Supply chains have spent a decade building visibility. Raw materials arrive with certificates, components carry lot codes, pallets are scanned at every dock door, and a shipment’s position is available to the hour.

A modern manufacturer can tell you where a bearing was cast, when it cleared customs, and which trailer carried it. Then the product is physically assembled, the single step where all of those traced parts become the thing the customer actually buys, and in a remarkable number of operations the data trail simply stops.

The joint that holds the product together is very often the least documented event in its entire journey from mine to doorstep, and that gap is starting to matter commercially.

Traceability Ends Where Assembly Begins

The blind spot is structural rather than careless. ERP systems know which parts were issued to a work order. Warehouse systems know when they moved. Quality systems know the inspection results at goods-in and at final test. What none of them typically records is the act of assembly itself: whether a given fastener was tightened, to what value, in what sequence, by which tool. The information exists for a fraction of a second inside a wrench and is gone. When a field failure or a customer complaint arrives eighteen months later, the investigation has certificates for every component and nothing at all for the step that joined them.

The Technology That Closes the Gap

Closing that gap is not a research problem. Connected tightening systems control each fastening against a programmed target, capture the full torque and angle signature of the joint rather than a single pass or fail flag, and stamp the result with the tool, the operator, the time, and the product serial number. Suppliers of engineered fastening systems such as Atlas Copco ITBA developed this capability for automotive and aerospace lines, where every safety-critical joint on a vehicle has carried a data record for years. The significant change for supply chain leaders is economic: the same instrumentation has become accessible at the scale of a single assembly cell, which puts joint-level traceability within reach of mid-tier suppliers rather than only the OEMs at the top of the chain.

Why OEMs Are Starting to Ask For It

Evidence requirements flow downhill. An OEM that logs every joint in its own plant has little patience for a tier-one supplier whose assembly record is a signature on a route card, and supplier quality agreements are beginning to say so explicitly. The commercial logic is easiest to see in recall economics. A manufacturer who can prove which serial numbers passed through a suspect station in a defined window recalls thousands of units. A manufacturer who cannot recalls everything built that quarter, and absorbs the cost difference, which routinely runs into millions. Joint-level records turn that from an estimate into a database query, and procurement teams have noticed that a supplier able to answer such a query in minutes is a measurably smaller risk than one who needs a fortnight.

The Regulatory Tailwind

Regulation is pulling in the same direction. Under the EU’s Ecodesign for Sustainable Products Regulation, digital product passports will progressively require products placed on the European market to carry structured, machine-readable lifecycle data, with the first mandatory passports arriving for batteries in 2027 and further product groups following through the decade. A passport is only as good as the records behind it, and assembly is one of the lifecycle stages where most companies currently have nothing to put in it. Organisations building their data architecture for product passports now are treating joint-level assembly records as an upstream feed rather than a separate initiative, because retrofitting that evidence after the fact is somewhere between expensive and impossible.

Visibility Cuts Both Ways

There is a wider lesson here about how unevenly visibility investment gets distributed. Research covered recently on this site found that UK firms hit by supplier cyber attacks struggle to map their exposure fast enough, taking 1.9 days on average to understand which suppliers were affected by an incident. Product quality has the same shape of problem. The question is never whether something will go wrong somewhere in a chain of hundreds of parties; it is how quickly and precisely the affected scope can be bounded when it does. Cyber teams are learning that lesson about supplier networks. Quality and supply chain teams are learning it about the physical product, and joint-level data is what bounding the scope looks like at the level of a single unit.

A Modest Line on the Capex Plan

What makes this blind spot unusual is how cheaply it closes compared with almost anything else on a supply chain technology roadmap. A connected tightening cell is a tool, a controller, and software, deployed one critical joint at a time, starting with whichever joint features most often in warranty claims. It needs no change to the product and no integration programme to begin returning audit-ready evidence. For operations leaders drafting 2027 budgets, the practical move is to put joint-level data on the supplier scorecard and one instrumented station on the shop floor, and let the first targeted recall that never happens pay for the rest.