Copy isn't subjective

Copywriting has done as much for our business this year as almost anything. The hard part is it feels subjective. Two people look at the same headline, disagree on which is better, and neither can prove it.

You can't settle it by asking customers either. People lie in surveys. Not on purpose, usually. They tell you what they think you want to hear, or what makes them sound like the person they'd like to be. Casual conversations with your target demo have the same flaw, even with your own customers. The reason someone gives you for buying rarely matches the reason they actually buy.

A headline test gets around all of it. It measures what people actually do when there's money on the line. When one line beats another, the preference is real, because someone backed it with a purchase.

So we stopped arguing and started testing. We've been working with Nate Lagos, and after a few rounds, one thing stood out: the lessons stack. Each test teaches you something real about the customer, and you carry it into the next one.

The homepage test

We started on the homepage. Not the big headline at the top, the smaller one that sits above the product block. It said "Best Sellers."

We changed it to "Go anywhere and do anything with Groove." Conversion went up 16%.

The lift was nice. What we learned mattered more. Our customers want to know their gear won't fail them. They go to work, they train, they get outside, and they need it to hold up. "Best Sellers" said nothing about that. The new line did.

Carrying it forward

Most brands treat every test like a fresh start. New page, new guesses, nothing carried over. That's the waste.

We took what we learned about the gear straight into the next test, on a new product.

New products are the hardest thing to write copy for. Something fresh, and no real idea yet how to sell it. The headline we'd been running was "Why it's the best in the world." We replaced it with a line built on what the homepage taught us: "Why it'll never quit on you."

It beat the old one by 91%. Almost double.

Same lesson, different page, and it won bigger the second time. We weren't guessing. We were using something we already knew worked.

It even works during a promotion

Last one, and this is the one that surprised me most. You'd think once a discount is on the table, the price does all the selling and the copy stops mattering. Not true.

We don't run many sitewide sales anymore, but Father's Day is one we keep. Most brands switch everything to promo messaging when a sale starts and abandon the angles that usually sell the product. We kept ours. We added a line for the person actually buying, who's usually shopping for someone else: "30 percent off for dads that deserve the best."

Profit on the sale went up 7%. That doesn't sound like much until you remember it's seven points of profit from one line of copy, on a sale we were already running.

The takeaway

Copy feels subjective until you put a number on it. Once you test, the argument is over. The data picks the winner.

What matters more is that the lessons stack. The homepage test gave us 16%, but the real prize was learning what our customers are afraid of. That lesson drove the 91% lift on a product page and the 7% on the sale. One lesson, three wins, because we treated each test as something to learn from, not a coin flip to win.

You don't guess your way to good copy. You test, you learn, and you take the lesson into the next one. Do that enough times and it stops feeling subjective at all.

P.S. If you want to try this, start with the smallest, highest-traffic piece of copy on your site you've never tested. A subhead, a button, a section header. Write one alternative based on what your customer actually cares about, not what you want to say about yourself. Run it. Whatever it teaches you, write it down and use it on the next test. The stacking is the whole point.