What we learned from testing 15 ads to recruit animal advocacy newsletter subscribers in Hungary

Last year, Farmállatvédelem worked with Campaigns Lab to test a question we had little evidence about: what kind of message in our culture gets people who care about animals to take the first step and subscribe to our newsletter? Most relevant communication research I’ve ever read origins from UK/US/Western Europe, so instead of just blindly taking over what works there, I wanted to test our own culrute’s responses and preferences.

So we decided to run a pilot Meta campaign to test several messagings.

I'm sharing the results because I think they might be useful to other animal advocacy organisations with a similar national culture.

What we tested

We ran 15 ads across four messaging themes on Facebook and Instagram. The campaign spent €1,130.52, reached around 178,000 people and generated 940 leads, at an average cost of €1.20 per lead. 

The four themes weren't different versions of the same message. We wanted to test different reasons why someone might decide to join. Each of the themes included 3-4 different messages.

The four themes:

Direct Ask: directly asking someone to sign up or get involved.

Positioning: testing different ways of framing what participation in animal advocacy could look like and why someone might want to participate.

Target Focus: speaking specifically to the identity or characteristics of the person we wanted to reach—for example, someone who already cares about animals—and inviting them to participate.

Revolution: using more energetic, movement-oriented framing around joining a larger push for a cage-free Hungary.

The results were quite different.

Direct Ask

The Direct Ask messages included things like “join us and help end farmed animal suffering”, “become a digital volunteer”, and an appeal to people frustrated that signing petitions is useless.

Together, these ads generated 293 leads at €1.27 per lead.

We found large differences between the individual messages. The strongest one was:

“People like you who care about animals are already volunteering with us. Want to know what we're planning next?”

That generated leads for €0.86, compared with €1.11–€1.87 for the other variants, and it performed significantly better on several of the metrics we measured.

Positioning

Here we tried messages such as “stand up for animals effectively rather than radically”, “help many animals through small, simple actions”, and “help the animals who need it most”.

Overall, this theme did somewhat better, with 277 leads at €1.02 per lead.

The ‘small-and-simple-actions’ message had the lowest cost per lead at €0.89. However, the differences between the messages in this group generally weren't statistically strong enough for us to confidently say that it would continue outperforming the others if we scaled it.

Something interesting here: “Help the animals who need it most” was much more expensive at €1.66 per lead, but it had the highest reaction and share rates in the group.

That was the first place where it became particularly obvious that the message people engage with publicly isn't necessarily the one that makes them sign up.

Target Focus

This was the strongest of the four themes.

Instead of primarily talking about our organisation, volunteering or the problem itself, these messages spoke directly to people who already saw themselves as caring about animals.

They included the messages:  “you don't need to be the loudest person to help animals”, “you're not alone—we can help animals together”, “if you love animals, turn that concern into action”, and “people who love animals from across Hungary are joining our campaigns—how about you?”

As a group, these ads generated 128 leads at €0.76 per lead. 

One message stood out in particular:

“People who love animals are joining our campaigns from every corner of the country. How about you?”

It generated leads for €0.46 each, while the other three Target Focus messages cost between €0.97 and €1.08. 

But again, it wasn't the winner on every metric. “You don't need to be the loudest to help animals” got a higher reaction rate (2.83% vs. 2.35%) and share rate (0.48% vs. 0.37%).

Revolution

Finally, we tried a more energetic language about joining a “cage-free revolution”, breaking out of apathy and achieving a cage-free Hungary together.

This was the most expensive of the four approaches for generating leads: €1.57 per lead.

The best individual version—roughly, “The cage-free revolution starts now. Will you join?”—cost €1.22 per lead. The others ranged from €1.63 to €2.01. 

So overall, this is what we saw:

The cheapest broad approach therefore generated leads at below one dollar, 2/3 of the price of the second best.

One thing that surprised us: engagement and sign-ups didn't always go together

This is probably the part of the experiment that I find most useful beyond the individual winning ad.

Our metric was signup per dollar, but if we'd judged the ads mainly by likes, comments and shares, we could have come to quite different conclusions. Some messages that did particularly well on engagement were relatively poor at generating newsletter sign-ups, while some of the best lead-generating messages weren't the engagement winners.

We also found that Facebook generated leads more cheaply than Instagram (€1.19 vs. €1.41), and Facebook Feed alone accounted for more than 80% of the leads.

There is also a fairly big caveat to everything I've written above: we've only measured the first step.

The 940 people we recruited are newsletter subscribers. They are not active supporters or volunteers.

And this is not out ultimate goal. What matters is what happens afterwards. We will use the results of this pilot to gather more leads and see how many newsletter subscribers go on to take digital campaign actions, become volunteers, or donate.

There are other reasons to be cautious too. This was one campaign, run by one organisation, in Hungary. Some comparisons were statistically significant and others weren't. Where they weren't, we can't confidently say that an apparent difference between two messages would persist if we ran them again.