Hamilton Feltman wrote: > I just want to know a price For DEVOLOPMENT of a device > (Schematics, prototype, and board layout). There has got to be a > somewhat round figure for the engineering skills to build projects > for "Another Company." What do you charge them? Of course there > are many variables, just needed a starting point. and Ben L Wirz replied: > Depending on the size of the company, it's reputation, and the type > of project, a engineering man hour can go for $20 to $200+ US. > Many independent small time guys get around $30/Hour if they are > pretty experienced. > .... > It's also often true with independant guys that you will pay for 1 > hour and get 2 hours of work if it's not something they are an > expert on. I.E. they won't charge you for their learning curve. Hamilton: What Ben says is true; the range is HUGE. Most guys in this area (Southern California) charge around $60/hour; I charge significantly more. There are two things you need to determine: The minimum that you MUST charge in order to make it worthwhile ("worthwhile", in this case, means "as much as you could make doing anything else that you enjoy equally well"), and the maximum that you CAN charge while keeping a straight face. I'll pretty much ignore the first issue... Only YOU can determine the minimum that you need to make. Just keep in mind that, as an independent contractor, you're going to have a LOT of expenses that you don't have as an employee. The second issue is worth some discussion: 1. In general, if you can handle only one small part of the product-development process (writing the firmware, for instance, or designing and laying-out the circuit but NOT actuially building PC boards), you can't charge as much as if you can handle the ENTIRE process. 2. On the other hand, if your particular specialty is pretty rare (even if it's just a small part of the whole process), you can charge more than someone who has a more-common talent. Hotshot antenna-design guys, for example, make a TON of money. 3. If you can MANAGE a project, rather than just follow directions, you can charge A LOT. "Dilbert" cartoons to the contrary, a good manager is worth his weight in gold. 4. There's nothing wrong with adjusting your rate in proportion to the product's value to the client. I have three rates, for example, for software development: The highest is for software that the client will sell in source-code form, the next is for software (or firmware) that the client will sell as part of a mass-produced product, and the lowest rate is for software that the client will use only in-house. I also have a "charity" rate that just barely covers my expenses... If I write code for a friend, the fact that I'm charging him SOMETHING ensures that we STAY friends. 5. As Ben mentioned, many consultants (including me, by the way) don't charge -- or don't charge MUCH -- for time spent climbing the learning curve. If you expect to have a lot of these unbillable hours, your rate needs to compensate. 6. Most employees aren't aware of their ACTUAL cost to their employer. In many cases, your salary isn't even HALF of what your employer pays for your work. If a potential client is trying to decide between hiring a full-time guy and outsourcing his development to you, your rate can be pretty high before it becomes unattractive to him. 7. I know I said I wouldn't go into the minimum-rate calculation, but you do need to take into account the amount of liability that you'll be taking on. I'm not talking just about legal liability here -- although that IS an issue, and some clients may require you to carry "errors and omissions" insurance (very expensive, by the way) -- but about the smaller kinds of liability... The support issues, etc. My standard contract guarantees free software bug-fixes forever, for instance. Lots of people tell me this is stupid, but I don't mind... I think the guarantee's important, and besides, it doesn't get used very often. When it DOES get used, though, even a very minor fix can end up costing me a lot of money. Just this week, I spent as much as I charge for two hours of my time on short-term equipment rental and FedEx shipping in order to make a 10-minute fix to a program for which I'd billed only ten hours in the first place... And this doesn't take into account the hour that I spent on the phone with the client. Not only did I have to "give back" 20% of the money I made on the job, but the half day that I spent in order to make that ten-minute fix generated no revenue AND pushed all my other projects back by half a day. You need to anticipate this sort of thing (even if your contracts aren't as generous as mine) and set your rates accordingly. Finally, there's one thing that a lot of people don't seem to understand when they make the transition from employee to consultant: There's a big difference between what they've been doing in their garage and what a client who's paying good money expects from them. When you design and build something yourself -- a "MIDI Octopus" clone, say -- you can afford to take all sorts of shortcuts. Your code needn't be thoroughly documented or tested, your PC board can be drawn in Microsoft Paint, your BOM need only show Radio Shack part numbers, you can ignore UL and FCC regulations, etc. When you build a REAL product for a REAL company, however, things are different... You'll need to have Gerber files, drill tapes, silkscreens, assembly drawings, etc.; your code will probably have to be written to a higher standard; you'll need to find REAL sources (and maybe second sources) for all the parts; your PCB design must conform to the client's design rules; you may need to get all the relevant regulatory approvals; you must provide SOLID documentation; etc. As Fred Brooks said in "The Mythical Man-Month", the difference between a "Program" and a "Programming Systems Product" is enormous. He was talking about computer software specifically, but the lesson applies to engineered products in general. To paraphrase Brooks, a "Program", which is what is commonly produced by a guy in his garage, is complete and ready to be used by its author on his system. To transform that program into a "Programming Systems Product", which is the only thing that large software companies want to ship, the program must be generalized, documented, formalized, tested both alone and with the other system components with which it will interact, etc. Brooks estimates that a Programming System Product costs nine times as much as a Program. I think his estimate is low. There's a reason we consultants charge so much; what we do is HARD WORK. There's a LOT more to say on this subject, but I'll stop here... I'm sure that there are any number of "So you want to start your own service business" books available at your bookstore; read one. -Andy === Andrew Warren - fastfwd@ix.netcom.com === === Fast Forward Engineering - Vista, California === === === === Did the information in this post help you? Consider === === contributing to the PICLIST Fund. Details are at: === === http://www.geocities.com/SiliconValley/2499/fund.html ===