We’ve updated this case study to be a little more digestible and the link is here. The version you’re currently reading combines two case studies and a much longer narrative. My suggestion? Read this one yourself and send the shorter version to your supervisor. Of course, before you do any of this, enter your email in the field at the bottom so that we can keep in touch.
@Mssg increases the results from Facebook Ad campaigns by sending the clicker into an automated conversation in Messenger – rather than sending the clicker directly to the landing page. In this case study we look in detail at how the Democratic Coalition Against Trump used @Mssg and Facebook Messenger to increase the number of donors and total donations per Facebook Ad dollar.
Executive Summary: Facebook Ads that send the clicker into Facebook Messenger can increase campaign results significantly. In this test it cost $12.56 – $25 to acquire a donor when they are routed directly to a landing page, and only $4.84 when the clicker routes to a conversation on Messenger. On an ROI basis, every dollar spent sending a clicker to a URL resulted in $1 – $1.99 of donations. Every dollar spent sending the clicker into a conversation resulted in $3.10 of donations.
Additionally, when the clicker is sent to Messenger the campaign generates much more supporter data – like email address. For every dollar spent on ads, the Democratic Coalition added 3 new email addresses to their list. Routing the clicker to the URL did not generate any emails accept for the emails of donors.
Methodology: This case study reviews two separate test happening mid-February to mid-March 2017. The first test was designed to find best practices for engagement and data collection over the messaging channel. The second test compared the results when a clicker is sent into a Messenger conversation vs. a landing page.
Each test begins with a single Facebook Ad Campaign and one Ad Set. Within that Ad set, 3 identical ads are created and these ads are all shown to the same targets inside Facebook. The goal is that the same group of people see the same ad at the same time. The difference that we are testing for is what happens after the ad is clicked.
The only difference in the ads, is that ads that route the user to Messenger tell the user that the action will happen in Messenger and the call to action on the Ad Button is “Send Message”. Ads that don’t link to Messenger do not mention Messenger and these ads can use any other call to action on the Ad Button.
Test #1: Starting a Conversation
With Messenger as a Destination Ads, when the user clicks they are redirected into Messenger instead of a landing page. When the user arrives in Messenger they are sent a Welcome Message, which is part of the Facebook Ad. Once the user engages with this Welcome Message, only then can the automated conversation start. Essentially with Messenger as a Destination Ads, the user needs to click twice – once on the ad and once on the Welcome Message.
The first test was designed to find best practices for Welcome Messages, measured by the percentage of people that engage with the welcome message after receiving it. The conversation rate is calculated as the number of people starting a conversation divided by the number of people that clicked on the ad, and presumably received the welcome message.
After the user engaged with either Welcome Message, the conversation would be the same. The user is asked for their email address, address and donation amount before the donation page is opened in Messenger and all of their information is auto-filled into the form.
Results: Messenger is a new space for all organizations, so discovering best practices can be really exciting and makes a huge difference for an organization’s results. Conversation rates were better when the Welcome Message is a structured message that includes an image, title, subtitle and buttons/choices. Asking the user to click their choice in a multiple choice setting works better than asking for a free form response. Furthermore, a simple, obvious, almost rhetorical question seems to work better than asking the user to actually make a decision. As can be seen in the winning Welcome Message, anyone targeted by this Facebook Ad is likely to respond in a very predictable way.
The best performing Welcome Message saw a 74% conversation rate. For every 100 people that clicked the Ad, 74 people started a conversation. In the first message of the conversation, the user was asked to respond with their email address. 60.5% of the people that started the conversation gave their email address. This math’s to a 46% conversion rate for email address for all ad clickers.
This first test helped us understand some techniques for getting users to engage and our first benchmark for data collection once the conversation started.
Test #2: Do Messenger Conversations Increase Donations?
The second test measured how sending the clicker into a conversation compares to sending the clicker to a landing page measured by donations and data collected from the clickers. Similar to the first test, this was a single campaign and ad set with nearly identical ads. The difference is where the user was taken after they clicked the ad.
The control group was taken directly to the donation page upon clicking the ad. The test group routed into a conversation where they were asked for their email, address and donation amount before the exact same donation page was opened. After the conversation, when the donation page was opened, @Mssg auto-populated the information we had on the potential donor.
The results are extremely exciting for Messaging as a channel to drive donations.
Comparing Conversations to URLs:
In this test, sending the user to a URL worked very well compared to traditional benchmarks. That’s to be expected given the current political situation and fundraising environment for Progressive Organizations. All that being said, the potential of sending clickers into a Messenger conversation is absolutely enormous. Organizations have been mobile optimizing pages for 5 years. After only a handful of tests, sending the clicker into Messenger was able to produce results 250 – 500% better than sending the clicker to the page.
This is just the beginning of conversations as a medium for data collection and donations. With more testing and optimization of this process results can improve further.
It’s also important to note that the partial data (specifically email) is valuable from a fundraising perspective. Emailing a fundraising ask to the people that gave an email, but did not complete the donation will result in an even greater ROI for the conversation tests in this campaign. This is an area that needs more testing.
If you do Facebook Ads and are looking to increase your ROI, we’d love to chat. Simply enter your email below and we’ll follow up. It’s also possible to email us directly at firstname.lastname@example.org