The store

Shopmicas is a fashion brand on Shopify, selling to shoppers in the US and Europe. Every time a new batch of products lands, the merchandising team has to decide how many impressions each new arrival gets in its first days on the site.

Before the test

That number had always come from experience. It wasn't wrong, exactly. The problem was that nobody knew whether it was right. Give a new arrival too few impressions, and a good style never gets the chance to take off. Give it too many, and a weak style holds a spot that a better product could have used. Since the number had never been tested, there was no way to tell.

What we changed

Our fashion e-commerce expert replaced the gut call with a calculated number, based on what we know about each new arrival and how similar styles have sold before.

How we tested

A 50/50 A/B test over 45 days. The primary metric was the share of new arrivals that made at least one sale within 30 days of launch.

What we found

That share went up by 15%. For the first time, the team could see what the number was worth, instead of arguing about it.

"I had never worked with a team this professional. They understood every metric and every situation I brought up. They built our whole new-arrival strategy with us: how many impressions a new arrival starts with, what the thresholds should be at each stage of its lifecycle, how each subcategory should be treated differently, and how to balance new arrivals against existing products. I had no idea it was this complicated. I used to think boosting would fix everything."

What they lacked wasn't features. It was features plus a strategy that could get them to their goal.