New Product Development (NPD) takes place within a web of connected actors who do not master fully the objects of the process, but rather they interpret the affordance - the en-abling and constraining framing by objects such as sketches, drawings, specifications, mock-ups and prototypes. The article draws on critical comments and extends Hutchby’s concept of affordance. Even if Hutchby does not study an artefact as a micro-level phe-nomenon from an engineering perspective and only considers the consumption of an arte-fact, we use him to study the development of artefacts in NPD. Hutchby’s approach con-sists of a functional and a relational dimension. We haul the functional dimension at both ends by employing an engineering and a sociological view. This is done by first viewing an artefact as consisting of different sub-systems and components and second by understand-ing design as a combination of reading and writing text in a situated context. The relational dimension is extended by employing American pragmatism. The empirical part of the arti-cle consists of four case studies of NPD in enterprises (under pseudonyms). Two cases fo-cus predominantly on technical issues of NPD, whereas two are more in the sociological direction. Alpha employs the artefact as a drawing and/or a mock-up in order to understand the customer’s situation, Golf applies a draft version of the customer’s CAD-drawing to translate the requirement to the toolmaker, Juliett utilises drawings and specifications to negotiate technical and non-technical aspects alike, while Sierra uses a CAD-system to support joint-development crossing organisational boundaries. The cases share features such as triadic relations; customer’s problem, product designers and artefact. For example, a wheelchair is presented to Sierra, the result of NPD-activities carried out by a professional golfer. Two product designers interpret the affordance of the wheelchair and compare it with the intended functionality explained by the golfer – we denote it intended affordance; the identified gap is used to rewrite the design. Next, the affordance of both proposals is discussed and the selected solution forms the basis of a new writing process. The cases also share the tension between and combination of hardness of some elements of the product and the malleability of others. Finally, the perception of an artefact is not an isolated and individualistic activity, but a process where human agency makes use of a network in which social and material factors constitute the interpretation. Our findings thus substanti-ate that artefacts frame the possibilities for action within NPD.