YouTube Ads on YouTube: What Works in 2025

Running a YouTube ad on YouTube means placing video advertising directly within the platform where your audience is already watching. You bid through Google Ads, choose your format, set your targeting, and your ad appears before, during, or alongside organic content. Simple in theory. Considerably more nuanced in practice.

The mechanics are straightforward. What separates campaigns that generate real revenue from those that burn through budget and produce a tidy report full of impressions is almost never the setup. It is the creative, the targeting logic, and the commercial discipline behind the decisions.

What Are the Main YouTube Ad Formats and When Should You Use Each?

There are five formats worth understanding properly. Most advertisers default to skippable in-stream because it is the most familiar. That is not always the right call.

Skippable in-stream ads play before or during a video and can be skipped after five seconds. You pay when someone watches 30 seconds or more, or interacts with your ad. This is the format most brands think of when they think YouTube advertising. It gives you the most creative runway, which is both its strength and its trap. The skip button is right there. If you have not earned attention in the first five seconds, you have not earned anything.

Non-skippable in-stream ads run for up to 15 seconds and cannot be skipped. You pay per impression. These work well for brand messaging where you need guaranteed delivery, but they carry a cost beyond the media buy: you are asking someone to sit through something they did not choose. Creative quality matters even more here, because there is no opt-out valve to filter disengaged viewers.

Bumper ads are six seconds, non-skippable, and priced on CPM. They are not built for storytelling. They are built for recall, frequency, and reinforcing a message that exists elsewhere. I have seen bumpers work well as a retargeting layer after a longer skippable ad has done the heavy lifting. Standalone, they are limited.

In-feed video ads appear in YouTube search results, on the homepage, and in the recommended video sidebar. They look like organic content. The user has to click to watch. This changes the dynamic entirely: you are paying for intent-driven views rather than interruption-based impressions. For content-led brands or campaigns where watch time and engagement matter, in-feed is underused.

Masthead ads sit at the top of the YouTube homepage and are bought on reservation. The reach numbers are significant, but so is the cost. These are brand plays for brands with brand budgets. If you are reading this to figure out where to allocate a mid-market performance budget, masthead is not the conversation to have right now.

If you are building a broader video marketing strategy, not just YouTube advertising, the Video Marketing hub at The Marketing Juice covers the full picture: from production and distribution to measurement and channel selection.

How Do You Set Up a YouTube Ad Campaign in Google Ads?

YouTube advertising runs through Google Ads. Your video creative needs to be uploaded to YouTube first, either publicly or as an unlisted video. Then you build your campaign from inside the Google Ads interface.

The campaign objective you select shapes what options Google surfaces. For direct response, you want leads or website traffic. For awareness, you want reach. The objective is not just a label. It affects bidding strategy, format availability, and how Google’s algorithm optimises delivery. Choose the wrong objective and you are optimising for the wrong signal from day one.

Once you have selected your objective, you will set your budget, bidding strategy, network settings, and targeting. A few things worth being deliberate about at setup:

Network settings: By default, Google will include YouTube search results, YouTube videos, and video partners on the Display Network. The Display Network inventory is often lower quality. Unless you have a specific reason to include it, exclude it at the start and test back in if needed.

Bidding: Target CPV (cost per view) gives you control over what you pay per view. Target CPM works better for reach-focused campaigns. For conversion-focused campaigns, Target CPA or Maximise Conversions hands more control to Google’s algorithm, which can work well once you have enough conversion data, and can waste budget when you do not.

Frequency capping: YouTube does not always cap frequency by default. Left uncapped, you can end up serving the same ad to the same person ten times in a week. That is not brand building. That is annoyance. Set a frequency cap, especially for non-skippable formats.

Placement exclusions: You can exclude specific channels, videos, and content categories. For most brands, excluding content in sensitive categories (tragedy, conflict, sexually suggestive content) is worth doing at campaign launch. Brand safety is easier to manage proactively than reactively.

I spent years managing campaigns at scale across multiple verticals, and the setup decisions that seemed minor at launch, frequency caps, network inclusions, placement exclusions, were consistently the ones that created the most expensive problems three months in. Do not rush the setup.

What Targeting Options Are Available on YouTube?

YouTube targeting is more sophisticated than most advertisers use it. The standard approach is demographic plus interest targeting. That is fine as a starting point. It is rarely the most effective approach for direct response.

Demographic targeting covers age, gender, parental status, and household income. Useful for filtering, less useful as the primary targeting signal. Knowing someone is 35 to 44 and male tells you relatively little about whether they are in the market for your product.

