We'll go through the whole proccess - from setting KPIs through creating Meta Ads to reporting. But firstly, let's look at what was accomplished:

🌝 Albert Schweitzer Stiftung — Germany

please note results are with VAT  (21 %)

→ Budget spent: 14,946 EUR

→ Regular donors acquired: 209

→ Cost per donor: 71.5 EUR

→ 12-month expected income: 20,155 EUR

→ Return on investment: 1.35x (first year)

alright, cool, so both organisations will gain investment from Meta Ads in the first year, that's usually how you measure success of the ads. Which brings us to how both organisations started. So what's the first step?

Defining success before touching Ads Manager

Before any creative or ads were built, both campaigns aligned on two key metrics:

1. Primary KPI
   Cost per donor: not clicks, reactions, or comments. The logic is straightforward. You want to calculate how much will the reuglar donor (acquired by Meta Ads) bring in a year. So you start with your average monthly gift (typically around 10 EUR) and multiply it by ten (not twelve as this already accounts for donor attrition over the year). The result is your benchmark for acceptable acquisition cost - so your investment will be back in a year.
   
   In the ASF campaign, this set the target at 100 EUR per regular donor. In OBRAZ, the target was 40 EUR per regular donor, reflecting lower Meta Ads costs in the Czech Republic compared to the Germany market and a deliberate need to hit ambitious benchmarks to demonstrate that Meta Ads can deliver at scale and justify further investment.

2. Secondary KPI
   Total number of donors acquired in the campaign window. ASF wanted to get to at least 100 regular donors through Meta Ads, OBRAZ set it at 50 regular donors.

okay, so I have my base number - cost per regular donor, I know how many regular donors I want to get, now I need what? The content plan:

Below you can see part of the creatives that were tested in the Obránci zvířat campaign.

We didn’t rely on one idea, but worked with several content principles and narratives in parallel.

The same image was often tested with different headlines, and the same headline was paired with different visuals.

Among the tested approaches were: personal story of a single chicken, strong visual contrast, close-ups of damaged or deformed body parts, detail shots focused on the eye, graphic representations of the space in which chickens live, and variations explicitly built around social proof and urgency. 

The goal of brainstorming like this is not to guess the winner, but to give the campaign enough strong, distinct angles so that both people and the algorithm have something to respond to.

“There isn’t much time left.
949 out of 960 people like you have already become supporters.
We are looking for the last 11 heroes. Help now. Donate.”

We also tested a few video creatives, and this one was clearly the most successful. The reason is simple and very human. It combines several strong signals at once: clarity, urgency, and social proof. The video shows a concrete number of missing donors, which immediately makes the goal tangible, this is not an abstract problem, but something that can actually be finished. At the same time, it clearly says that there isn’t much time left, creating natural pressure to act now, not later. People can instantly understand what is happening, how close the campaign is to success, and what their donation would change.

Also, you can see it's not some big production. Basically just a simple video with a headline. Nothing complicated.

This piece of content generated 7 regular donors with a crazy low cost per regular donor at 12,3 EUR.

oh I am spoilering the results of some creatives already. well so let's skip few steps and already talk about how the content performed...

because I know, you want to know :P and then we'll go back to step-by-step

so what performed well?

It actually got really interesting.

We tested 27 different creative concepts. The performance gap between best and worst was not subtle, it was massive. Some ads brought in dozens of donors. Others, despite significant investment, brought zero.

Let's look at the data.

The clear winners

Three concepts stood out significantly above the rest.

and what didn't work?

okaay what about ASF?

With ASF we tested even more creatives. It was 59... yep, that's right. More than 20 of them brought zero conversions. But the ones that worked, worked well enough to make the entire campaign profitable.

One key difference from the Czech campaign: ASF ran their entire campaign with a matching offer. Major donors committed to matching donations, which gave every creative an extra persuasion layer: "your donation gets multiplied."

