From Saw & Sew to Datafurn

When I designed the Saw & Sew chair, the goal was individualisation at scale. The response was strong, with early sales and press.

But producing and selling a highly individualised product at scale proved hard in practice.

The wall

People responded to the chairs I had designed. They couldn’t picture their own.

How do you catalogue and configure near-endless variations so that each one looks real and can actually be made?

The tools I found were built for more rigid product structures. Their boundaries were fixed around the offering, not the resources that decided what was possible.

So I decided to build that infrastructure.

What Saw & Sew did

At Saw & Sew, every configuration of the chair was bounded by design rules before the model was generated. A single parametric file could produce endless variations of a product. When a customer made a choice, an aesthetic style for example, the configurable inputs and the 3D metadata were filtered by the rules of that choice. I called this on-the-fly CAD.

The app could generate a photorealistic catalogue of products from its own structure, and customers could configure each one in near-endless variations. It ran in production.

The missing layer was manufacturing.

What it taught me

The principle was to map a parametric file’s inputs and outputs to a relational database. That connects parametric design to real-world data.

In principle, it makes an infrastructure for near-endless product variation possible.

Datafurn

I found I enjoyed the technical problems more than product design. So I moved from designing products to building the systems that make them efficiently.

That led me to the gap between design and production. The approach I’d built works upstream, at the design stage, which is where I think that gap is best addressed.

Datafurn adds the layer that was missing: it brings manufacturing data into parametric design. A CNC shop’s machine, material and tooling limits, held as data, with each design generated from the shop’s measured stock and checked against its machines and tools at the design stage.

This all started from a conviction: made-to-order is about efficiency, not luxury. I began with one product built from components. Datafurn’s aim is the whole chain, from design to production, and the limits at each step. Machine code is next.

Winston Ferguson, founder