Analytics for Video Ad Serving

Daniel Hoelbling-Inzko
. 8 min read
- Bitmovin

- Bitmovin
Advertising is a controversial topic for most people, but the reality is that today’s internet wouldn’t be the same without advertising and how it impacts our daily lives. 
Compared to traditional advertising, online advertising has huge benefits for advertisers and consumers alike. Targeting a specific audience to ensure your ad dollars are wisely spent is a huge deal. So instead of having a large audience watch the same ad on linear TV and just hoping that your intended target audience is watching, targeted online video can specifically target just a few people who are most likely to convert and who may also find value in the ad being shown to them. A better experience all around.
When advertising works it’s a win-win for both advertiser and consumer. But the unfortunate truth is that video experiences on the internet are sometimes less than stellar. When we do fall short of that ideal goal of relevant, technically flawless presentation, customers not only get really annoyed with the ads but also with the whole experience.

The Key to Successful Ads

Imagine yourself, spending 10 seconds staring at a blank screen while the ad servers try to figure out what ad to display, just to then finally see a barely working ad presented in terrible video quality. Chances are the actual content following that experience starts out at a pretty big deficit.
Making sure that ads presented on your platform are working, presented in a timely manner and relevant to users is a huge deal nowadays. Better performing ads also tend to generate more conversions, so knowing which ad systems and ad types work best for your platform is also hugely beneficial.
So keeping an eye on your ad performance is quite critical, especially if big parts of your revenue depend on video advertising. Unfortunately, in most cases you’ll be limited to the tools your ad server provides you, and even worse you’ll have to check multiple sources to get an accurate picture of what’s going on. Add in the inherent complexity of online video and advertising workflows and you quickly get overwhelmed.
Bitmovin Analytics is built to give you visibility into your video workflow, show you actionable intelligence about your platform health, and make your life easier.
Adding in advertising analytics was a logical next step. Our initial thoughts were that while video ads are more constrained, they are still in essence videos and the technical challenges are largely the same. Or at least so we thought.

Understanding Ad Performance in Real Time

We were wrong – not about the video ads being essentially videos, but more about how much cool stuff you can measure about video ads and their performance. So it took a bit longer than expected, but we’re really proud to announce our new Analytics Ads Dashboard that just launched in our Bitmovin Dashboard.

Try for Free
So what did we come up with? First off we had to cross off all the standard things you’d expect to see in your analytics solution: Number of ad impressions, completion rates, and click-through rates (CTR).

- Bitmovin
Graph showing the completion rate distribution over time

But where it gets really interesting is with our ability to break down all of these metrics and slice and dice it to get to exactly the data you want to see.
So you could, for example, look at completion rates overall based on country of origin in relation to ad technologies used. Or whether or not users actually saw the ad (time in viewport), if you see this becoming a problem you can start making changes to the design of your site to avoid this.
- Bitmovin
Completion rate, broken down by ad-position

Maximizing Ad Monetization

Another thing which obviously is hugely impactful is looking at all the reasons why some impressions were NOT monetized. It can be very helpful to see how many times your audience failed to load an ad SDK, or how many times the ad server returned an invalid/unparsable manifest. Especially if you are running your own internal ad server that waterfalls to external ones.

Optimizing Ad Waterfalling

Another obvious target for optimization is the ad click rate as well as the position the ad was clicked on. You obviously want to prioritize ads that have high click-through rates and thus generate revenue for you, while resorting to ad servers that have very low click-through rates only at the very end of your chain. The more ad servers are involved in your ad waterfall the longer it usually takes for the user to load. So optimizing ad waterfalling is a key factor in improving the user’s experience on your platform.
But even just breaking down click rates by ad position can also be a huge help. If your audience responds much better to mid-rolls then why would you take the hefty performance penalty of doing a pre-roll before you were able to pique their interest?

Get Started with Bitmovin’s Ad Analytics

Using the new ads feature is quite simple and requires no special configuration. Once you login to your Bitmovin account, you just have to deploy the Analytics collector to your website and deploy the new code. No special configuration is required, aside from the Bitmovin analytics configuration.

Try for Free
If you are already using Bitmovin Analytics you simply need to upgrade your analytics collector to version 2.1 (and Bitmovin Player v8.1) and you should be all good.
Once data shows up you just head over to our Dashboard and click the Ads to explore all the new metrics coming in about your audience.
Obviously, this is just the beginning. Like with every product, we have built a flexible API that allows for a lot of use cases we might not have even thought about. Give it a try and let us know what you think. We can’t wait to see what cool use cases this will enable for users!

Daniel Hoelbling-Inzko

Daniel Hoelbling-Inzko

Daniel is an Infrastructure Architect at the video infrastructure software company, Bitmovin. Daniel is responsible for the design and operations of the encoding infrastructure, enabling complex video workflows for companies like Red Bull Media House, New York Times, Sling TV and others. Daniel has been driving Bitmovin’s migration to containerized architecture on Docker and Kubernetes.

Related Posts

- Bitmovin

Creating a Live Streaming Setup with Android Devices and Bitmovin

Join the conversation