Making an effective YouTube ad is no walk in the park, you’ve got to make sure you get your targeting right, select the appropriate ad type(s) and campaign structure, not to mention, above all, getting your creative absolutely perfect. Thankfully, Google has created the ABCD guidelines for creating effective YouTube ads, which can come in handy when working on a new ad – but can still be a little hard to check and verify if you’ve correctly put them into place.

In this article, we’re going to take a closer look at the ‘B’ of the ABCD guidelines – Branding and how it can be measured with YouTube’s Brand Lift metrics. According to Google, the key to branding success in a YouTube ad is to brand “early, often and richly”. This approach absolutely makes sense, as verified in our neuromarketing research studies with Alpha.One. However, if you don’t have clear benchmarks – it can be hard to know just how early, how often and how richly.

Of course, showing your brand is important – but how can you do it in a way that is subtle yet effective? So that you don’t overwhelm your audience with the presence of a logo, but are still able to drive brand equity through improved brand recall.

This is where Junbi.ai comes in – we have developed a proprietary score for predictively measuring attention to your brand in a YouTube ad – which we call the brand attention score. Through early studies, we have found that there is a strong positive correlation between improved brand recall and a higher brand attention score on Junbi.ai.

What is the brand attention score on Junbi.ai?

The brand attention score is a measure of how much visual attention is paid to your brand at each moment of the ad.

We calculate the amount of visual attention by first creating an attention prediction heatmap of your video ad. This is a custom-built and trained neural network, trained with real eye tracking data, that can predict visual attention on any image or video with 95% accuracy in a matter of seconds (or a few minutes for videos). So you can think of this a like an eye-tracking heatmap, but without the need for participants.

The predicted attention to your brand is then measured on a frame-by-frame basis throughout your ad, as our custom AI models search for your brand and/or logo in each frame, then use the data from the attention prediction to quantify the amount of attention that is overlapping with the area where the brand has been detected. This is visualised in the example below where the ‘Lancôme’ brand has been detected in a frame of a YouTube ad featuring Zendaya.

Why do we measure brand attention on YouTube ads? It’s actually a very simple answer – if your viewers don’t see your brand, how will they know the ad is from you? If you’re trying to grow your brand awareness and brand equity as an advertiser, it’s essential that you don’t just include your brand in your ad in the bottom corner, but make sure that your audience is really noticing your brand and paying attention to it.

On top of the second-by-second score, the Junbi.ai brand attention score will also be displayed as a weighted average score, normalised to our database of thousands of real YouTube ads, so you can not only see if your brand is being noticed, but you can see how your branding weighs up against the competition. As seen in the Budweiser ad below, seems that the King of Beers is also the King of Branding with a normalised score of 82!

Why should you pay attention to brand lift?

So we’ve mentioned how important it is to get your brand noticed. But what is the tangible impact of this? How can you show your boss that your Junbi-optimised ad has led to a measurable positive impact in your YouTube advertising?

That’s where brand lift comes in. For those of you unfamiliar with Brand Lift, it is a free tool provided by Google to measure the effectiveness of video campaigns related to building brand equity and brand awareness.

Brand lift is available for in-stream and bumper ads, and includes metrics such as ad recall, brand awareness and consideration – as opposed to traditional metrics like clicks and views. This is particularly important as traditional metrics can serve as good indicator of how many people have seen and engaged with your ad, but does not shed much light on the effectiveness of your campaign when it comes to awareness and consideration

The data for the brand lift metrics is collected through surveys which are shown to both YouTube users who have seen your ad, and users who did not. From this data, YouTube will provide advertisers with aggregated brand lift metrics including (but not limited to) # of lifted users, absolute brand lift, relative brand lift and more. This will help you to answer questions like how many people remember watching your ad, and if they are more aware of your brand after having seen your ad on YouTube.

With this knowledge, you can better understand the dynamic perceptions of your target audience in a matter of days, so you can even adjust and optimise your campaigns mid-flight to get the most out of your ad spend.

How can you measure attention on your brand in Junbi.ai?

Getting your brand noticed is important, and now advertisers have the brand lift metrics available to measure real-world impact and effectiveness. But how can you use Junbi.ai to better predict which ads will have a higher brand lift? Let’s go through it step-by-step.

Step 1: Upload your ad

Simple as that, just upload your ad to junbi.ai, give your prediction a name and select which brand you want to test for. This is an important step, as the brand you define here will be the brand that our predictive models search for in the ad and will use as the basis for the brand attention score.

Step 2: Select a viewer mode

To ensure your ad is tested in a realistic context, Junbi.ai provides advertisers with the option to select a viewer mode to test in. This is very important for defining one of our other metrics – the ad breakthrough score. Currently we support 3 viewer modes: Desktop, Desktop Theater and Mobile (Stay tuned for updates on CTV and shorts)!

Simply select the viewer mode you wish to test for, and click next.

Step 3: Select an ad type

Select either skippable in stream, non-skippable in stream or bumper ad. This will determine which sample of ads your ad will be benchmarked against, as we only compare ads of the same type to ensure the most accurate results. We’ll automatically check the length of your ad, and if your ad doesn’t fit the length requirements for certain ad types – those options will be greyed out.

Step 4: Create your prediction

Once you have entered all the details about your ad, simply click the next button and let Junbi.ai do its magic! Your result should be ready in 5-15 minutes, depending on the length of your ad.

Step 5: Learn what works and optimise

Once your results are ready, your output will be visualised on the results page. For brand attention, the most important thing to look out for is the yellow line on the ‘effectiveness drivers over time’ graph. This visualises on a second-by-second basis how well your brand is being noticed on a scale of 0-100%. As you can see below in the Budweiser example, this ad does a great job of getting attention on the brand, particularly at the beginning and end of the ad.

This is actually an excellent example of what to do right, especially because this ad is a non-skippable in stream ad, meaning that the viewer will see the brand right at the beginning, in the middle and the end of the ad. This is the ideal way to include your brand in an ad, thanks to a behavioural heuristic known as the primacy and recency effect, meaning that people will remember what is shown at the beginning and end of the ad the best.

To accommodate this behavioral heuristic, Junbi’s brand attention score will place a greater weighting on the branding score in the first 2 seconds an last 5 seconds of the ad – which is why this Budweiser example has received an excellent rating for brand attention!

Tips and best practices

Obviously there is no one single best way to include your brand in an advertisement. You do not exactly have to copy what Budweiser has done in the example above, but we do have some general ‘rules of thumb’ that serve as guidelines for improving branding effectiveness:

• Don’t rely too much on placing your brand in the corner. This can help to get some additional attention on your brand, but it doesn’t get noticed nearly as much as when you show your brand clearly and centrally in the ad.
• Show your brand early! It’s never too early to show your brand in your advertisement, regardless of the ad type and campaign goals. The last thing you want is for viewers to get confused about who the ad is from. You might end up boosting the results of your competitors if you’re not careful.
• For non-skippables, get a lot of attention on your brand in the last 5 seconds of the ad. Since viewers cannot skip the ad, this will be the last thing they see before their video starts. Make sure your brand is part of what they remember.
• Be creative! We’re not trying to encourage all creators to follow a set formula for showing the brand. You can show the brand on a product, on clothing, on a building – it doesn’t matter, as long as it’s in there and it’s getting attention.

That brings our brief guide on using the brand attention score to boost brand lift on YouTube to an end. If you’re interested to learn more about Junbi, feel free to book a demo with one of our specialists or use our free ad testing tool to test your own YouTube ad.

Happy testing!