Cost of implementing late-stage changes to a product

I recently came across a case study of a portable solar lamp designed for developing markets. Apparently, when they ran field trials of the device, it was discovered that there was a big demand for the capability to charge cell phones off the lamp - even people in the poorest countries on the planet regularly own mobile phones.

I was wondering if anyone had an rough idea of how much a design change like this would have cost relative to the overall project cost, compared to if it was brought in as a feature well before field trials (say after the first prototype)?


A change in scope will always affect cost and timing. But your question question pertains to when that change of scope occurs.

Will I need to change production tooling? That can be a big chunk of change. But in many other cases, it is not.

How much redesign will be needed? Accommodating the feature in the first place would require time, so you are looking at the difference in time. It could be minimal, could be a lot.

Do I need to validate the change with end users and to what extent. Again, could be big dollars. Or not.

In this case, it seems you would only need to add a USB port to connect to a phone. Power will be DC so there is no need for an inverter. Doesn’t seem to me as a major change.

Yeah - this is an impossible question to answer generally since it all depends on the impact to the design.

To quantify something that specific you’d need whoever was managing the program to understand the manpower cost, tooling and design costs, and impact of delays.

If something like you mentioned can be made with a simple tweak to the circuit board and a small cutout added to the tooling, then it may be a very small issue. A few thousand dollars in tooling tweaks plus the costs associated with paying the engineers and vendors to respin the boards.

To add a USB port late on something like a cell phone? Could be hundreds of thousands of dollars of regulatory costs, antenna testing, and months of delays. So no great way to gauge without all the specifics.

Cheers, thanks for the help guys!

Generally, the later the implementation, the greater the cost/time.