YouTube Advertising is kind of like the Mutual Funds of marketing channels: Someone told you that it was a good thing to do, you put a lot of money into it, but you really have no idea what the hell to do with it or whether or not your choices make sense.  Yet you soldier forward, thinking “well…everyone else is doing it, it must be good, right”?

While it’s hard for anyone to claim to be an expert in YouTube advertising these days (it’s still early days IMO) I’ve gone through enough trial and error to impart some knowledge.  There are plenty of posts that will tell you the basics of how to set up a campaign like here and here, so I’ll skip that part. This post addresses the audience targeting options and strategy.

The “Try Everything” Fallacy 

In my experience, Google Adwords representatives and digital marketing gurus will tell you to “Try all the targeting options in YouTube and then optimize towards what works” when setting up YouTube targeting.  While this method works great if you have a large budget an lots of optimization resources, I think this can be a frustrating approach if you are at a small, resource constrained startup.  Also, it can be difficult to optimize campaigns where you may or may not have truly significant signals to iterate off of (there will be another post on this).

I’ve bought large-scale media campaigns on TV and I believe that when thinking about targeting on YouTube, it’s best to approach the buy first from a media buyer’s perspective – using their rules of thumb –  not a digital marketer’s. In any upfront media buying, you have to think about key marketing principles of branded marketing prior to launching and plan your buy accordingly.

This means first thinking about who your customer is and what content on YouTube they might watch.  If you are making a media campaign for Bank Of America, for example, this could be a broad target as most of America needs a bank account (although it is arguable that even here, you could be more efficient). However, for most businesses, you want to start with content that is what your target customer would enjoy.  I’ve found there is generally plenty of scale on YouTube, even when you get very particular with targeting.

Don’t know who your target customer is yet?  Figure this out before you do YouTube (or any branded) advertising.  Run a past customer survey, or something. You wouldn’t buy TV  or radio without thinking about what message you were giving to what kind of people, so I wouldn’t recommend doing this on YouTube either.

Let’s do an example:  Say that you are a nation-wide running gear boutique chain that sells online and in retail stores.  How would you approach targeting on YouTube?  

First let’s know our customer.  In the case of this particular business, let’s assume that casual runners and fitness enthusiasts are the target customer.  People who are looking to run their first 5k, lose 15 pounds, and perhaps do some yoga on the side. So we’ll be focusing on the type of content that this consumer would enjoy to run our ads on.

Next let’s think about YouTube targeting options. For the purposes of this study, let’s say we want to run In-Stream ads (the ones that run before any video you watch, not the ones you click on in a search).

Since you already know the basics of setting up YouTube campaigns, you know you can target in the following ways:  Keywords, Topics, Placements, Interests, and Demographic Groups.  Placements are the most restrictive, followed by keywords.  Then come interests and topics.  Demographics are what you’d layer on top (things like age, gender, parental status as determined by the Google gods. We won’t spend much time on these specifics in this post).

Before you get excited about trying all of these kinds of targeting and seeing what works best, I would do a quick experiment.  Look up what Google thinks YOU are interested in and how they would serve you ads.  If your interests / topics are as crazy broad as Google thinks mine are, you’ll understand why this kind of targeting may not make sense for most businesses.

Go here to check it out:

This is what Google thinks my interests are:

Screen Shot 2016-11-25 at 5.28.45 PM.png


You may not know me, so I will let you know that I have next to no interest in combat sports or folk & traditional music. And while many of the other interests do describe me, they are so broad. I like “news”, “travel”, “books”, and “events & listings”.  This is starting to sound like the world’s most generic online dating profile, no? Could be anyone!

So unless your business is gigantic, I would argue against using topics or interests.  (The difference is that interests looks at the person Google thinks you are, Topics looks at the content you are looking at).  Even if a Topic is relevant to what you are advertising on, keep in mind that if the audience is extremely large you may violate one of the biggest laws of media buying:  achieving reach and frequency. Basically, in order to be successful in marketing, you must reach your users more than once.

This leaves Keywords and Placements as targeting options for our Running Store business.  

Because keyword and placement of targeting is more specific, we should do as much research as possible on our target consumer before getting started.  I would recommend surveying our current customers to ask them what sort of running or fitness related content they enjoy.  If you don’t know the content space, it will be hard to add in the right placements and keywords.

Once we’ve done our research we may find that our consumers enjoy watching cross-fit videos, watching running and healthy living vlogs, and watching nutrition-based videos like recipes for “whole 30” and the like.  If they are really ready to buy, they may watch running shoe review videos. So we are almost there.

First, let’s build out our keywords.  The way keywords work in YouTube is that they’ll match in a number of ways.  It will match with the content of the current video (the title, the keywords included in the description, etc.).  OR it can match with the content of the previously watched video.

I would recommend testing keyword targeting. We will build out adgroups like “running bloggers”, “whole 30”, “cross fit”, “running shoe reviews”, and the like with videos and an appropriate creative in it. This will target people with a true intent while they are in the mind-set of thinking about running.

Remember, even with YouTube keywords, you can do broad match and phrase match. I would recommend phrase match for most and would also recommend looking through and excluding irrelevant search terms daily.  Got to make sure you are not flushing money away.  Go to the video targeting tab, and then go to “search terms” to see what is triggering your ad.


The last thing you want is to come back to your campaign a couple weeks and a few thousand dollars later and realize all of your ads ran on Justin Bieber music videos. Which unfortunately has happened to a lot more often than most marketers would like to admit – including yours truly.


Placements are the most targeted form of YouTube advertising of all.  In my opinion they are the way to go for most businesses dipping a toe into the world of YouTube ads. The great thing about placements is that you know  and can determine exactly where your ads are playing.  In the case of this running store example, I would recommend making a list of popularly visited running related YouTube channels.  You can do this by simply searching the web or YouTube with keywords like “Running”,  “Running Gear”, “Running for Women”, “Running and Yoga”, etc.  Yes, it is extra work, but in my opinion, time well worth-it.

Soon enough you’ll have a long enough list to search for the YouTube channels and add them as placements!  Woohoo!


Here’s where you add placements. (go to video targeting tab –> placements –> + placements).  Create a couple adgroups in different placement channel categories and see how they run.

From here – congratulations, you are running your first YouTube campaign. Or at least you’ve done the targeting part!

You may be asking at this point:  “But Margot, how can I find new placements and expand my reach beyond just placements?”.

I would answer that 1) YouTube is a huge universe, it’s unlikely, even if you are targeting just on a few highly trafficked channels that you won’t get a high volume of views and 2) Sure, you are welcome to start experimenting with YouTube targets as well.  My advice though, is to get the fundamentals done first: Think about who you are targeting, think about what they like to watch on YouTube, and get your ads there, first.

Have fun!  🙂