean incrementality tests without swarms of data scientists.
So how do you stay ahead of the curve and keep driving your growth marketing campaigns forward? First, constantly question the events you are optimizing for. And second, don’t neglect any effort.
If you use the same optimization event forever, it won’t do the performance potential of your campaign any favors. By experimenting with flow changes and running tests on new events, you’ll get a head start. When iterating over the stream, think about user behavior and events from the user’s perspective. What flow events, if added, would correlate with a high propensity conversion segment?