Women's Refuge

Safe Night is Women's Refuge's best-known website, supporting people escaping family violence. Until this appeal, there was 1 way to give: a direct donation, either a single gift or a regular one.

That worked, and it capped growth at how many people the charity could reach itself. Every donor had to find Safe Night first.

For an annual Winter Appeal that has to grow each year without a matching increase in paid media, that ceiling was the problem to solve. The limit sat in the number of people the charity could reach, so that is where the work went.

The appeal Women's Refuge wanted to build was a campaign named 'It Takes Seven', named for the fact that it takes a woman experiencing family violence an average of 7 attempts to leave. Supporters would sign up to a physical challenge and ask their own networks to sponsor them, so each one becomes a small fundraising channel.

Women's Refuge had never run a campaign of this kind, so there was no record of how many supporters would take it up, which channel would recruit them, or what a fundraiser does once signed up.

The campaign was also launching on a new domain with no traffic history and no social proof. Every visitor arriving for the first time had to sort themselves into 1 of 3 paths on a single visit: sign up to fundraise, sponsor a friend or donate. The appeal ran to a fixed calendar window, so there was no room for a slow first month.

Key challenges

Women's Refuge had never run a campaign like this. Nobody knew whether supporters would fundraise, or what it would cost to sign them up. That made it a test, and a test is only worth running if you can read the result.

So we built it as its own website. If we had bolted it onto the existing donation pages, its visitors and its donations would have been mixed in with the giving that was already happening, and nobody could have said what the new idea did on its own.

We didn't build that website from scratch. We used one of Raisely's ready-made fundraising templates, so sign-ups, emails, activity tracking and fundraiser pages already worked. That let us spend the time on testing instead. We walked every path a visitor could take, start to finish, before it went live.

The campaign recruited 600 people to fundraise on Women's Refuge's behalf, and Winter Appeal donations roughly doubled year on year.

Splitting Facebook in 2 made its contribution readable. The June campaign recruited fundraisers at low cost. The July campaign, judged on its own goal, returned around 7.5 times what was spent on it.

The June spend kept paying out long after the ads stopped, because each fundraiser it brought in went on to ask their own network for sponsorship, using the social posts, posters and QR codes we supplied.

The emails ran at 35 to 45% open rates across all 6 sends, above typical nonprofit benchmarks. Email did the most work in this campaign, turning existing donors into fundraisers with networks of their own.

Safe Night had 1 way to give and no evidence that a second would work. We left the first one running and built the second as a contained test: its own website, on proven software, checked before launch, with email recruiting the first fundraisers and paid social split into 2 separate jobs.

Inside a single appeal window, 600 supporters signed up to fundraise, participants logged 29,147 km, and Winter Appeal donations roughly doubled year on year. Women's Refuge now has a second way for people to give, and knows what it costs to recruit the people who drive it.