Interest and affinity audiences are Google’s pre-built segments based on long-term behaviour patterns. They are broad by design. They work for awareness campaigns where reach matters more than precision.

In-market audiences are more useful for direct response. These are people Google has identified as actively researching or considering a purchase in a specific category, based on their recent search and browsing behaviour. The category definitions are not always precise, but they signal intent in a way that affinity audiences do not.

Custom intent audiences are where things get genuinely interesting. You build these by entering search terms that your target customer is likely to use. Google then targets people who have recently searched for those terms. If you know what your customer types into Google before they buy, you can reach them on YouTube with video creative that speaks directly to that moment. This is one of the more underused targeting tools available in the platform.

Customer match lets you upload a list of email addresses and target those people on YouTube. Useful for retargeting existing customers or suppressing them from acquisition campaigns. The match rates are imperfect, but the intent signal is strong.

Remarketing audiences built from YouTube engagement, including people who watched a certain percentage of your videos, visited your channel, or interacted with your cards, are particularly useful for building sequential campaigns. You can show a 30-second brand video to cold audiences and then retarget viewers with a more direct response message. The funnel logic is sound. The execution requires planning.

Placement targeting lets you specify exactly which YouTube channels or videos you want to appear on. For niche B2B audiences or highly specific interest groups, this can be more precise than any audience segment. If your target customer watches three specific YouTube channels, you can put your ad on those channels. That is a level of contextual targeting that does not get enough attention.

What Makes a YouTube Ad Creative Actually Work?

This is where most campaigns are won or lost, and where most of the budget goes to waste.

The first five seconds of a skippable ad are not an introduction. They are the entire argument for why someone should keep watching. Most advertisers spend those five seconds on a branded opening, a logo reveal, or a scene-setting shot. By the time they get to anything resembling a reason to watch, the viewer has already skipped.

I have sat in enough creative reviews to know that the people making these decisions often have not watched their own ads the way a real viewer watches them: on a phone, mid-video, with their thumb already hovering over the skip button. Watch your ad that way before you approve it.

A few principles that hold across formats:

Lead with the problem or the payoff, not the brand. The viewer does not care about your brand yet. They care about what is on screen. Give them a reason to care before you ask them to care about who you are.

Design for sound-off and sound-on. A significant portion of YouTube viewing happens without audio. If your ad only makes sense with sound, you are losing a meaningful slice of your audience. Subtitles are not optional.

Make the call to action explicit and early. Do not save the CTA for the end. Put it in the first 30 seconds and repeat it. If someone skips at 31 seconds, they should still know what you wanted them to do.

Test multiple creative variants. YouTube’s creative testing tools are functional. Use them. A single video creative tells you almost nothing. Two or three variants with different hooks, different structures, or different offers tell you something you can act on.

Match the creative to the audience and their position in the funnel. A cold audience needs to understand who you are and why you are relevant. A remarketing audience already has that context. Showing the same creative to both is a creative failure, not a targeting failure.

When I was at lastminute.com, we ran a paid search campaign for a music festival that generated six figures of revenue in roughly a day. The campaign itself was relatively simple. What made it work was that the messaging matched exactly what people were already looking for. YouTube creative works the same way: alignment between what the viewer is thinking and what the ad says is worth more than production value.

How Does YouTube Ad Measurement Actually Work?

YouTube measurement is one of the areas where the most commercial damage gets done, quietly, over time.

The default metrics look good. Views, view rates, CPV, reach. They are real numbers. They are also incomplete proxies for business outcomes. A high view rate tells you people watched. It does not tell you they bought, remembered, or changed their behaviour in any way that matters commercially.

View-through conversions are the specific area where I would urge the most caution. A view-through conversion is recorded when someone sees your YouTube ad (without clicking) and then converts on your site within a defined attribution window. The default window is often 30 days. That means someone who watched three seconds of your pre-roll ad and then bought your product a month later via a Google search might be attributed to your YouTube campaign. That is not wrong exactly. It is just not the whole story.

If you are using view-through attribution to justify YouTube spend, make sure you understand what you are measuring. Look at incrementality where you can. Compare performance in markets where you are running YouTube against markets where you are not. That is a more honest read than a conversion report that includes every tangential touchpoint.

Google’s own measurement tools, Brand Lift, Search Lift, and Sales Lift, are more useful for understanding the actual effect of your campaign on awareness, consideration, and purchase behaviour. They require minimum spend thresholds and take time to generate statistically significant results, but they give you a more defensible picture of what YouTube is actually doing for your business.

