Evaluating Marketing Campaign Effectiveness Through Data
Marketing used to be an art. These days, it’s becoming more like a science.
One of the tasks I’ve done as a marketing data analyst intern in my current position is reviewing/measuring campaign performance. Campaign managers have campaigns that they are running. My role is to go into Google Ads and Google Analytics to help review and optimize campaigns through data. Today, I will share some of the metrics I typically look at when analyzing campaigns, how I interpret them, and the feedback I would give to the campaign managers.
Google Ads
Ad Impression:
impressions refer to the number of times an ad is displayed to the users. If the impressions are low, it means the ad is not reaching enough people. We can check if our keywords have good volumes behind them. If not, we can expand the keywords by researching on SEMrush. We could also try increasing the bid amount.
CTR (Click-Through-Rate):
If an ad has a low CTR, that means whatever search keywords we are using and the ad we’re putting out there are considered irrelevant to the users. The ad copy is unattractive and doesn’t speak to the users, so they will not click on it. We need to consider revising ad copy/messaging to make it more compelling and relevant to entice the target audience to click on the ad. We could also optimize the keywords we use by removing irrelevant ones and adding more relevant ones.
Bounce Rate:
Bounce rate refers to the percentage of visitors who navigate away from the website after viewing only one page. A high bounce rate could indicate that the landing page is not relevant and engaging to the visitors, or there could be technical issues with the website hindering user experience. Typically, a bounce rate above 60% for search ads is a red flag for me; 80% for display ads.
Conversion Rate:
We look at the conversion rate to measure a campaign’s effectiveness in achieving the desired outcomes or conversions, such as filling out a form, subscribing to a newsletter, or downloading our white paper. The benchmark for every company and industry is different. At my company, we typically want 1% (but would love to have 2%). With all the traffic that comes to the website, we have to be converting at a 1% rate to meet our revenue goal every year. That’s just the standard for our team.
Ad Assets:
Ad assets refer to the various components of an ad, such as headlines, descriptions, images, and videos. For example, if a particular headline or image is performing well, we may choose to use similar elements in future campaigns. On the other hand, if an ad asset is performing poorly, we may choose to remove it or test a different variation to improve results.
Cost:
We typically have a set budget for a campaign. We need to monitor how much an ad cost over a period of time and ensure it doesn’t exceed our budget. If it does, we must stop the ads or ask for more budget if it’s performing well.
Take the below Search Ad as an example:
We are converting at 3.8%. The cost is right on track (depending on your budget). The bounce rate is super low at 12.70%. With that, the keywords, the ad messaging, and the landing page are all working well together. Why do I say that? because our CTR is pretty good, and the bounce rate is low. The conversion rate is good as well. I will tell the campaign manager that this is a great ad so leave it as is.
Google Analytics
Above, we were just looking at the performance before users reached our website. We were just making sure that our ads are serving and nothing weird is happening. Now we will go to Google Analytics to see the user behavior on our site. What’s happened once they got on the page, is there anything we need to react to as far as how they are consuming the content? or any of the audiences are performing particularly well/poorly?
Source/Medium:
Source refers to the website or platform that referred the visitor to the website, while medium refers to the source category, such as email, social media, or search engine. It helps to identify where the traffic is coming from and how visitors find the website.
Users:
By understanding how many unique users we have, we can get a sense of the reach of our campaign and the potential size of the target audience. Additionally, by comparing the number of Users to other metrics such as Sessions and Pageviews, we can gain insights into the engagement and behavior of the audience on our website or within our campaign.
Pageviews:
It provides insights into the number of times users have viewed a particular page on a website, which can indicate the level of interest or engagement with the campaign.
Page/Session:
Page/Session indicates the average number of pages a user visits during a single session on the website. Generally, A higher Page/Session value suggests that users are finding the website content relevant and engaging and are exploring the site more thoroughly. However, on the flip side, it could also mean they are not finding the information they’re looking for, continuing to search and search because we are not answering their question.
Bounce Rate:
Explained above. Again, we typically want the bounce rate to be lower than 60% for search ads and 80% for display ads for my team.
Conversion Rate:
Explained above. Again, we typically want to have at least 1%, but different companies have different benchmarks.
Company/Visitor Balance:
If the campaign managers are targeting a certain audience, it’s interesting to know who is on the page.
Next Page:
Looking at what pages users are going to next can help identify if users are finding the information they are looking for on the website or if they are encountering roadblocks in the form of dead-end pages or confusing navigation.
Take the below landing page as an example:
A huge bulk of the traffic (94.35%) comes from our Google Paid Ads (google/cpc). That means the Google Ads strategy is working for the landing page with 5900 users and 14758 pageviews.
The bounce rate (23.79%) is super low. That means the messaging and the ad are working really well. *Remember to also look at how many users there are to see if the conclusion is viable. Every page is different, and everything going into the page is different.
All-encompassing, we are getting a lot of traffic, but the conversion rate (0.5%) is not really there. If generating leads (filling out the form) is the goal of this landing page, the feedback I would give to the campaign manager is - the ads are working because traffic is coming from the ads, and it’s a good user experience because it has a very low bounce rate. However, they are not necessarily converting, so something is missing on the page.
Now we can think: Usually, in the first touch point, users are not ready to buy the products or talk to a salesperson yet, so they are not filling out a form. They may need more information to help them make a decision and convert. Should we add a video on the page that gives them educational content? Do we actually not want them to convert on this page? Could we design a user journey where we land them on an education page and then lead them to a converting page? Ultimately, it will be up to the campaign manager to decide what changes happen on the page. Here, analysts are just providing insights about if the ads are working, getting traffic, and if they are converting.
Brief Conclusion
The big thing about being an analyst is taking the numbers and telling the story. When the ads are getting good traffic, what does that mean? Is it because the keywords have good volumes behind them? Do the ads themselves speak to the people? When they get to the landing page, did they stay on? There is something that users like about the page, but they don’t want to make that final push to fill out a form, which is an ultimate goal. Now, why is that?
Sometimes the page can be perfect, and they are not converting; maybe users are simply not ready yet. Our product (we sell software) isn’t cheap, it’s a big investment for them, so they need to know more than one page to decide if they want to go further. We dig in by looking at the data to figure out those puzzle pieces, put them together and reason through it.
