One of the hardest things to learn when you work in data is this: Not everything needs fixing.
I know. Heresy.
I’m also a big believer that most things can be fixed. But over time I’ve learnt that the real skill is deciding what to tackle, what to leave alone, and where an 80% improvement is more valuable than chasing an impossible 100%.
If I’m cleaning data, I never guarantee perfection. With large datasets there will always be edge cases, oddities and things that don’t get picked up first time round. The important question isn’t whether an 80% fix is acceptable. It’s understanding what that 80% changes.
Take poor quality supporter data. Communications don’t reach people, returns are high, response rates are diluted and organisations keep carrying the same issues through every campaign.
So what happens when you improve 80% of it? fewer people receive communications they shouldn’t, fewer complaints arise, response rates become more meaningful, decisions are made on something closer to reality.
Is it perfect? No. But it’s materially better than where you started. And sometimes that’s the decision – not to fix everything, but to fix the things creating the biggest friction.
Here’s what I consciously wouldn’t fix. I wouldn’t chase issues I can’t consistently identify at scale. I wouldn’t automatically split joint records unless it was affecting fundraising activity. And if an organisation only communicates digitally, I might not prioritise perfect postal addresses beyond what’s needed operationally.
That doesn’t mean those things don’t matter. It means I’ve consciously decided not to tackle them yet. That’s the bit I’ve learnt over time.
Good data management isn’t about building a perfect database. It’s about understanding where imperfect data is causing genuine problems – and putting your energy there first.
Because sometimes the smartest thing you can do…
…is leave something alone.
If your team is trying to improve data quality, untangle processes or work out where to focus effort first, I’ve got some space over the coming months to help.
And if you know a team trying to fix everything all at once and needing help prioritising, I’d really appreciate an introduction.
#CharityData
#DataManagement
#CRM
#Fundraising
#DataQuality
