20% Lift in Donor Conversions? Yep.

Is this right? This can’t be right.

I read the email again. 

Yup, I’m pretty sure I’m reading this right.

I grab a couple folks for a quick Meet and gut check. “I just got the results for the USO test of Conversion Engine. We lifted donor conversions by 20%.

After a stunned moment of silence, the questions pour out. None of us believe what we just heard. Not only is a 20% lift in conversion rate nearly unheard of for established fundraising programs, it wasn’t supposed to happen. The best case scenario we hoped for was that our version of USO’s donor acquisition flow would perform as well as the control.

You may be scratching your head right now. We weren’t expecting a lift?! Any?! This seems like a good time for me to tell you a bit more about what we were trying to do and learn.

USO’s donor acquisition flow is unlike any flow I’ve seen in the 20+ years I’ve been in this field. It compresses months of email cultivation down to a 5 minute web-based journey. It’s shockingly effective. It was also very difficult for USO to put into practice. Their flow incorporates multiple systems and integrations, all of which were aging. Staff struggled to build out new flows without extensive technical support. This high overhead cut down on the USO team’s ability to innovate and experiment. We couldn’t have that!

Enter Conversion Engine. 

Based on our years of experience on nonprofit giving catalogs – including with USO – we know that landing pages are central to successful fundraising. Like many, we were focused on streamlined, disciplined landing pages, which converted people with as little friction as possible. But USO had another donor acquisition flow that seemed to buck this trend. And not by accident. They had meticulously tested into a flow that was most likely to immediately convert folks who gave their emails into donors. And with a high rate of sustaining donors to boot.  This gave us the starting point for our brand new fundraising platform: Landing pages AND journey building that can’t be done anywhere else. No big.

Soon, we had some very promising software, but it wasn’t yet validated with live production data. And USO had a very effective acquisition strategy, but it required an unruly combination of tools which made it difficult to iterate and spin up new variations. And so our test was born! 

Our only objective was to see how our early alpha version did against their existing system. We identified one of their existing flows used for landing  Facebook ad-driven traffic, and replicated it within the Conversion Engine. To the donor, the control and the Conversion Engine flows were virtually identical; same creative, same sequence of pages, and the same sustainer upsell. But under the surface, the Conversion Engine’s flow was completely different. Since all the user-facing elements were identical, the best we hoped for was the Conversion Engine flow to perform as well as the control. And we were prepared for it to not perform as well, due to the Conversion Engine’s early stage of development. 

Then we absolutely blew the control out of the water. 

But why? How? So many questions!

Turns out that bouncing between systems resulted in enough latency for users that they bailed after providing their email address and, to a lesser extent, when they were directed to the donation form later in the flow. The donor journey on the Conversion Engine is all in one system, where there is significantly less latency between these key moments of the donor journey. System response time is something that we all know is important, but to see the financial implications laid out so clearly was truly remarkable. 

And even more importantly, Capellic now has critical performance benchmarks to measure against when building out new features and flows for our clients. 

Where are things today?

The lift we saw in our first test continues to hold up with other flows where the performance bump matters. Not surprisingly, there was suddenly a lot of urgency from USO to build out as many flows as possible on the Conversion Engine. That’s happening now.

In case you’re wondering, not every test was a success. This is a good thing and critical for us as we learn and prioritize what we build and when. For example, Conversion Engine recently lost (but just barely) a test against a control standard donation form. The culprit? The control had ApplyPay, and ours didn’t (ApplePay was in QA). We also had some secondary copy in a more prominent location than the control did, which probably distracted from the primary donation value proposition. Both of those issues are being addressed now. 

With a few of these remaining updates in place, the Conversion Engine will be in a solid beta state. And that means…we’ll officially unveil it at NTC in April 2023! If you’re interested in the Conversion Engine platform and a program to operationalize development of the ideal donor journeys for your organization, please let us know. We love to work with organizations and agencies who love to experiment and find new ways to improve their fundraising results! What’s the right ask, for the right audience, at the right time, and why?