We are often contacted by companies that are still early in their product design & development projects and need to estimate their cost, so they ask us for production RFQs.
Of course, that’s understandable. They need to plan for the amount of money to raise, as well as the timelines they need to work to.
They usually write something of the sort to us:
“Before moving forward with design work, I’d like to obtain quotes to estimate the cost to get a prototype, and also to know about the production unit cost.”
The problem is, when it comes to unique (new-to-the-world) products with mechanical and/or electronic parts, this is far from realistic.
Why don’t early Production RFQs make sense?
Let’s take 2 examples of why early production RFQs may not make a lot of sense, and why.
a) Simple product
Let’s say you want to design a flower pot made of plastic, and it only includes a few components (1 or 2 plastic parts to be injection-molded, and packaging). You have drawn an initial ‘concept design’ that shows the desired shape.
From there, you need to select a plastic supplier, explain what you want, and you could get an initial, rough estimate. There is probably still some work (industrial design, then structural design, then final DFM review & adjustments), but it is not likely to change the whole project dramatically.
The supplier’s quotation will only be final once the 3D drawings of the pot have been confirmed, along with the material & finish information, though.
b) Electronic product
Let’s say you want a flower pot that alerts you when the soil is getting dry. (Most products we work on have more features than this, but let’s keep things straightforward.) And let’s say you have already got a local graphic designer to prepare the aesthetic look of the product. There are still many things to confirm, such as:
- Can it actually be made with high-volume production processes? An industrial designer can usually help with this.
- What microcontroller to base the design work on?
- How will the information be gathered on the device? What type of sensor, and in what location?
- How will the data be sent (wifi, Bluetooth…)? How often? To what other device (e.g. to a mobile phone with Bluetooth, but that requires a mobile app)?
- How will the device be powered? Is it on a rechargeable battery? How long between charges (if there is a continuous wifi connection, that will take a while)?
- If a mobile application needs to be developed (in addition to the firmware embedded in the device), what will its functions be? Only sending an alert? Also, keeping a history of past data and displaying it in a nice manner? Have the UX & UI been developed?
We could keep going on and on, but this is probably sufficient to drive the message: there are lots of unknowns that need clarifying, and engineering work is needed for those clarifications.
Depending on the direction the design goes, the amount (and timelines) of the engineering design work are deeply impacted – probably by a multiple of 1 to 5 or more.
In addition, the cost of the components is also impacted –by a multiple of 1 to 2, at least.
That’s why working on “quotations” or production RFQs makes no sense until all the important questions have been answered.
All a design team can do early on is “rough estimates that are absolutely not quotations”, and that’s only after they have spent time answering some of the questions. Any estimate given ‘off the cuff’ is going to be wildly inaccurate and avoiding that is probably better.
As we always say, a new product introduction project is more about DELIVERY in the early phases, and prediction accuracy becomes relatively high only after the product design has been frozen. But you need to spend a lot of time and money to get to that point… That’s the unfortunate reality of product development.
Rough estimates have a place, though
Now, a rough estimate is quite helpful. If you embark on a project that is estimated to necessitate somewhere between 200,000 and 500,000 USD, as well as 6-12 months of development, you need to be aware of it and plan accordingly.
Similarly, if the cost of the critical components already adds up to 34 USD (per unit), that’s also helpful – you can make a few assumptions, and bet that the full unit will cost 50 USD per piece (out of China, before shipment), work up to your total landed cost, add your sales multiple, and conduct some market research based on that.
The need for some engineering work to estimate feasibility and to get pointed in the right direction
Experienced companies will usually suggest an intermediary phase, before working on fully functional prototypes and all the design details, to get clarification on all the “big questions”. In our company, we call it the feasibility study.
We wrote an entire page about this, and recorded a video on that topic — click here to know more.
What if you insist on running an RFQ early in your project?
Despite our warnings, we see companies asking for production RFQs. And some of them come to us 6 to 18 months later, having made a limited amount of progress, and they tell us about their mistakes.
From our observations, one of these two outcomes will happen:
- The supplier will decide to be aggressive and quote under very optimistic assumptions, to ‘get a foot in the door’. In over 90% of cases, for a complex new product, that leads to a frustrating situation where the supplier spends their budget, then slows down, and then asks for more money. The founding team is in for a very nasty surprise.
- The supplier will be cautious and for good reasons* (the customer typically comes up with more requirements over time, and the nature of development work is a succession of problems that can’t be forecast ahead of time and that sometimes consume a lot of resources). The quotation is very high, scaring away the customer.
The unfortunate consequence is that, in most of these cases, aggressive suppliers get the business and disappoint the customer, while serious and cautious suppliers don’t get the business… until the customer comes back to them.
*Note: No company wants to lose money on the development of a startup product because production is not warranted (having a great idea is not enough for success) unless they are considering using the same technology for their own products as a sort of investment.
Main takeaways
If you are developing a new electro-mechanical product and haven’t made it to a fully functional prototype yet, here is our advice:
- For your version 1.0, try to simplify the product. A product with just the essential feature(s) will go to market faster, with a lower investment, and at a lower unit price, than a more sophisticated product.
- As Tony Fadell wrote in his excellent book Build, the objective of version 1.0 is to get feedback from the market, so that a more sophisticated version 2.0 can be a big success. (It is explained here.) That’s very true for new-to-the-world products in general.
- As a founder, you need to start working on the technical side of the product. You need to learn about the key technologies. You need to know where the “big assumptions to test” are hiding.
- It is better to run the feasibility phase first to clarify the “big questions”first, as I explained above.
- If you can’t do it yourself, you need to find people who will work on an initial, rough, semi-functional ‘proof of concept’ prototype. Without that, no manufacturer will take you seriously.
- If you need to raise money from investors and they keep telling you “Show me first a fully developed prototype that works & looks like the production units”, ask them questions to understand what risks they are trying to address. Ask if some earlier milestones, such as a prototype with the 1 or 2 key features, would be acceptable. Serious hardware-focused investors can often be flexible.
