Shopify’s best selling sort ranks products by past sales, so a new product with zero sales history starts at the bottom, gets little visibility, and therefore makes few sales. That feedback loop is the cold-start problem. The fix is a blended sort that weights product newness alongside sales, giving new items a temporary visibility window to prove themselves.
How the feedback loop works
Best selling is backward looking by definition. Position drives visibility, visibility drives sales, and sales drive position. Established products enjoy a self-reinforcing cycle at the top while new products face the same cycle in reverse: no history, so no position, so no impressions, so no history. The sort order is not broken, it is doing exactly what it says. The problem is that what it says ignores everything you know about a product except its past.
What a blended recipe changes
A merchandising recipe scores every product on multiple weighted signals instead of one. A typical blend for this problem combines recent sales velocity with a product age signal that favors newness. New products enter with a meaningful score from the age component, land in visible positions, and start generating real data. As they age, the newness contribution decays and their earned sales performance takes over. Strong new products keep their spots, weak ones slide down on merit.
Does this hurt your proven sellers?
Not if the weights are sane. Newness should be one ingredient, not the whole dish. Your established top performers still score high on the sales components, so they stay prominent. You are reserving a slice of premium visibility for products that have not had a chance to compete yet, which is exactly what a good merchandiser would do by hand, just without the hand.
This is one recipe in Dynasort. Install it from the Shopify App Store or see how it works.