The most common delay in CGI projects is not production, it is waiting for data. This is what is needed, what is optional, and what happens when none of it exists yet.
Three things are ideal: CAD data of the product, the artwork as print or vector data, and a handful of reference images showing how the result should feel. With those three, a project can start immediately.
If one is missing, that is not a blocker. It only shifts the effort: whatever is not supplied has to be reconstructed.
The usable formats are STEP, IGES, OBJ and FBX. STEP is the normal case out of product development and usually the best basis, because dimensions and edges are exact.
What is often overlooked: CAD from engineering is optimised for manufacturing, not for images. It frequently contains internal parts, screws and tolerances that are never visible, and it lacks the small edge fillets that make light catch in a render at all. That preparation is part of the work, and it is one reason the first image takes longer than the second.
Then the product is rebuilt, from dimensioned drawings, good photographs from several angles, and information on height, diameter and wall thickness. This works well and for simple shapes is barely slower. For complex geometry with mechanics it becomes the larger part of the job.
A product image lives on its label. The artwork is needed the way it goes to print: as PDF, AI or EPS with vectors, not as a JPG pulled from a slide deck. An upscaled screen JPG is immediately visible in close-up and ruins an otherwise good render.
Equally important is the information the artwork file does not carry: where the spot varnish sits, where the foil is, what is matte and what is gloss, where the embossing runs. Those differences are exactly what creates the impression of quality, and they cannot be read off a flat file.
Three to five images are enough, and they do not have to come from your category. More useful than "premium and modern" are concrete examples with one sentence each on what you like about them: the light, the background, the composition, the type of reflection.
Counter-examples help just as much. One image labelled "definitely not this" often saves an entire round of revisions.
That works too. From sketches, dimensions and a description, a product that does not physically exist can still be built. That is precisely why CGI is used in launch projects: the Vela case study shows a skincare product visualised before production began.
The difference is not feasibility, it is the amount of alignment needed. The less that is fixed, the more gets decided during the process.
STEP and IGES are the normal case and usually the best basis. OBJ and FBX work as well. What matters is less the format than whether the geometry is clean and complete.
Yes. From dimensioned drawings, photographs from several angles and material information the product can be rebuilt. For simple shapes this costs little time, for complex mechanics it is the larger part of the work.
As vector or print data, so PDF, AI or EPS. An upscaled JPG from a presentation shows in close-up and costs quality. Alongside it, the information on finishes such as spot varnish, foil or embossing.
Describe the product, the scope and the deadline. It usually takes a few sentences to say what is needed and which route makes most sense.
Start a project ↗