Google Discovery Ads: 2026 Strategy for 2x ROAS

Listen to this article · 12 min listen

In 2026, you can’t just throw money at digital ads and hope for the best. You need precision. Google Discovery Ads are a serious option for tapping into user intent across Google’s massive network, showing up as visually rich, native ads right in the feeds where users are already looking for new things. These ads go beyond simple search queries to get ahead of user needs. So, how do you actually use this intent-based targeting to get significant returns?

Key Takeaways

  • To give Google’s machine learning enough to work with, you have to feed it at least three headlines and five descriptions so it has enough creative variations for optimization.
  • You can get a 15% to 20% higher conversion rate compared to broad targeting by segmenting your Discovery audiences with custom intent and lookalike audiences built from your own customer data.
  • Set aside 10% to 15% of your total Google Ads budget for Discovery campaigns. This gives you enough spend to get meaningful data and test performance without risking too much.
  • Use high-quality, varied images and videos. You need at least one square image (1:1), one field or “field” image (1.91:1), and one portrait image (4:5) to get your ads placed correctly across all the different feeds.
  • Keep a close eye on your Discovery campaigns by checking the asset performance reports and tweaking bids based on conversion value, with the goal of hitting a 2x return on ad spend inside the first 90 days.

Understanding the Discovery Feed Ecosystem

What sets Google Discovery Ads apart is that they don’t show up in search results. They appear in places where people are just browsing content, the Google Discover feed on your phone, the YouTube Home and Watch Next feeds, and even the Promotions and Social tabs in Gmail. Search ads react to what someone types into a search box, but Discovery Ads are meant to catch users while they’re exploring. This change in placement means you need a totally different approach for your creative and audience targeting, one that’s built around strong visuals and predictive intent instead of just matching keywords.

Under the hood, Google’s machine learning is what makes Discovery Ads tick. These algorithms look at a person’s past behavior, what apps they use, videos they watch, and their search history to figure out what they might be interested in. It’s about anticipating what a user might want to discover next, not just reacting to what they’re searching for right now. For an advertiser, this gives you a chance to introduce your products to an audience that’s open to new ideas but might not know what they want yet. The trick, of course, is to make creative that grabs them based on these predicted interests without being creepy or intrusive.

Crafting Compelling Creatives for Discovery

The visual element of Google Discovery Ads is everything. High-quality images and videos aren’t just a nice-to-have. They’re the main driver of engagement. I’ve seen campaigns crash and burn because the advertiser just dumped their old display assets into Discovery. Big mistake. These feeds are built for rich, immersive content, and your ads have to match that vibe. Think about what people are already looking at in these feeds, news stories, interesting videos, personalized recommendations. Your ad has to fit in with that content while being compelling enough to get a click.

A good Discovery ad campaign requires a deep bench of creative assets. If you look at the Google Ads documentation on Discovery campaigns (find it at support.google.com/google-ads), it clearly states you need multiple headlines, descriptions, and image ratios. You should be providing at least three different headlines (up to 40 characters) and five descriptions (up to 90 characters). For your images, you absolutely need a square (1:1), a field (1.91:1), and a portrait (4:5) version. This variety is what allows Google’s algorithm to run tests, mix and match combinations, and figure out what works best across all the different placements. Not providing this creative diversity is like trying to cook a gourmet meal with only salt and pepper. You’re just limiting your potential.

Telling a Story with Your Creatives

Past the technical specs, the story your creative tells is what really matters. People using Discovery are in a mode of exploration, so your ads need to tell a story or offer a benefit that makes them curious. Instead of just blasting a sales message, think about how your product solves a problem or makes someone’s life better. A travel company, for example, could run a stunning photo of a remote beach with the headline “Discover Your Next Adventure,” which is a lot more effective than just “Cheap Flights.” This works because it taps into the user’s desire to find something new. And while they aren’t required, short, punchy video clips (under 30 seconds) that show the product in action or create an emotional connection perform extremely well, especially in the YouTube feeds.

Using Intent-Based Targeting Strategies

The real muscle in Google Discovery Ads comes from their sophisticated intent-based targeting. Keywords are the backbone of search campaigns, but Discovery works by understanding user behavior and what they’re likely interested in. This is the point where your own customer data gets combined with Google’s massive data set. The whole point is to find users whose online behavior shows they’re likely to convert, even if they aren’t searching for your product right now. This is completely different from old-school display advertising that just sprays ads everywhere.

One of the best targeting tools in the box is custom intent audiences. You build these by giving Google a list of relevant keywords, URLs from websites your customers visit, or even apps they use. For instance, if you sell high-end hiking gear, you could build a custom intent audience using keywords like “best backpacking trails,” URLs from popular outdoor blogs, or a review site like “ultralight camping equipment reviews.” Google then finds people who’ve recently been looking at that kind of content across its network. This method lets you capture demonstrated interest with a lot of precision, even before that person has typed your brand name into a search bar.

Another powerful approach is using lookalike audiences, which Google calls “Customer Match” audiences, built directly from your own customer data. You can upload your customer email lists or website visitor data, and Google will find other users who have similar online behaviors and interests. This is a huge advantage because it lets you find more people who act and think just like your best customers, which is how you scale. According to a HubSpot report on marketing statistics (blog.hubspot.com/marketing-statistics), businesses using lookalike audiences tend to see a 15% to 20% jump in conversion rates over broader targeting. Getting this precise is everything. You’re actively finding more people who resemble your existing customers instead of just guessing who might be interested.

