Whether it’s your first stab at creating a Facebook campaign, or you’re a seasoned vet, the importance of account organization is crucial in keeping track of your account’s health, as well as keeping your sanity intact. To get the most out of your advertising campaigns, segmenting your audience into logical clusters is key to long-term success. Outlined below is an effective approach to segmenting your campaigns in a way that makes sense for efficient testing and optimal advertisement organization.
It’s worth mentioning that this post is not designed to help you decide what your target audience should be, as everyone’s ideal will be different. This post is simply used to show you an easy way to organize and segment your account for optimal performance and clarity. That being said, we won’t worry about segmenting by broad or precise interest-based categories – instead we can focus on our less ‘psychographic-specific’ targeting options.
By now you’ve probably noticed Facebook’s fluctuating ‘reach’ number off to the right-hand side of the ad creation interface. This number is worth paying attention to early on in your ad creation, as it will allow you to see how many people in your overall target demo are in each of the segments you will create. When creating various campaign segments it is important to make sure they are in logical chunks that you could pause if the performance starts to drop, or perhaps increase bids on campaigns that are turning better results for your campaign.
Segment by Gender
This is always an easy place to start – which gender performs better with your current creative? In no way am I suggesting that you should halt all advertising to either gender, but in many cases different creative will resonate in different ways between genders.
Segment by Age
There is a wide range here, so you can decide for yourself how you want to split up the ‘13-64’-age range. In the past, I’ve used segments of eight years (13-20, 21-28, 29-36, etc.) to determine the top- and worst-performing age segments for my Facebook ad campaigns – but this range may vary depending upon the product or service you are using. The idea here is to get comfortable with the data that is reported back to you once these segments have been tested. If you wanted to get very granular with your targeting, you can target every age separately; just make sure your reach is high enough to give you statistically significant volume.
Segment by Location
This is going to vary widely on the reach of your product or service. If you want to advertise to consumers on a national level, splitting this segment up by region or state would be a wise place to start. You can quickly identify which states will be top performers during your campaign, and stop targeting the locations that are underperforming. If you are advertising on the local level, you may want to consider segmenting by city or even zip codes around your location.
Segment by Bidding Style (CPC, CPM)
This is something too many advertisers forget to test when starting a new campaign. Just because you’ve seen a lot of success bidding for clicks (or impressions) doesn’t necessarily mean it will be the best option for each new campaign. If you optimize your campaigns for a strong CTR first, sometimes switching over to a CPM model can result in a better cost per conversion for the campaign – something to be mindful of in your testing.
Segment by Placement (Right Column, News Feed)
The News Feed has almost always performed better for direct response campaign in my experience, but there certainly is a trade off here to be aware of: competition. Placement position is the real estate of the web, and if more advertisers want their ads to appear somewhere, the price tag will be higher for that position. Keeping that in mind, you may experience lower CTRs on the right column placements, but the average CPC/CPM for this placement can oftentimes be lower than the much sought-after News Feed placement.
Segment by Device (Desktop, Mobile)
Long-gone are the days of the outrageous CTRs that early-adopting advertisers experienced on mobile devices, but this doesn’t mean that there isn’t still plenty of opportunity to see solid performance from advertising on these devices. One thing to be mindful of here is your website’s mobile experience – is your website responsive? If the answer is no, you may decide that limiting yourself to just a desktop placement is your best option.
All of these segments are split into separate clusters, allowing us to compare performance to their relative counterparts, but something that deserves more exploration is the idea of multivariate testing. (This is simply the idea of testing multiple layers of segmentation within the same campaign.) For example, do males on mobile devices perform better than females on desktop?
As you can see, this gets more complex the more segments you decide to add to your campaign, which is why I recommend you start with the larger clusters first and work your way down from there. If you narrow your reach down to less than 20 users, it doesn’t matter how great that segment performs if it can’t be scaled. Finding a good balance between reach and effectiveness is very important for campaign longevity.
One of the most important takeaways here is to make sure you are not assuming what the best targeting criteria will be – or, even worse, lumping all of your targeting together into one campaign, leaving you incapable of determining which criteria in your targeting settings are increasing or decreasing the overall performance of your campaign.
Obviously this list is not all-inclusive when determining which segments you can optimize your ads around, just some of the larger clusters I’ve had personal success working with. I highly encourage you to find and test different targeting segments to work with, and proceed to share your results in the comments below!