The goal of the game is to run a full Works racing team. This means everything to build a car is done in your own factory. (At the grassroots level, these principals will still need to apply but on a smaller scale)
Hiring Designers unlocks the ability to start designing your own parts. In the early game this might be a new design of Roll Cage, at the late game it might be a new Engine. Computer simulations and Wind Tunnel time can contribute to it's overall performance, reliability and lifespan at a cost (money and time) and are optional.
Newly designed parts don't have a specific performance values. For example, you'll know it's performance score is between 60 and 80. If you used a windtunnel and extra compute in the design process then the range might be more like 70 to 75. The more this range can be reduced during design the more confidence you'll have on the part you're creating, you finalise the design whenever you're ready to move on.
Designed parts have to be manufactured. Parts take time to make and you'd be competing for resources within your own factory. If you have a three car team and you want them all to run your latest design you'd have to batch build one for each car plus as many spares as you'll need. Machines and mechanics in your factory get to work, machines need raw materials, maintenance and a little cleaning but they'll get to work on your batch of parts.
To work out exactly how your part performs, or to know when it breaks. you have to use it!
Mechanical parts can be machine tested (push it until it breaks to find out it's performance / durability at the cost of the part itself), aerodynamic parts need testing on a track, that could be an expensive onsite test circuit... or just send it to the race and see how it does.
At best, you know exactly how fast your part is, when it's performance drops off, exactly when the risks of critical failure appear
At worst, you're racing blind with only a vague understanding of what your car is capable of Two choices:
Design > simulate > manufacture > test > learn > race with certainty
or
Rush a Design > don't simulate > manufacture fast > don't test > don't learn > race with uncertainty
Both could win the title but they are very different paths