AI-driven ad automation: Tiktok Smart+ vs Meta Advantage+ vs Google App Campaigns

AI-driven ad automation platforms that promise to do all the work for you are suddenly everywhere. It seems like every major platform is now using AI to automate app promotion campaigns. What they’re also doing at the same time is aggregating all of the many components of their ad supply and attempting to match all those placements and properties with advertisers’ demand.

Google started it with UAC, or Universal App Campaigns, way back in 2015. (UAC, of course, is now Google App Campaigns.) UAC promised to help app developers promote their apps across various Google platforms, including Google Search, Google Play, YouTube, and the Google Display Network, while doing minimal setup and management.

AI vs ad automation

AI was not as much part of the story back then as ad automation, although the company did highlight that it was using machine learning to drive app installs and in-app actions. 

More recently, Meta brought in Advantage+ in 2022 to compete. Just this month, Tiktok released Smart+ campaigns. (And Apple is rumored to be testing a similar functionality.)

For all of them, the pitch is similar: our AI-driven ad automation systems will make it easy for you to grow.

Upload ads, insert money, relax.

There’s obviously some truth in that. The question, of course, is how much, and whether it’s not just easier to spend money. Or, worst case scenario, waste money fast. 

The value for Google and Meta specifically, however, is also that they can seed ads into perhaps less-desirable corners and alleys of their various properties, thereby monetizing everything they do and giving them considerable opportunity to try to meet advertiser expectations for ROAS and ROI but not necessarily massively beat them, which would then waste monetization opportunities. How much this benefits paying advertisers is questionable.

Be that as it may, AI-driven ad automation systems are a thing. They’re here. They’re not going anywhere.

So I wanted to look at each and compare them, plus add some context from mobile app marketers in terms of what they’ve experienced using them. TikTok Smart+ will be the hardest platform to do that with, simply because it’s the newest, but there are many gray or not-fully-disclosed components to all of the platforms.

Ad automation: TikTok Smart+ vs Meta Advantage+ vs Google App Campaigns

We’ll look at most of the features of the various AI ads systems, including how much work you need to do to set them up. (Hint: in most cases, it’s not that much.)

Note that while all of these ad automation platforms currently say they’re using AI to generate creative as well as (in some cases) mixing and matching your own creative, this is not the same as full-on generative AI ad creative creation, which is generally something that Amazon, TikTok, Google, Microsoft, and Meta are working on, but typically don’t release to everyone, everywhere, right in the ad tools dashboards.

Expect that to get integrated over time.

Summing up

There’s a lot to look at between all of these ad automation options.

TikTok Smart+ is obviously the newest kid on the block, and TikTok’s ad tools in general are younger than both Meta’s and Google’s. There’s some value in that: they’re often simpler, for 1 thing. But TikTok Smart+ and other TikTok tools are also often simpler just because TikTok ads show on TikTok only … as opposed to Meta’s entire ecosystem of apps and properties, and Google’s massive universe of owned and partnered publishers.

All of the ad automation systems make life easier for mobile marketers.

But they also all come with some costs.

Just accepting all the defaults of a super-simple system has its risks, as I chatted with Replug’s Lorenzo Rossi about in a recent Growth Masterminds episode. Sometimes the easy button in Advantage+ makes setting up ad campaigns too easy, and he’s been bit with audience expansion beyond a cohort that would actually convert:

The hard (and expensive) lesson: not having all the dials and levers of a complex advertising platform at your disposal makes advertising easier, but it doesn’t always make it better. 

Another challenge: giving the platforms all the creative possibilities they want makes it easier for them to run your ads in locations that don’t convert nearly as well. Think Google Display Network versus Google Search. Or Messenger versus Facebook. In some cases, it’s better to limit the creative you provide so that you force the machine to use it where you will get better results.

One example: YouTube.

I recently chatted with Ashley Black, an ex-Googler who spent most of her decade with the tech giant in the mobile user acquisition space, about creative built specifically for Google:

YouTube has some very specific ad opportunities that might be performing extremely well for you. To ensure you hit them, you have to be very specific about which kinds of creative you upload into Google App Campaigns.

“I think one of the big things that people don’t realize is that YouTube, especially on iOS, is a massive portion of inventory for Google,” says Black. “And you have such different types of video content on YouTube.”

Sure, there’s shorts and long-form videos. There’s also bumper ads, pre-rolls, mid-rolls, partially skippable and not skippable ads, sponsored cards for in-stream discovery, and all the non-video ads that YouTube includes.

Without the right formats and sizes, some of these opportunities are just out of bounds for Google App Campaigns, meaning 

As always, your mileage may vary

The various AI-driven ad automation systems are constantly being updated. The AI is getting better and better over time, and generally so are the tools, levers, and analytics they offer.

The real question is which perform better and provide greater ROAS.

For that, keep an eye out for our refreshed ROI Index, coming out in Q1 of 2024.