How I set up retargeting campaigns to get a 300% ROI (minimum)

Do you ever get ads for products you looked at on Instagram a few days ago? Or maybe you added something to your cart, didn’t end up purchasing, and now you’re getting ads for a 25% off coupon to complete your purchase. This is the power of retargeting.

Retargeting (also called remarketing) is simply advertising to people that have already interacted with your brands or website. Generally, you show your ad to people that visited your website, but didn’t complete a key action (purchased, submitted a form, etc).

Retargeting is some of the lowest hanging fruit you can ever target, regardless of your business type, industry, or who your customers are. Compared to prospecting campaigns, your costs (and associated cost-per-acquisition) are going to be fractional. See some examples below…

In this post, I’m going to show you the exact starting structure I use for creating profitable, sustainable retargeting campaigns. For simplicity’s sake, I’ll be using an eCommerce company as a running example, though this structure can be applied to nearly any business (Blogs, B2B, local businesses, etc.)

How does Facebook retargeting work?

By now, I’m guessing you know what the Facebook Pixel is and what it does. Essentially, every time someone visits your website (with the pixel installed), you have the potential to target them when they visit any site within the Facebook ad network (Facebook, Instagram, Messenger).

The key here is that advertising to these people is MUCH cheaper than if you were trying to target them under a cold audience. Additionally, because these people have already visited your website and theoretically know about you, you’re likely to enjoy:

  • Better clickthrough rates (CTRs)
  • Cheaper cost-per-clicks (CPCs)
  • Higher relevance scores
  • Lower cost-per-acquisitions (CPAs)

I hope you’re beginning to realize just how powerful retargeting can be! However, for some businesses, it doesn’t make sense to immediately try and sell their products in a retargeting ad, they still need to be ‘warmed up’ a bit more before being sold. Which leads us to our next section…

Dynamic Retargeting vs. Normal Retargeting

Dynamic retargeting is showing people the exact product that they viewed/added to their cart, rather than just a ‘static’ ad. This is done by pulling your entire store’s catalog into Facebook, and Facebook will display the appropriate products as necessary in your ads. Below is an example of what these ads look like:

The products in this carousel will automatically be the product that the user last viewed, or added to their cart.

Dynamic Retargeting ads need to be set up with the ‘Catalog Sales’ objective within Facebook. With this, you’ll get access to the following targeting options, which makes the process super simple.

There are tons of catalogs that allow you to pull your entire store’s inventory into Facebook with ease. For Shopify, I’ve used and recommend Flexify. For other platforms, just Google ‘<your eCommerce Platform> Facebook Catalog’.

I highly recommend utilizing dynamic retargeting. In almost every case I have found it to outperform ‘normal’ retargeting ads. The hyper-specificity of the format is just too powerful to ignore.

How many retargeting steps do you need?

The vast majority of businesses can get away with just 1 retargeting step. That is, immediately show your product/form to people that either visited your website (and didn’t purchase) or abandoned cart. However, for higher-ticket stores, or B2B businesses, it often makes sense to add more steps.

Generally speaking, the more expensive/more investment your product or service requires, the more retargeting steps you’re going to need. These additional steps can be the form of articles, to try and educate them further, or lead-gen ads, to try and get them into your autoresponder sequence.

Let’s say you’re selling a high ticket consulting service ($5,000-$10,000 range). You’re retargeting sequence could look something like this:

Step 1. Retarget all website visitors who never joined mailing list, with a white paper locked behind an email submit form

Step 2. Retarget everyone who downloaded the white paper but never booked a consultation call, with an ad directing them to book a consultation call

Step 3. Retarget everyone who booked a consultation call, but didn’t end up purchasing, with an ad directing them to purchase

Compare that to an eCommerce store selling a product that costs, say, $50, who could just do the following:

Step 1 (of 1). Retarget all website visitors who added an item to their cart, but never purchased, with an ad directing them to finish their checkout.

Think about what makes the most sense for your business. When in doubt, go with fewer steps. Adding too many steps can dilute the process and spike your costs without much return.

