09. Measure and Optimize your Video Advertising Campaign

We ran a similar video campaign for a few days, so now we can evaluate the results. By now, you are pretty familiar with this process as it has many similarities with how you evaluate campaign results for Search and Display campaigns in Adwords. As you will see in the results, this campaign used keyword targeting. I set up two ad groups, I used the same creative (our trailer video), but I targeted using a narrow keyword list focused on education in one ad group, and a large keyword list of over 900 keywords, all focused on digital marketing and digital marketing education in the other ad group.

Campaign Set-Up Summary

As a starting point, let’s recap our marketing objective and the basics for this campaign:

  • Our ultimate marketing objective is to generate interest and conversions for the Digital Marketing Nanodegree. With this video campaign, we don’t expect people to immediately convert and buy the degree, however, we would like to drive visitors to the website. We know that it usually takes a few touch points to push the user further along on the buyer’s journey, and video is a good way to catch people’s attention, generating awareness and interest in the course. If I get people to view the full ad, even if they don’t come and visit, I believe that can still help in building awareness. So, I am interested in the views I get for this ad, as well as the number of clicks.
  • We assumed that we can pay a maximum $0.10 per view while maintaining a positive ROI. As you know, this assumes that I can ultimately convert enough people to buying the degree to make my return on the advertising spent positive.
  • I spent a total of $777 on the campaign (the campaign ran a little over a week, with a $100/day budget

Evaluating overall campaign success

Let’s take a look at the high level results of our campaign.

Evaluating ROI
The campaign generated 27,593 video views and 1760 clicks to the landing page (www.udacity.com/digital-marketing). The cost per view came down to 3 cts, which is considerably less than I was willing to pay. The CTR was 1.19% on average, which is quite high. That leads to a CPC of $0.44. So, let’s see where that leaves us in terms of ROI.

It cost me $0.44 per click that leads a user to the landing page. If I assume a conversion rate of 0.2% (2 in every thousand visitors converts and becomes a student), that means I will pay $220 per conversion.

$0.44/0.2%=$220.

Now, every person who converts pays $999, and let’s assume the profit is $300 per degree, then I can conclude that the ROI on this campaign was positive. That said, paying $220 per conversion is not exactly cheap.

Keep in mind though that I have generated many views with this campaign, and as we indicated before, we have probably touched many users who are at the very start of the customer journey. These users may need a little more time to convert, so there is a good chance that the campaign had an effect that I will only see later in time and that is not measured well by looking at the clicks.

Evaluation of Different Ad Groups

Next, let’s take a look at the performance of the two different ad groups I set up. In this case, I kept the creative constant (same video), but I specified different target groups. Both targets were based on keywords, but in one group “interested in education”, I choose a set of 78 keywords focused on education (with keywords that included ‘education’, ‘course’, ‘training’) and one a group “video DMDN Launch” with a broader target including over 900 keywords that touched on broader marketing concepts (e.g. SEO, Search Marketing etc) and that also included a long list of tail keywords.

While the View Rate was a little bit lower for the “video DMND launch” adgroup at 18.38%, the CTR for this adgroup is considerably higher at 1.41%. As a result, the cost for these ads was also considerably lower at $0.37 per click vs. $0.64 per click for the “interested in education” adgroup.

Given that my targets are similar for both ad groups, but that I used different keyword lists to get to those targets, it is a good idea to pause the adgroup that doesn’t deliver as well, in this case the “interested in education” adgroup.

Fine tuning the keyword list

Of course, as with display campaigns and search campaigns, it is worth it to further optimize within the keyword list that you used to target. So, in our example, I can try to further optimize the performance in the “video DMND launch” group by optimizing the list of keywords for that adgroup.

Two tactics to use are:

  1. Pause keywords with high CPC
  2. Expand the list of keywords with keywords similar to the best performing keywords

Next Concept