Naia Spaces
Upload a parts list. Naia works out what every part is made of and matches it against your own data: the supplier quotes, material specs and test results you already have. Then see in minutes what an alternative does to cost, environmental footprint, sourcing and risk.
Book a conversationFive screens from Spaces, running on an example product: a 25-litre day pack.
Not just the sustainability lead.
Head of product, VP engineering, chief design officer, general manager, anyone accountable for a product decision that would be expensive to get wrong.
Upload the parts list, get a figure for every line, try the alternatives. It takes a meeting, not a quarter. The figure is the hard part: a line has to become a real material and process before any number means anything.
One line of many, exactly as the engineer wrote it.
The line does not say. Naia infers both from the weight and the product around it.
From whatever catalogue you have. This one matched at low confidence.
An uncertain match comes to you, not past you.
Every line gets a figure like this one, and together they make the product’s total. From there you try the alternatives: change a material or a supplier and see what it does before anyone commits. Confirm a match once and the next parts list carrying that part starts from it, inside your own installation.
An AI estimate never stands alone. The system checks the material against the real catalog first and shows what it finds, so a sourced, catalog-matched option sits right next to the model’s guess, not in place of it.
When the system is not certain, it does not decide quietly. It surfaces every candidate side by side, confidence and source attached, and a person confirms or switches before it locks in. Garbage in, garbage out gets caught here, because nothing is trusted on the model’s word alone.
No AI sits in the calculation itself. The arithmetic, the source lookup and every aggregation are ordinary deterministic code, and the model is never asked to do the sum. It proposes and it retrieves; a person accepts.
No autonomous agents run loose in your data either. We build the system around the AI so it stays held in place, and that is a deliberate choice. The data this work touches, cost structures and product development that has not launched, is among the most protected material a company holds, including internally.
Environmental impact is the dimension with a real method behind it. A reference catalogue, PEF method and life-cycle stages, so a CO₂e figure per unit resolves to a source you can open and check. It is the headline number and the one we stand behind.
The other six are estimates, and the product says so on screen rather than in a footnote. Cost, capacity, sourcing, implementation and risk all depend on data that lives inside your company: your quotes, your supplier confirmations, your bench tests. So rather than print a confident figure, Naia shows what each number is worth right now and exactly what would tighten it.
In construction the equivalent already exists. Molio’s price data covers cost there the way a materials catalogue covers CO₂ here, which is why that product can put a price on a model on day one. In product development no such catalogue exists, and pretending otherwise is the fastest way to lose a room.
Two delivery formats. The choice depends on the size of the decision.
One day at your offices, on a single product or product line. We pin down the decision in the morning, then spend the rest of the day setting Naia up on your own data. By the time we leave, the key choices are worked through and you have a directional recommendation you can defend.
A longer piece of work with a fixed scope, for a portfolio or a group of products. Before work starts we agree what it covers and go through your data. Then we model several scenarios, with the method tuned to your priorities. You end up with a documented result and people in-house who know how to use it.
Either way, we deliver it alongside your own team. It is not software handed over for you to figure out alone.
The same method runs in construction together with Molio, at Danish architecture and engineering practices, on real projects. That is where it has been sharpened, and where it has had the most numbers through it.
The team behind it has backgrounds spanning LEGO, Vestas, Grundfos and Arla. Product decisions that would be expensive to get wrong are not new ground for us.
See what it looks like in constructionBring a decision that would be expensive to get wrong, and we will show you what the frame looks like laid over it.