Now that we’ve covered the basics of retargeting, we’re on to the main event…

My exact framework for creating profitable retargeting campaigns

Before proceeding with any of this, you must be absolutely positive that your pixel is installed AND firing correctly on key events (add-to-cart, form submissions on lead forms, etc). If it’s not working correctly, none of our retargeting campaigns will work correctly.

It’s easy enough to verify this. Within the Facebook Ads Manager simply go to the ‘Measure & Report’ tab and click on ‘Pixels’.

Click on the title that says, ‘<Your Businesses> Pixel’ and scroll down. In here, we can verify that everything is firing correctly. The screenshot below shows a properly installed and functioning eCommerce pixel (running on Shopify).

As with all things paid traffic, segmentation is of the utmost importance. I stress this a lot, and for good reason. Unless we have distinct ad-sets each targeting a unique area of traffic we want to remarket to, we will not have control over our results.

Think about it like this. Below is a (very crude) image of 2 example ad set structures. Which one is better?

Theoretically, both of these examples are targeting the same people. The only difference is that the 2nd example is more ad sets. Seems like a lot of unnecessary work, right? Wrong! Unless we have these segmented ad sets, we only have the average of how all of these audiences are performing. Our abandoned cart visitors could be converting great, while our website visitors are doing poorly. Unless we had them each in their own ad set we would never be able to see exactly which audiences are working well for us, and which are not.

Hopefully that all made sense, because this is the basis of how we are going to be structuring our retargeting campaign.

For the majority of businesses, I recommend doing a 4-part, 2-week segmentation. Similar to the example mentioned above, our main focus is going to be website visitors and abandoned cart-ers. You can have as many conversion events as you’d like, but these are often the best places to start for eCommerce businesses. For B2B/blogging sites, you can replace the abandoned cart event with website visitors that visited a key page on your website, like your contact page, but never submitted the form.

Without further delay, here’s what the ad set structure looks like:

  • Website Visitors:
    • Last 1 Day
    • Last 3 Days
    • Last 7 Days
    • Last 14 Days
  • Abandoned Cart
    • Last 1 Day
    • Last 3 Days
    • Last 7 Days
    • Last 14 Days
Why can’t these display in proper descending order Facebook???

The logic behind this is pretty simple. The more recently someone interacted with your brand or website, the more likely it is they’ll remember you if you show them a retargeting ad. This is why the maximum range I set for this structure is 14 days.

Sure, we could increase our reach and spending if we increased this to a month, maybe even longer. But the results we would get from attempting to remarket to these customers wouldn’t even be worth the ad spend. It’d be far more worthwhile for us to focus our attention on the most vital prospects: the ones that still remember us!

We’re not done yet though. Because we’re segmenting on overlapping timeframes, there are some exclusions we need to set up. We want each ad set to be targeting each timeframe exclusively. For example, the ‘Abandoned Cart, Last 3 Days’ technically covers the same timeframe as the ‘Abandoned Cart, Last 1 Day’ ad set. For this reason, we need to exclude the 1 Day ad set inside the 3 day one.

If you’re using dynamic retargeting, you can set exclusions by ticking the box that says ‘Custom Combination’, like in the example below:

If you’re just using custom audiences, set up the targeting like the example below:

Interpreting Results & Optimizing

Because we have this segmentation set up properly, it’s incredibly easy for us to see what timeframes are working for our product. As you might expect, the shorter time frames are generally going to perform best, since your product, service, or website is most ‘fresh’ in people’s heads.

However, just because your results are technically ‘worse’ on the later timeframes, you’ll still be getting a great ROI. For most businesses, it makes sense to continue running these ‘poorer performers’ so as to not limit the potential reach of your retargeting campaign as a whole.

As far as optimizing goes, your best bet is to experiment with different ad formats, and your potential offer. If you have an eCommerce store, dynamic product ads (DPAs) can be used to show the exact product people looked at/added to their cart, which generally outperforms generic images. Experiment with single image ads, carousels, and videos. Consider adding in a discount code, or some other ‘bonus’ to incentivize people to complete their purchase.

What time ranges have worked best for you when doing retargeting? Let me know in the comments below 🙂

Leave a Comment