Early in my career, when I moved from SEO into a Growth PM role, another growth PM gave me advice that’s stuck with me since: always ask what the smallest possible version of a change is. The smallest increment toward the thing you’re trying to achieve. The smallest thing you can ship so you can learn if it works. The idea is to get a cheap, clean signal before you commit real dev time or budget to something.
I still think that’s good advice in most cases, but there are exceptions, and this summer was one.
Before summer, product marketing launched a new product positioning. We’re a mostly inbound-driven company, so the website is the most important surface where potential customers form an opinion about who we are. My job for most of the summer was making sure the site actually reflected that.
Most of that was straightforward: page by page, ship the smallest change that gets the language right, check the bigger calls with leadership, move on. I worked with a Claude Project to flag pages that still used language from the old positioning, so I didn’t have to find them all myself.
Then I got to the homepage.
The homepage isn’t a page I get to edit freely. Our CEO is personally invested in it and we work closely with him when we update the page. So this wasn’t a “ship it and see” situation.
One thing we learned over the past year or so, is that if you update a homepage section by section, you lose coherence. Instead of one page telling one story, you end up with a patchwork of sections written at different times, each anchored to what the positioning happened to be at that point.
A homepage built out of incremental edits gets messy, because a homepage isn’t actually a stack of independent sections. It’s one narrative wearing the shape of several. And a narrative doesn’t survive being edited one paragraph at a time.
So instead of improving and updating the “old positioning” sections one by one, my designer and I stepped back and looked at the homepage as a whole. We rebuilt it as one coherent piece, waited until our product marketer was back from vacation and could weigh in properly, then presented it to our CEO together as one story, not three separate edits.
It went well. The new homepage is going live soon, along with a new main nav, something I’d wanted to do for a long time.
The smallest-possible-change advice still holds for most growth work: test a new channel, eperiment with a new way to explain your product, or test out if a new feature gets traction. It breaks down for anything that’s really one narrative dressed up as multiple sections. A homepage. An onboarding flow. A pitch deck. You can’t patch a story incrementally. At some point you have to step back and rewrite the whole thing at once.