Module 3 - Targeting methods

What targeting options are available for YouTube video campaigns?

There's a huge range of targeting options on YouTube, and the trick is not to overlay too many of them at once or overthink it too early. What we're trying to do is test different ways of getting our video campaigns in front of the right people.

The most basic starting point is demographics: age groups, genders, parental status, household income if you're in the States, and then more detailed demographics based on interests and traits. If you haven't set all of this up at the very start, that's fine, because you can learn from the data what's working and apply more detailed targeting to the next campaign. It's better not to get too detailed too soon.

How do interest, affinity and in-market audiences work?

Interest targeting is straightforward: you pick topics and look at the kinds of pages people visit and the other things they're interested in, without always tying yourself too tightly to your exact topic. Think about the other things your potential customer might like. Affinity audiences build on how people have already interacted with certain topics, and you can create your own custom affinity audiences based on which brands people have researched, what sports they follow, and similar signals. Life events targeting covers things like an upcoming birthday, a wedding, moving house or graduating.

In-market audiences are particularly good for YouTube, because they let you reach people who are already researching and searching for products and services like yours, for example people who've been visiting competitor websites, which Google is able to identify and build into audiences for you. Custom intent audiences take this further, letting you reach viewers at the exact point they're deciding to buy, which is crucial on YouTube because an in-stream ad interrupts someone's viewing, whereas a discovery ad reaches someone who is already searching for a topic and is more likely to engage because they sought that content out themselves.

How does remarketing and customer match targeting work?

Remarketing audiences let you reach viewers based on their past interactions with your videos: which videos or playlists they watched, which topics interested them, how far through they engaged. Website and app remarketing extends this to people who've been on your website or interacted with your app, or with particular categories of app from the app store. All of this comes back to conversion data, and the tools only matter if you're actually testing within those audiences rather than sitting back and hoping for the best.

Customer match lets you upload your own customer data and build similar audiences around it, but you typically need a certain level of spend and a genuinely large customer list before Google will enable it, generally several thousand rather than a hundred or two hundred, because if you upload a small list Google may only match a fraction of it, leaving too small an audience to serve ads against effectively. Similar audiences then let Google expand your reach further from your existing audiences. That costs more budget, but similar audiences can sometimes outperform the audiences you built them from, which is particularly useful once you're ready to scale a campaign.

Why does overlaying keywords, topics and audiences matter?

Refining your audience, rather than relying on a "spray and pray" approach with off-the-shelf audiences, is what most YouTube advertisers still aren't doing, and it's what can halve your ad cost. If you're spending thousands of pounds a month, that's a genuinely significant saving, and it's exactly the kind of detail that puts you ahead of competitors who haven't bothered to refine their targeting.

Beyond audiences, you can also target by placement, channel and specific video, and by topic, where Google has categorised an enormous range of subjects, sometimes surprisingly niche ones. Keyword targeting still matters too, and the keywords you choose need to be specific and niche. The real power, though, comes from overlaying these methods: combining a keyword someone is searching for with a topic or affinity or intent audience they belong to. An ad performs far better when topics, keywords and audiences intersect and overlay, rather than simply running a set of ads against a flat list of keywords the way you might with traditional PPC.