MCSMN Blog

Share this:
October 29th, 2015

Demonstrating ROI of Social Media Ads for Events

By MakalaArce

“What’s the ROI?” is a question social media advocates are often asked by those allocating resources. Can social media platforms truly drive bottom-line results?

Preparing for September's first-ever international Healthcare and Social Media Summit in Australia provided us an opportunity to track and demonstrate the financial benefits of advertising on social platforms.

We started by defining our audience to include physicians, nurses, communicators and marketers. We theorized that by advertising the Summit on key social media platforms, we’d reach those who'd be interested because they were already using social media.

To be most effective, we needed to optimize two factors:

  • The performance of our ads on each platform, and
  • The mix of advertising, increasing ad spending on the platforms that performed best.

At the beginning, we used native platform metrics to optimize ad images, text, and headlines, testing different versions to see which ones led to the most clicks. After the first week, the cost-per-website click (CPC) on Facebook was $1.36, while we were paying $3.71 on LinkedIn and $1.79 on Twitter.

So we should have increased spending on Facebook because it was giving us the most clicks for the money, right?

Not necessarily, as the second part of our analytics journey showed. While we had optimized ads to maximize clicks to the Summit registration site, we couldn't yet show which clicks, if any, had led to registrations. In other words, what if the more expensive clicks were really a better value, because a higher percentage of them converted to a paid registration?

We hadn't yet completed the technical work that would enable us to track registration conversion rates by referral source. So, we then added Google Analytics tracking using UTM (Urchin Tracking Module) parameters differentiating key variables we wanted to compare, and implemented tracking codes on the event site. (To learn more about custom UTM parameters, visit Google’s support page.)

By including a unique UTM parameter for each of the three platforms within our ads, we learned via Google Analytics that our social media ads led directly to 18 registrations from these platforms:

  • Facebook: 7
  • LinkedIn: 9
  • Twitter: 2

Sample Ads

The overall cost per registration (CPR) through social platforms was approximately $400, which we considered worthwhile given the $880 in revenue resulting from each. The CPR for LinkedIn was about half of that for Facebook, with Twitter a distant third.

LinkedIn was the clear winner for our professional audience, so we shifted a higher proportion of our advertising spending to it. Had we just looked at the first half of the analytics picture (the CPC), we would have incorrectly reallocated more spending towards Facebook and away from LinkedIn. We wouldn’t have known that Facebook clicks weren’t yielding as many registrations.

We also wanted to compare effects of geotargeting. Campaign 1 broadly targeted Australia (and included New Zealand on LinkedIn). Campaign 2 focused on cities closer to Brisbane (e.g., Coffs Harbour, Gold Coast, Sydney).

Targeting more narrowly on Facebook more than doubled the cost per registration, while the same focus on LinkedIn cut the CPR by more than 60 percent. And while LinkedIn was the overall platform winner with an average CPR of $250, once we honed in on a more targeted audience, the average CPR decreased to $180.

Another key take-away: the more targeted (Campaign 2) click-through traffic to the event website from all three platforms had longer average session durations (time spent on the site), which indicated a higher level of interest.

In the end, the ROI resulting from our 18 registrations directly attributed to advertising was at least 113 percent.

Finally, because of indirect effects and because we weren't able to track clicks all the way through to registrations at first, it’s important to note that this was our minimum advertising ROI.

Our post-Summit survey asked registrants how they learned about the event and 32 respondents credited a social platform, 14 more than we'd tracked directly. Some probably saw organic social media posts, while others may have become interested because of our ads, but registered on a different device. For example, we would not have captured someone who saw a Facebook ad on mobile and later registered via desktop. Survey Results

And while Twitter had the lowest number of registrations we could prove as coming from ads, more survey respondents cited it as playing a role than Facebook and LinkedIn combined.

Measurement really only matters when it helps us make a decision and change our actions according to the data. In this case, it clearly did. UTM tracking made it possible to continuously measure ROI for these campaigns and adjust spending based on what we learned.

We’re excited to further apply these analytics tools and continually refine our tactical use of advertising on social media platforms.

Tags: advertising, Case Studies, metrics, Metrics & Analytics, ROI, Strategy, Tactics & Best Practices

Liked by Ben Forstie, Lisa Wylde, Andrea Kelley, Dan Hinmon, MCSMN Director

Comment


Chris Boyer
@chrisboyer

Posts: 105
Joined: Feb 07, 2011
Posted by @chrisboyer, Oct 31, 2015

Makala, when you say ROI was 113% what exactly do you mean? Can you share the formula that you used to hep us understand? Also, post event “how did you hear about us” surveys are very suspect, as they are anecdotal at best. How would you suggest organization’s track multi-channel ROI brought by social media (and other marketing tactics)?

Liked by Gina


MakalaArce
@MakalaArce

Posts: 71
Joined: Jan 31, 2011
Posted by @MakalaArce, Nov 3, 2015

ROI = Return on Investment. This is calculated by taking the profit (revenue subtracted by cost), and divide that by the cost. This number is converted to a percentage. For our specific case, we calculated the profit by starting with the total intake of money resulting from the advertisement registrations and subtracting our total spend on advertisements. Then, we divided this number by the total spend on advertisements.

To clarify, our survey was done at time of registration, which would likely make it easier for registrants to remember. That said, I agree that surveys are anecdotal. The question on tracking multi-channel ROI is a big one. In our case, we added “&utm_source=” differentiators in the tracking code to distinguish whether the traffic was coming from Facebook, Twitter, or LinkedIn. There is a link to a free builder in the blogpost for actually building these. Additionally, I think tracking multi-device ROI is challenging. Approaches and challenges to this are briefly introduced here: https://www.ftc.gov/news-events/events-calendar/2015/11/cross-device-tracking.

Liked by Lisa Wylde


Lisa Wylde
@Lisa_Wylde

Posts: 54
Joined: Jun 08, 2011
Posted by @Lisa_Wylde, Nov 3, 2015

Thanks for this analysis Makala. I think it’s important to remember that as this was an inaugural event here in Australia and we didn’t have the tracking set up right from the start, we may have missed some people out of the final analysis. It also doesn’t take into account people who may have registered after reading LinkedIn posts made by External Advisory Board or Mayo Staff members. I know anecdotally of a couple of people who registered after reading one of my posts rather than seeing the LI advertsing. But it’s great to know you’ve got the tracking set up now to help us promote MayoInOz16 to a bigger and better audience and monitor where they come from.


MakalaArce
@MakalaArce

Posts: 71
Joined: Jan 31, 2011
Posted by @MakalaArce, Nov 4, 2015

I agree! The above tracking doesn’t include organic results, nor any registrations before tracking was set up. That’s great to hear about those anecdotal examples. Thanks for sharing!

Please login or register to post a comment.

© Mayo Clinic Social Media Network. All Rights Reserved.