Everyone agrees we need to do something about family violence. Politicians promise tougher sentences. Advocates call for better support. And quietly, in courtrooms across New Zealand, judges keep making the same order: complete a non-violence programme.

In 2024/2025, 2,933 people finished those programmes. That's eight people every single day being released back into their communities after court-mandated intervention designed to stop them from being violent again.

Think about that number for a moment. Nearly three thousand New Zealanders — mostly men, almost certainly — sat through group sessions, filled out worksheets about anger management, and learned alternative ways to handle conflict. Then they walked out the door. (Source: Ministry of Justice, family-violence — 1.Non-violence programmes)

What we don't know is what happened next. Did it work? Are their partners safer? Did they reoffend? The Ministry of Justice tracks who completes the programmes, but not whether completion actually changes behaviour. We count the people who show up. We don't count the violence that stops — or doesn't.

This matters because completion rates tell us nothing about effectiveness. A programme with a 100% completion rate could be utterly useless at preventing violence. Or it could be transformative. We're spending court time, taxpayer money, and — most importantly — asking victims to wait and hope, without knowing if any of it reduces harm.

And the scale is significant. Nearly 3,000 completions in a single year suggests either a lot of people are being ordered into these programmes, or the same people are cycling through repeatedly, or both. Without baseline data from previous years, we can't tell if this number is rising, falling, or stable. But we do know this: whatever we're doing right now, we're doing it to thousands of families every year.

The real question isn't how many people finish these programmes. It's whether the person who hurt someone in 2024 is less likely to hurt them again in 2025. And right now, we're flying blind. We're treating family violence like a checklist item — programme completed, box ticked, case closed — when it's actually a question that can only be answered months or years later, in the quiet moments when violence either happens or doesn't.

Three thousand people completed court-ordered programmes this year. We should probably start asking what that actually achieved.