WTF Happened To Funnels?

Top-of-funnel paid media tactics and using traffic and engagement optimizations to fill your funnel.

Hi everyone! Welcome back to another edition of The Marketer’s Playbook. Today I’d like to get into something a little different. This is a study of marketing concepts and an analysis of how this works in action. Today, we look at the idea of running Top of Funnel campaigns on Facebook without optimizing for conversions 😱

To start, I need to credit Jake with kickstarting this conversation, and Akvile, who inspired him (you should definitely follow them both). In Andrew Foxwell’s private Slack group (we will get into that later), there was a large thread surrounding this topic, making me want to dig deeper into it. It started with Jake asking:

This is a really interesting thought. Before digital marketers got really comfortable with Facebook’s ability to optimize for conversions in prospecting, using higher-funnel goals such as traffic was much more popular. With many agencies/freelancers working to ensure direct returns for clients, it makes sense that more nebulous optimizations were left in the past - at least for a while. Not only that but there are several reasons for businesses to have steered clear:

• For low AOV products, especially if you have a low LTV as well, you don’t have the capital to spend on acquisition hoping to make that up later - therefore, you need to get conversions very early into the funnel

• Many SMBs don’t have the capital to put up for a full-funnel play. The amount of time between paying for the acquisition and the revenue coming through the site is very important. The longer that amount of time is, the harder it is for small businesses without a lot of capital to float this marketing strategy. If the revenue doesn’t hit right away, it’s more of an investment, which can act as a deterrent.

• These funnel-filling strategies are generally ‘brand’ plays - think of car companies, for example. At a certain level of scale, it becomes a reasonable strategy to pay for awareness so that people remember your name when they see you out in the world. In this way, you will generally see incremental ‘lift’ in lieu of direct-return performance. If you’re an SMB, building that equity in those early days isn’t as impactful because if your product isn’t there to remind the potential customer that you are familiar, the equity you paid for fades more quickly.

Fundamentally, the mentality of optimizing for purchase immediately at all stages of the funnel exists because it has been proven time and time again to be an extremely profitable and viable strategy.

Unfortunately, the days of Facebook Ads being dirt-cheap are gone. With all of the changes happening in the landscape of digital marketing, many advertisers are revisiting old strategies that are tried and true. As everyone is probably aware by now, Facebook is having one hell of a time gathering off-platform data points, which makes it much harder for them to optimize towards those actions. Purchase, for example, being the furthest down the funnel, is one of the metrics being hit the hardest by iOS14.5+. Due to this, Facebook has been working to improve their platform, both on the efficiency of optimizing towards those lost metrics and also piecing that together with on-platform metric optimizations. For example, optimizing a campaign for Post Engagements (which happen on-platform) versus Add To Cart (which happen off-platform). Anything that happens on-platform Facebook has much more control over, including full visibility for reporting purposes.

So, things are changing, but that’s not to say that purchase optimization at the top of the funnel is out, right? Right. I, among many others, am still seeing decent performance coming out of purchase optimization at the top of the funnel. However, I think it’s time to start reintroducing some of these further up-funnel optimizations into the mix. Not only can you more cheaply reach a larger audience, but it dramatically increases the size of your retargeting pools. Recently, I wondered what it would look like if I added a Post Engagement campaign alongside my Purchase optimized campaigns.

It was time to test it.

Take a look at the screenshots below. To put this concept to the test, I ran a Post Engagement optimization campaign with a Purchase Lookalike audience for a balance of sheer size and slight segmentation to keep the audience relevant.

When was the last time you saw metrics this cheap? In fact, they are so cheap that they are being rounded up in the dashboard. The actual cost per engagement on this campaign was $0.007. That’s half a million new people in my middle-of-funnel retargeting campaign in just social engagers alone, never mind the additional actions that took place, such as link clicks, view contents, etc.

To break it down further, I have substantial data on video plays. Not only that, but I have 20,000 full play-throughs. If I can convert even 0.2% of those alone at a $100 AOV, I will more than cover my costs on this amount of spend. It’s worth noting that these videos are reasonably long, so the 3s video view mark is much before the 25% mark, so it’s understandable that there is a bit of a drop-off there. Either way, this was optimized for Post Engagements and not video views, so I would be interested to see how this changes if I swap that optimization.

Another thought I had for better understanding this data was looking at unique post engagements. The idea for this came from the fact that if someone ‘liked’ a post, then commented on it, Facebook would count that as two post engagements. So, I created a custom metric to see how many people you add to your funnel.

To add Unique Post Engagements to your dashboard, you need to create a custom metric with the following calculation:

Post Engagement * ( Post Engagement / Reach )

Leading into BFCM, this can be an incredibly valuable tactic where a large part of most people’s current strategy is building those retargeting audiences to hit hard come November. I will keep testing this, but the early results are very promising. If you have the capital to sustain converting potential customers after their initial touchpoint, then Facebook is still a seriously cheap channel.