A lot of people discuss and debate retargeting. There is no right way to do this, but I want to stress that regardless of HOW you do it, it's important to do it. Once you figure out your funnel and you got a winning product/adset/ad combination, don't forget to retarget your visitors.
Even if you have a loss leader front end funnel (free + shipping), retargeting can make all the difference in your campaigns.
Yes, this means post event retargeting too. More on that later (not so much for leads, but for purchases)
Before diving into retargeting and retargeting funnels, I want to stress that it's very important that you have your ads and audiences figured out. You don't want to use retargeting until you have a good working front-end offer.
Furthermore, you want to retarget users who drop off at a particular point in your sales funnel and nudge them further to the next stage. Each stage in the funnel is designed to push them along to the next stage. Even the final stage is supposed to continue pushing them along.
For example, you can nudge them along to take an upsell if they dropped off at the first upsell.
In any case, you should ALWAYS retarget your users when you are hitting them and ready to scale. This should be setup so that users travel through a process and machine. Each component will need optimization, but you should be keen on putting several ad variations in each ad set so your retargeting doesn't get stale.
Retargeting can take on many forms. There is DPA, and more intensive retargeting funnels that I call Advanced Segmented Retargeting. Obviously DPA is pretty easy to setup, and you can use it along side this technique.
If you’re like most people you have a misunderstanding of retargeting and feel it's a one-size-fits-all approach. You see, retargeting is not one of these things to take lightly. It’s literally leaving money on the table.
If you're running ecommerce product funnels, your Facebook retargeting ads should be returning 5-10x ROAS on your products compared to your front end ads.
In a lot of cases that I see, the retargeting is just slapped together capturing all users that take a particular action. For example, all users who added to cart in your ecommerce store in the last 30 days. This captures all customers with only 1 buying intent.
What if you can actually segment users by levels of intent and distance they are from your funnel? What if we can segment users with retargeting the same way we can segment users with email?
Indeed, Facebook Ads does offer a way to do this, but it’s not readily available…or readily obvious. It takes some understanding of how to segment your audience within the platform so that you can identify what the levels of intent & time are for your business. To do this right, you need to include/exclude many audiences from your funnel. If you visualize your funnel and the customer journey, the inclusion/exclusion becomes very obvious, but can make your head hurt.
The below example is intended for eCommerce, but the same principles can be applied to virtually any Facebook Ad campaign.
The first step is time.
Obviously, time is universal so we don’t need to define time, but you do need to identify how much time passes before showing the user a particular offer. Ok fine, you got me. Time is relative to the observer according to Einstein’s Theory of Relativity and can change and fluctuate. Let’s keep this discussion about Facebook ads and not get involved into physics. We can table this part of the discussion for another day, and perhaps even another site. We aren’t physics professors, just media buyers and advertisers. Hold this in your mind, we'll get to it later.
Then we have level of intent. As with any advertising campaign, this is super important to understand.
What I mean here is how much engagement has the user gone through with your site. Since standard events are included in every Facebook pixel and they always include eCommerce events, let’s start there.
Here's the example There are the 4 types of standard events that come with every pixel as it pertains to eCommerce: View Content (VC), Add to Cart (ATC), Initiate Checkout (IC) and Purchase (P). We’ll go through each one.
VC refers to the user viewing the content that is presented to them on the screen. In other words, did this user see the page?
This doesn’t tell us what they saw or how much they saw, so this puts the user in a low level of intent category.
Next up is the ATC event. This refers to the user actually clicking on the item, perhaps selecting variants, and adding to the cart. He or she may have read the description, seen some images and saw the price. They may have decided to purchase or just wanted to add it to their own internal wishlist and save it for later.
By default, ATCs have a higher level of intent, but may not show buying intent.
This is because users can add to cart to save the product for later use. However, this does mean that they interacted more with your site than the VC users so it’s a Medium level of intent.
After ATC comes IC. IC refers to the user clicking through the cart page and going to the checkout screen. They may have started the checkout process or they may have just hit the checkout page and dropped off. In either case, these users show a much higher level of intent. Especially, if the user inputted their credit card and it got declined. This would put this person in the High level of intent.