This shaped the messaging throughout. While OBRAZ relied purely on emotional appeal and urgency, ASF could add a concrete value proposition: your money goes further.

Like with OBRAZ, we didn't rely on one idea. Multiple narratives and visual approaches were tested in parallel:

First-person chicken voice: the chicken speaks directly: "I can't take it anymore!", "I'm suffering. Please help me.", "I was born to suffer."

First-person supporter voice: flipping the perspective: "I want chickens to not suffer anymore!", "I want to save chicken lives!"

Matching message prominent: leading with the multiplier effect: "Your Euro makes the difference now!", "Your donation + 100€ on top for the chickens"

Urgency + social proof combination: progress bar showing "Only a few days left!" with "Already 625+ supporters joined!"

Graphic suffering imagery: very direct visuals of collapsed, dying chickens

Scale shots: wide factory images showing thousands of chickens

Low-barrier "just 1€" messaging: in the extended phase: "Do you have 1€ left for me?"

The same image was often tested with different headlines, and the same headline with different visuals. The goal was to give the algorithm and audiences enough variety to find what resonated — especially critical when pixel tracking was unreliable.

soo what actually worked for ASF?

Was 59 creatives too many?

Honestly? Probably yes.

Here's the math. To properly test whether a creative works, you need to let it spend roughly 2× your target cost per donor. For ASF with a 100 EUR target, that means ~200 EUR per ad set before you can make a clear judgment.

With 59 creatives spread across multiple ad sets, you'd need a massive budget to give each one a fair chance. In practice, many creatives got killed early or never got enough spend to show their true potential. Some of those "zero conversion" creatives might have worked with more budget, we'll never know.

The learning phase myth doesn't help here either. You don't need Meta to complete its "50 conversions per ad set" learning phase for ads to work. That would require unrealistic budgets for most NGOs. What you need is enough spend to see signal: and 2× your target cost per donor is usually enough to tell if something has potential or not.

My recommendation for most campaigns:

→ Minimum: 10 distinct creative concepts → Maximum: 30 creative concepts → Sweet spot for most budgets: 15-20 creatives

This gives you enough variety to find winners, but not so many that your budget gets spread too thin to learn anything.

The 59-creative approach made some sense for ASF given their larger budget but for most organizations, it's overkill. Start with 10-15 strong concepts, test them properly, then iterate based on what you learn.

okaaay back to the step-by-step. We have the numbers we will be tracking, we have the content ready - now we need... a communication plan!

Above you see part of the ASF year-end communication timeline. It shows, in a very practical way, how the campaign is meant to move forward across social media (organic and paid) and mailing. Thanks to this, you’re not just “posting and advertising”, but you have a sense of rhythm: when the campaign starts, when it gains strength, and when it should feel more urgent.

You can clearly see the moments where it makes sense to push harder, add urgency, or connect multiple channels at once to increase the overall effect. In practical sense: if your ads are not working well in the beginning you can still know, you will have a communication window where you can add urgency to your creatives and lower the price per regular donor.

The plan doesn’t need to be perfect, its main job is to give you direction, so you’re not making every decision on the fly.

okay so you numbers done, content done, communication timeline done... now let's start and set up the ads? Well before that, you need to know, HOW you will be able to tell, if they are successful.

Before any ad goes live: the invisible foundation

Here's something that doesn't show up in the final results, but made those results possible to measure in the first place.

Both OBRAZ and ASF had their tracking infrastructure ready before launching a single ad. This sounds obvious, but it's where many campaigns fail before they even start.

The setup included:

UTM parameters on every single link. Every ad variant had a unique UTM campaign tag that would follow the user all the way to the donation form. Without this, we'd have no way of knowing which creative actually brought which donor.

CRM configured to capture and report on UTMs. The donation system needed to store those UTM parameters. When someone donated, we needed to see not just "donation from Facebook" but "donation from facebook/cpc/post_8".

