How New Forest Clothing Stopped Hand-Picking Recommendations For 6,000 Products

About New Forest Clothing

New Forest Clothing is a family-run business that started as market stalls in southern England in 1978 and grew into shops in Ringwood and Salisbury. In 2016 they launched their own label with a single wax jacket, and it now sits alongside brands like Le Chameau, Musto, Deerhunter and Härkila.

That mix is what makes merchandising hard: around 6,000 products across jackets, layers, trousers, footwear and accessories, changing every season.

The Challenge

The store had been using Rebuy. It started well, but over time, as co-founder Paul puts it, the support and the relationship dwindled - and the team stopped feeling the app was proving its worth.

Meanwhile, the catalogue had outgrown the way recommendations were being managed. They had to be added manually, product by product, and every new product meant going back into the app to add them again.

"Going in and putting in those manual recommendations at a product level, that was a huge task for us. It's something we couldn't really maintain every time we uploaded a new product."

Benji

Head of Digital Marketing

The Solution

New Forest Clothing wasn't looking for another widget. They wanted AI choosing the recommendations instead of a person. Benji says that was what set Selleasy apart from the other options they considered.

They defined what the recommendations should do. The Logbase team built it.

They define, we build‍

The team shared their requirements for each widget, and our team prompt-engineered Selleasy AI around them.

Designed to fit the store‍

The widgets were built to match the store's existing theme rather than dropped in as generic components.

AI they can control‍

The team can adjust the recommendations through the AI prompt in the settings, without raising a support ticket.

Benji's summary of the split was simple: they hardly had to do anything. The widgets worked the way the team wanted and fit right into the site. The styling, he says, needed no input from them at all. 

Results

New Forest have seen a definite increase in average order value, with more customers checking out with several items rather than one.

Every recommendation on the store is now generated by AI. In the first three weeks, Selleasy recommendations influenced 5.4% of the store's orders. 

The manual recommendation work is gone. Recommendations are generated from the AI's instructions, so adding a new product no longer creates another task.

Asked whether they would recommend Selleasy to another outdoor retailer looking at upsell apps, the answer was that they wouldn't hesitate.

"You've delivered on what we've asked for perfectly well, and it seems so far to be having the desired result for what we're looking for. Hence the reason we're coming back to you, looking at other apps you do."

Paul

Co-founder

Thinking About Changing Your Upsell App? 

Three things pushed New Forest Clothing to look. All three are worth checking against your own setup.

1. Who maintains your product recommendations?‍

If someone is entering them product by product, that job grows every time your catalogue or brand list does. For a store with a large catalogue, AI recommendations is more efficient.  

2. What happens when you upload a new product?‍

If recommendations don't populate on their own, you may be building a backlog without realising it.

3. Has your support relationship changed since onboarding?‍

Good service at the start is one thing. What matters is whether you can still get help when you need it.

If any of these sound familiar, it's worth a look at your current setup.