Refining Audience Discovery with In-Market Segments

On top of custom intent and lookalikes, Google gives you a whole menu of pre-defined in-market segments. These are Google’s pre-packaged audiences of people who are already actively shopping for specific things. For a car dealership, targeting the “Autos & Vehicles > Motor Vehicles > SUVs” in-market segment puts your ad in front of people who are showing strong signals they’re about to buy an SUV. In-market segments aren’t as specific as custom intent, but they give you a big, pre-qualified audience, which is perfect for products with mass appeal. Layering demographics like age, gender, or parental status on top of these segments tightens your targeting even more, making sure your ads get in front of the right people.

Budgeting and Bidding for Discovery Success

You have to be smart about how you budget for Discovery Ads. A lot of advertisers start with a tiny test budget, which seems safe, but it’s usually too small to get any real data. I tell my clients to put 10% to 15% of their total Google Ads budget here to start. This is enough spend to let Google’s machine learning actually learn something, optimize, and show you what it can do. These campaigns need a steady flow of data to optimize themselves, and if you don’t spend enough, you’re just starving the algorithm.

As for bidding, Google Discovery Ads run on automated bidding strategies. Your best bets are usually “Maximize Conversions” or “Target CPA” (Cost Per Acquisition). If your conversion tracking is set up correctly and you have some history, “Maximize Conversions” is a great place to start, telling Google to just get you as many conversions as it can for your budget. Once the campaign has run for a bit and you know what a good CPA looks like for your business, switching to “Target CPA” helps you keep your lead or sale costs in a profitable range. For any campaign that’s about generating leads, setting a realistic Target CPA is a must. If you don’t, you could end up paying way too much for leads and killing your profit margin.

Why Conversion Tracking is Non-Negotiable

Let me be clear: these bidding strategies are completely useless without solid conversion tracking. You have to properly install Google Ads conversion tags on your site for all the important actions, purchases, form fills, or even smaller things like a newsletter signup. If you don’t have accurate conversion data, Google’s algorithm is basically flying blind and can’t optimize for your actual business goals. This is a foundational requirement, not a mere suggestion. I see this all the time: businesses launch campaigns without double-checking their conversion tracking, and they just end up burning money and getting garbage reports.

Measuring and Optimizing Performance

Getting the campaign live is just the start. You have to constantly monitor and optimize to maximize your return on ad spend. The Google Ads interface gives you pretty detailed reports for Discovery campaigns, so you can see how things are working. You need to live in the “Asset report.” This report is gold. It shows you exactly how your individual headlines, descriptions, and images are performing, which is what you need to guide your next round of ads. Is a certain image getting tons of clicks but no conversions? It’s probably eye-catching but not actually relevant to what you’re selling. An asset with a lower CTR but high conversions, on the other hand, is hitting the right audience.

Go beyond just the assets and look at your audience segments regularly. Are your custom intent audiences working better than your in-market ones? Is one specific segment delivering all the high-value conversions? Use that information to move your budget around, tighten up your targeting, or build new audiences. For example, if you find that an audience interested in “luxury travel” is converting way better than one interested in “budget travel,” you should probably split them into separate campaigns or ad groups with tailored creative and bids.

A/B Testing and Iteration

Good optimization also means you’re always running systematic A/B tests. Don’t be scared to try out new headlines, different descriptions, and fresh images. You’d be surprised how small changes can add up to big performance gains over time. For example, you could test a headline that talks about a unique feature against one that promotes a discount to see what your audience actually responds to. Every test you run on a Discovery campaign gives you data that sharpens your understanding of the audience and makes future campaigns perform better. I recommend setting a goal of a 2x return on ad spend within the first 90 days, and from there, you can focus on making small improvements through constant testing.

Discovery Ads are great because they can reach users way earlier in their buying journey, sometimes before they’ve even thought about a specific product. This proactive style of advertising means you have to shift your thinking from reactive keyword bidding toward predictive audience engagement. If you make this shift and really commit to high-quality creative and smart targeting, you’ll find Discovery Ads become a go-to tool in your digital marketing arsenal.

What is the primary difference between Google Discovery Ads and Search Ads?

The main difference is user intent and where the ads appear. Search Ads show up in Google search results to answer a direct query. Discovery Ads appear in content feeds like Google Discover, YouTube, and Gmail, targeting users based on their inferred interests and browsing history, not an active search.

What types of creative assets are essential for a successful Discovery campaign?

You need a good variety of creative assets. For each ad, provide at least three headlines and five descriptions. Most importantly, you must have high-quality images in at least three aspect ratios: square (1:1), field/field (1.91:1), and portrait (4:5) to ensure your ads can run in all available placements.

How can I target specific audiences with Google Discovery Ads?

You have a few powerful options. You can use custom intent audiences (built from relevant keywords, URLs, and apps), lookalike audiences (based on your own customer lists), and Google’s pre-built in-market segments to reach users who are actively researching products like yours.

What bidding strategies are recommended for Google Discovery Ads?

Automated bidding strategies work best. Start with “Maximize Conversions” if you have enough conversion data. Once you’ve established a profitable cost-per-acquisition, switch to “Target CPA” to maintain control over your costs and profitability. Both depend on accurate conversion tracking.

How frequently should I optimize my Discovery campaigns?

You should be checking in and optimizing regularly. Review your asset performance reports weekly to see which creatives are working. Analyze your audience segments to shift budget toward top performers. And you should always be running A/B tests on your ad copy and images to find ways to improve performance.

Editorial Team

The Editorial Team at Paid Media Studio is a group of writers and editors who cover marketing, brand, and growth. They research each story and publish under a shared desk byline — experts in their fields, working as one team.