For direct response campaigns, track clicks, landing page visits, and downstream conversions with the same rigour you would apply to any paid channel. Connect Google Ads to Google Analytics. Set up conversion tracking properly before you launch. These are basics, but they are still missed more often than they should be.

Resources like the Semrush YouTube SEO study offer useful data on organic performance patterns, which matter for measurement context even when your primary focus is paid. Organic visibility and paid performance are not independent variables on YouTube.

How Do YouTube SEO and YouTube Advertising Work Together?

Most advertisers treat YouTube paid and YouTube organic as separate workstreams. They are managed by different teams, measured separately, and rarely informed by each other. That is a structural inefficiency worth addressing.

YouTube is the second largest search engine in the world. People use it to find answers, compare products, learn skills, and make purchase decisions. Organic video content that ranks well in YouTube search captures that intent without a media cost attached. Paid advertising can accelerate reach and target specific audiences, but it does not replace the compounding value of organic visibility.

The practical overlap between the two disciplines is significant. Keywords that perform well in YouTube search are often worth targeting in paid campaigns. Videos that generate strong organic engagement, high watch time, good click-through rates from thumbnails, are also likely to perform well as paid creative because they have already demonstrated that real viewers find them worth watching.

YouTube SEO involves optimising your video titles, descriptions, tags, and thumbnails for discoverability. It also involves watch time, which is the primary signal YouTube’s algorithm uses to determine whether a video deserves to be recommended. Buffer’s guide to YouTube SEO covers the mechanics in detail, and Mailchimp’s YouTube SEO resource offers a solid complementary perspective on the fundamentals.

From a paid perspective, a channel with strong organic content is a better foundation for advertising than a channel that exists only to host ad creative. The organic content builds trust and context. When someone sees your ad and then visits your channel to investigate further, what they find matters. A channel with 12 unlisted ad videos and no organic content is not reassuring.

I have seen brands invest heavily in YouTube advertising while neglecting the channel entirely as an organic asset. The paid campaigns work in isolation, but they are working harder than they need to. Every pound or dollar of paid media is doing a job that a well-built organic presence could partially offset over time.

What Budget Do You Need to Run YouTube Ads Effectively?

There is no universal floor. YouTube advertising can technically start with a few pounds a day. Whether that generates anything useful is a different question.

For brand awareness campaigns using CPM bidding, you can generate meaningful reach at relatively modest budgets if your targeting is precise. For direct response campaigns optimising for conversions, you need enough budget to generate sufficient conversion volume for Google’s algorithm to learn. Too little budget in a conversion-optimised campaign means the algorithm never exits the learning phase and performance stays volatile.

A rough working principle: if you are running a conversion-focused campaign and your target CPA is £50, you need enough daily budget to generate at least a handful of conversions per week. Below that volume, the algorithm is guessing. Above it, it starts to find patterns.

Production cost is the other budget consideration that often gets underestimated. YouTube ads require video. Good video costs money to produce. If your creative budget is zero, your options are limited to repurposed content, user-generated material, or self-produced video. All of those can work. None of them should be an afterthought.

Early in my career, when I was told there was no budget for a new website I needed, I taught myself to code and built it myself. The outcome was functional, if not polished. The principle transfers: resource constraints are not always reasons to stop. But they do change what you can reasonably expect. A £500 production budget and a £500 media budget will not produce the same results as a properly funded campaign. Being honest about that at the planning stage avoids a lot of post-campaign disappointment.

For testing purposes, a minimum meaningful test on YouTube is probably three to four weeks of spend at a level that can generate statistically meaningful data. What that number is depends on your CPV or CPM, your conversion rate, and your target CPA. Work it backwards from the outcomes you need, not forwards from a number that feels comfortable.

How Do You Use YouTube Ads Alongside Other Video Channels?

YouTube does not exist in isolation. Most brands running video advertising are also running or considering video on Meta, TikTok, LinkedIn, or connected TV. The question of how YouTube fits into that mix is worth thinking through deliberately.

YouTube’s core advantage is intent. The platform is search-driven. People come to YouTube looking for something. That makes it particularly effective for mid-funnel content, product comparisons, how-to content, and anything that benefits from longer engagement. A two-minute product demonstration has a home on YouTube in a way it does not on TikTok or Instagram Reels.

Instagram Reels and TikTok are interruption environments. People are scrolling, not searching. Short-form, high-impact creative that captures attention in the first one to two seconds works better there. The creative logic is different, the audience behaviour is different, and the measurement is different. Repurposing the same video across all platforms without adapting it is a common mistake that produces mediocre results everywhere.