Finally, there is a purchase event. Since this user is a customer, they, by default, show high levels of intent. Let’s say these users are our customers and we understand by default customers have the HIGHEST level of intent. I mean, they gave us their credit card information and show massive trust.
The purchase event can be further divided into upsells and downsell drop off purchase events. I won't get into this too much as it gets even more granular than what we are already discussing.
Ok, now that you understand levels of intent, let’s talk about time again. Showing the offer at the right time to the right user is very important in a successful Facebook Retargeting campaign. Time can be broken down like this:distance-from-funnel
- 0-3 Days
- 4-7 Days
- 8-14 Days
- 15-30 Days
- 31-60 Days
- 61-90 Days
- 91-180 Days
Then your levels of intent:
Now combine them for each ad set
- -0-3 Days x VC
- -4-7 Days x VC
- -8-14 Days x VC
- -15-30 Days x VC
- -31-60 Days x VC
- -61-90 Days x VC
- -91-180 Days x VC
- -0-3 Days x ATC
- -4-7 Days x ATC
- -8-14 Days x ATC
- -15-30 Days x ATC
- -31-60 Days x ATC
- -61-90 Days x ATC
- -91-180 Days x ATC
- -0-3 Days x IC
- -4-7 Days x IC
- -8-14 Days x IC
- -15-30 Days x IC
- -31-60 Days x IC
- -61-90 Days x IC
- -91-180 Days x IC
- -0-3 Days x P
- -4-7 Days x P
- -8-14 Days x P
- -15-30 Days x P
- -31-60 Days x P
- -61-90 Days x P
- -91-180 Days x P
You can have 4 Campaigns here. VC, ATC, IC and P and the time represents an ad set. Each of the above combinations represents a new custom audience. Within each ad set, I recommend creating 3 different ad variations. All in all you are left with 84 ads within your ad sets in total. Not all your ad sets will survive and not all your ads will be winners. Hence, why we test with multiple variations. You can even consider using dynamic ad here to make things even more fun.
Each ad copy should be unique and present a new offer to the customer depending on what product they committed to. If it's purely front end products, offer them a discount ladder that gradually increases or decreases (test both directions). If it's a sale and a user dropped off the upsell sequence, you can offer them the product that they dropped off at with a discount ladder. Each time, coupled with a level of intent, can be presented as a separate offer to the customer.
It's important to segment your users based on intent because not all users respond to the same offer, especially if their intent and journey looks different.
Furthermore, if you show them multiple styles of offers and multiple types of ads to the same customer you will limit banner blindness and expose users to different messages from the same company. Showing all users the same ad over and over and over increases the frequency and the chance the user is going to ignore your marketing.
In fact, they may even hide all your future ads diminishing your relevance score and making you pay more for your ads. Then you can show different offers to different users who will respond differently to each ad. Some ads will have 10+% CTR, where as others will have less than 1%, depending how engaged the user is and how close they are to your original message. Over time and frequency, the message you continuously ram down your customer's throats will get stale. This method keeps things fresh.
So think about crossing each time period with a different level of intent. And promoting the right part along the funnel depending on where the user dropped off in your funnel. ASR can take on many forms and isn't constrained to one particular style. But you can vary several offers within it. Each level of intent and time combination should have its own offer associated with it that makes sense according to your initial front end offer.
You don't have to get as granular with your audience as I like to, but you can certainly experiment with a broader time period segment. Furthermore, you don't have to segment by as many intents either. If you are running a lead campaign, you can just do pre-lead and post lead campaigns to encourage them to convert.
If setup correctly, this can become a marketing machine that just churns out results for you. ‘
I also have a diagram that will help you better understand how to segment your audiences based on intent and time. This download shows you what to exclude/include in your Facebook Audiences so you do this right and you don't include/exclude the wrong audiences and break the cycle. It goes over the above examples and how to create the correct audiences per product. Again, this can be applied to any offer type, but it is fairly easy to visualize with an ecommerce example.
What questions do you have about Facebook Retargeting? Please write them below!