A reporting sheet updated daily. Both campaigns maintained a simple spreadsheet pulling CRM data by UTM source. Every morning, the team could see: which creatives brought donations yesterday, what's the running cost per donor for each variant, what should we scale up or pause today.

If you're new to this: check a Meta Ads Guide where you'll find more info on this.

Below you can see the logic in how the Excel Sheet can look like so you can clearly see what works:

phew.. right.. numbers → content → comms plan → reporting setup → well, actually now we can finally set up the ads

Finally! Let's talk about actual Meta Ads Manager stuff.

Campaign structure

Both organizations kept it simple. One campaign with multiple ad sets, each ad set targeting a different audience.

Campaign: Year-end 2025 (Conversion objective)

→ Ad Set 1: Engagement (people who engaged with FB/IG page)

→ Ad Set 2: SupporterList (custom audience from email list)

→ Ad Set 3: DonorsList (custom audience from past donors)

→ Ad Set 4: Advantage+ (letting Meta decide)

→ Ad Set 5: Website visitors last 180 days

⚠️ you can see below that now all the creatives were within one campaign/ad set. Especially when you are testing more creatives, you should aim at a maximum of 8 creatives in one ad set.

👏 engagement target audience surprised me - it proved the best for Obránci zvířat.

Okay so everything is set up. Ads are running. Now what?

disclaimer: long-read below. but worth it - spoiler: i had struggles.

Here's what actually happened during both campaigns — the struggles, the pivots, and the moments where it all came together.

OBRAZ: the slow start and the urgency

First 2 weeks, cost per donor hovered around 40 EUR which is GREAT result, but we wanted to be really ambitious and this was our border price. Sometimes we were slightly below, sometimes above. There was no comfortable margin.

The creative with a eye close-up was doing most of the work. It brought consistent results, but at 49 EUR per donor at that time, over-all numbers stayed in that uncomfortable "maybe this will work, maybe it won't" zone.

Then came the final week.

We launched the urgency creatives: "Only a few days left. 621 people already joined. We need 700."

The effect was immediate.

Cost per donor dropped from ~40 EUR to under 20 EUR. Conversion rates jumped. The same audiences that had been responding moderately suddenly converted at much higher rates.

The urgency phase was roughly 30% of total budget but brought disproportionately better results. If we had known this would happen, we might have allocated even more budget to the final days.

The lesson? If your early results are just okay (not great, not terrible) don't panic. The urgency phase can dramatically change the numbers. Both OBRAZ and ASF saw their best results in the final days when deadline pressure kicked in.

It's possible that the weeks of emotional content beforehand helped prime the audience. People had seen the cause, they knew what it was about, and the urgency just gave them the final push. But we can't say that for certain. What we can say is: budget some patience for the early phase, prepare your urgency content in advance, and don't kill the campaign before you've tested what deadline pressure can do.

ASF: when Meta can't see your results

ASF's campaign started with a different kind of problem.

The ads were running. Money was being spent. But Meta Ads Manager showed almost nothing. A few conversions here and there. Cost per conversion looked astronomical: 200, 300, sometimes 500 EUR.

If you only looked at Meta's dashboard, you'd think the campaign was a disaster.

But donations were coming in. The CRM showed new regular donors every day. The UTM tags confirmed they came from the ads for a much lower price.

So what was happening?

The pixel was broken. Not technically broken, it was installed correctly, firing on the right pages. But German users reject cookie consent at very high rates. No consent = no tracking = Meta can't see the conversion.

Only about 10% of actual donations were being recorded.

This created a real problem. Not just for reporting, for optimization. Meta's algorithm learns from conversions. If it can't see conversions, it can't optimize. The ads were essentially running blind.

The pivot: Conversion → Traffic

After about a week of this, I made a decision: switch some campaigns from Conversion objective to Traffic (Landing Page Views).

Why?