For brands thinking about where to allocate video budget, YouTube tends to perform well for considered purchases where the buyer needs information before converting. Fast-moving consumer goods with impulse purchase dynamics often find more efficient reach on social video. The right answer depends on your category, your customer, and what stage of the funnel you are trying to influence.

Video also extends beyond social platforms. Adding video to email campaigns can improve engagement meaningfully, and using video for lead scoring is an underused application in B2B marketing. YouTube advertising feeds the top of that funnel. What happens after someone clicks matters just as much as the ad itself.

If you are thinking about video marketing as a discipline rather than just a channel, the broader Video Marketing resources at The Marketing Juice cover the strategic and executional questions across the full video landscape, not just YouTube.

What Are the Most Common YouTube Advertising Mistakes?

After managing significant ad spend across multiple industries, the mistakes I see most consistently on YouTube are not technical. They are strategic and creative.

Treating YouTube like television. TV creative is designed for passive, captive audiences. YouTube viewers have a skip button and a choice. The creative conventions of TV advertising, the slow build, the brand reveal at the end, the emotional arc over 30 seconds, do not transfer cleanly to a skippable environment. Advertisers who repurpose TV ads for YouTube often wonder why performance is disappointing. The format demands different creative thinking.

Ignoring the post-click experience. A well-constructed YouTube ad that sends someone to a generic homepage is a waste. The landing page needs to continue the conversation the ad started. If the ad is about a specific product, the landing page should be about that specific product. This is basic, and it is consistently ignored.

Over-relying on Google’s automated recommendations. Google’s campaign optimisation suggestions are not neutral. They are designed to increase spend. Some of them are genuinely useful. Others will expand your targeting, raise your bids, or add placements in ways that benefit Google’s revenue more than your campaign’s performance. Review every recommendation critically before applying it.

Not testing creative systematically. One video is not a test. Running the same creative for three months and concluding that YouTube does not work for your brand is not a valid conclusion. Creative fatigue is real. Audience response varies. Testing is the only way to learn what actually moves your specific audience.

Measuring the wrong things. View counts and view rates are vanity metrics if your objective is revenue. I have seen campaigns celebrated internally because they generated millions of views while the downstream conversion data told a completely different story. Align your measurement to your objective from the start, and be honest when the numbers do not support the narrative.

Underinvesting in targeting refinement. Setting up a campaign with broad targeting and leaving it to run is not a strategy. YouTube targeting requires ongoing refinement: adding negative placements, excluding audiences who have already converted, building remarketing layers, and adjusting bids based on performance data. Campaigns that are not actively managed tend to drift toward the path of least resistance, which is usually not the most commercially efficient path.

How Do You Scale YouTube Ad Campaigns Without Losing Efficiency?

Scaling paid media is where a lot of campaigns go wrong. What works at £500 a week does not always work at £5,000 a week, and the failure modes are predictable if you know what to look for.

The first scaling constraint on YouTube is usually audience saturation. If you are targeting a defined audience with a single creative, you will eventually exhaust that audience’s tolerance for the ad. Frequency rises, performance drops, and the temptation is to increase bids to maintain volume. The right response is usually to refresh creative, expand targeting, or both.

Budget scaling should be gradual. Doubling a campaign budget overnight disrupts the algorithm’s learning and can produce erratic performance for days or weeks while it recalibrates. A 20 to 30 percent budget increase at a time, with a period of stability in between, gives the algorithm room to adapt without losing the optimisation it has already built.

Creative diversification becomes more important at scale. At low budgets, a single strong creative can carry a campaign. At higher spend levels, you need a library of creative variants to maintain performance across different audience segments, different times of day, and different stages of the customer experience.

Audience expansion at scale needs to be deliberate. Moving from a highly targeted custom intent audience to a broader in-market audience to reach more people is a valid scaling strategy, but it changes the nature of what you are doing. The conversion rates will likely be lower. The CPAs will likely be higher. That is not necessarily a failure. It is a different trade-off, and it should be planned for rather than discovered after the fact.

Having grown an agency from 20 to 100 people while managing hundreds of millions in ad spend across 30 industries, the scaling principle I kept coming back to was this: the things that make a campaign work at small scale are almost always the same things that make it work at large scale. Creative quality, targeting precision, landing page relevance, measurement discipline. Scale amplifies what is already there. If the foundations are weak, more budget makes the problem bigger, not smaller.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions