When Small Changes Aren’t Enough

We just restructured AQUAVIEW’s information architecture. The whole thing. Map-first to dataset-first. It wasn’t a sprint’s worth of work—it was a decision that touched almost everything we’d built over the past year.

Which raises the question: when do you make a big change versus iterating with small ones?

Scrum wants you to work incrementally. Lean wants you to minimize waste. Both of those push you toward smaller, safer adjustments. And most of the time, that’s right. You tweak, you test, you learn, you adjust.

But sometimes incremental changes just paper over a structural problem.

We spent months making our map interface better—clearer filters, better interactions, faster load times. All good changes. But they didn’t fix the underlying issue: our users wanted to browse datasets, and our information architecture forced them to think spatially first.

You can optimize the wrong structure all you want. It’s still the wrong structure.

Here’s what made us pull the trigger: our sprint reviews kept surfacing the same friction points. Not “this feature needs polish” friction. More like “I don’t think this is how people actually work” friction. That’s a signal that incremental fixes won’t solve it.

Lean talks about sunk cost. We’d built a lot around the map-first approach. But sunk cost isn’t a reason to keep building on a foundation that’s not working. The waste isn’t what you already spent—it’s what you’ll keep spending if you don’t change direction.

So we changed it. The map is still there (it’s oceanographic data—geography matters) but it’s not the entry point anymore. Datasets first, spatial context second.

It’s messy. We’re rebuilding navigation, rethinking workflows, adjusting how everything connects. But the structure finally matches how people actually want to use the thing.

I think the lesson—if there is one—is that agile doesn’t mean you only make small changes. It means you’re willing to make the right change, even if it’s big, when the evidence points that way. The sprint reviews gave us the signal. Lean principles gave us permission to act on it.

Sometimes you need to rework the foundation, not just repaint the walls.