Conversion objective optimizes for... conversions. But if the pixel can't see conversions, Meta has nothing to optimize for. The algorithm is guessing.

Traffic objective optimizes for landing page views. This doesn't require the pixel to fire on conversion, just on page load. And that worked more reliably.

The trade-off: → Traffic campaigns reached more people for less money (lower CPM) → But they weren't "optimizing" for donors, just for clicks → So in theory, quality might be lower

In practice? With broken conversion tracking, Traffic campaigns performed comparably. Sometimes better. The lower CPM meant more people saw the ads, which compensated for less precise targeting.

Living without real-time data

The uncomfortable truth about ASF's campaign? I couldn't see what was working in real-time.

With OBRAZ, you could check Meta Ads Manager and get a rough sense of which ads were converting. The data wasn't perfect, but it was directional.

With ASF? Meta showed almost nothing useful. The dashboard was essentially fiction.

So every morning became a ritual:

1. Open the reporting Excel Sheet

2. Update it with Meta Ads spends

3. Calculate cost per donor for each creative

4. Compare to what Meta said (laugh at the difference)

5. Decide what to scale, what to pause

This was the only way to know what was actually working.

It meant decisions were always 24 hours behind. If a creative started performing well on Tuesday, I wouldn't know until Wednesday morning. If something was burning money with no results, same delay.

You have to be comfortable with that uncertainty. And you have to be ready to make dynamic decisions based on incomplete information.

Some days the CRM showed 5 new donors from one creative. Great, scale it. Next day, zero from the same creative. Was it a fluke? Audience exhaustion? Random variation? Hard to know. You make a judgment call and move on.

The insight: you need to be ready for dynamic decisions

Both campaigns taught the same lesson, in different ways.

OBRAZ showed that performance can change dramatically based on timing and messaging. The campaign that looked great but still were on the edge of the target price for three weeks became highly profitable in the final days. If they had panicked and cut budget early, they would have missed the best results.

ASF showed that you can't always trust the platform. When your tracking breaks, you need backup systems and the willingness to make decisions without perfect data.

In both cases, the teams that succeeded were the ones who: → Had tracking infrastructure in place before launch (UTMs, CRM reporting) → Checked real results daily, not just Meta's dashboard → Were willing to pivot strategy mid-campaign → Didn't panic when early results looked uncertain

Meta Ads isn't "set and forget." It's an ongoing conversation between your data, your instincts, and your willingness to adapt.

So, was it worth it?

If you take one thing from this deep dive, let it be this:
there is no single “perfect setup” for Meta Ads fundraising.

What worked here wasn’t magic, hacks, or secret targeting tricks. It was clarity about goals, patience with early uncertainty, and the discipline to let real data guide decisions.

Both campaigns succeeded for different reasons. OBRAZ showed how timing and urgency can unlock performance that wasn’t that visible at the start. ASF showed how campaigns can still work even when platforms fail to show the truth if you build your own way of seeing results.

Let's look at the numbers one more time.

OBRAZ spent 3,450 EUR and got 124 regular donors. At an average monthly gift of ~8 EUR, that's roughly 10,000 EUR in the first year alone. ROI: 3x.

ASF spent 14,946 EUR and got 209 regular donors. First-year income: ~20,000 EUR. ROI: 1.35x in year one. These donors will keep giving in year two, three, and beyond.

Both campaigns hit their targets. Both will be profitable. Both proved that Meta Ads can work for regular donor acquisition, even with a broken tracking, even with uncertain early results, even with 60% of creatives bringing zero conversions.

The key wasn't having perfect conditions. It was having: → Clear KPIs before starting → Multiple creative angles to test → Tracking infrastructure that didn't depend on Meta → Patience to wait for the urgency phase → Willingness to adapt.

If you're considering running Meta Ads for regular donor acquisition: yes, it's hard. Yes, it's uncertain. But with the right setup and the right mindset, it works.

Good luck.