These days, ad-buying platforms offer you a breadth of targeting options and not a ton of guidance around how to do it efficiently. Many people believe that granular targeting yields better results, but from my personal experience I’ve found that this isn’t always true.
Don’t get me wrong, I think there is a time and place for granular targeting, but it isn’t always the right way to go, especially in the beginning. Below are a few things to think about when you’re getting ready to start a new campaign.
Budget is and will always be your biggest factor when deciding how granular you want to be when targeting. You want to make sure that with every campaign you launch, you are gaining statistically significant data surrounding each of your goal Key Performance Indicators (KPIs). Your daily and monthly spend play a huge factor in determining the significance of your learnings from each campaign.
For example, if you only have a very small budget, say $5 a day to spend on advertising, and your Cost Per Impression (CPM) varies between $5 to $10 a day, you will barely be able to reach enough people to get some actionable learnings.
CPMs categorically increase the more granular you get with your targeting, so if you are budget constrained, try broader targeting (an audience size of at least one million people), and report against age, gender, and location to understand the better performing outliers within your target audience. This will allow your money to go further.
On the other hand, if you have millions to spend on advertising, I highly suggest running multiple, more granularly targeted campaigns. The higher your budget, the more you can break down your audiences into granular subsets and still gather substantial data. This can allow you to test copy and creative by location or understand the difference in purchase behavior between men and women, which will help you make more informed decisions in the next campaign you launch.
Little To No Data (New to Market/Small Businesses):
It isn’t uncommon for many companies to go to market with an idealized customer in their head with little to no data showing that this customer profile is actually who is purchasing their product. Although it would be great for all of your customers for your new fashion line to be in their mid-20s, living in the top metro areas, making an $80,000+ annual salary, all with executive level positions, the chances of that being true is highly unlikely.
Don’t get bogged down on who you want your customer to be and instead start broad testing different targeting theories and let the data guide you. For example, take your age bands above and below your theorized customer age, start nationwide, and try a couple of different interest and behavioral layers that will allow you to make informed decisions.
Prove a Theory:
Granular targeting is helpful if you want to prove a theory. For example, you are looking through your analytics and see that your company’s per session value in Seattle is $5 more than it is in Los Angeles, but you only have 100 sales from Seattle and you have 10,000 sales in Los Angeles. One hundred sales isn’t a very significant amount when you compare it to Los Angeles, but the per session value makes Seattle seem like a promising location for growth. How do you figure out if this theory is true or not?
The fastest way for you to gather supporting data is to run a controlled test in Seattle and see if that trend holds true with additional traffic.
Unique Business Conditions:
The most obvious use for granular targeting is unique business conditions. If your product is only for women, not sold nationwide, or only usable by a certain trade, then granular targeting allows you to target effectively and spend efficiently by cutting out all of the extra fluff.
The biggest caveat to this is scalability. If you have a very niche product that can only be used in a specific location, by a specific gender, or a particular income level, then your audiences can be so small that you reach them all within a week and then become stuck.
In general, the importance of targeting are the learnings. Make sure that your budget and audience size are substantial enough to allow you to gain actionable data. Don’t make targeting decisions based off of an idealized customer, start broad, with a large enough budget and key KPIs in mind, and gather data that allows you to launch new, more intelligent audiences.