No Script, No Fear, All Opinion
RSS icon Home icon
  • The Economics Of Selling Your Hardware Project

    Posted on May 28th, 2014 EEVblog 51 comments


    I’ve had quite a few people ask why I am not selling my µCurrent through international distributors like I did last time. Well, that’s a good question, and I’ll answer that as well as breaking down the economics of selling your own hardware product, what price you should sell your product at, and the pros & cons of using distributors.


    The reasons I am not selling the µCurrent through distributors (resellers) this time around are as follows:
    1) I make a lot more profit selling it myself. And, if I’m honest, the idea of a distributor making more than I do per unit kind of bugs me. Even if it means more work for me.
    2) I have spent many months and a lot of effort setting up an Australia Post commercial account, and a system that allows me to sell and distribute my product easily and cheaply. Primarily for the Kickstarter campaign, but I was also thinking for the future. To now not take full advantage of that would be silly.
    3) The vast majority of sales for my µCurrent come from my large audience base and my own marketing, SEO, and word of mouth etc. Very few people who know nothing about it will stumble across it on a distributors site and decide to throw one in their shopping cart. It’s a fairly niche product.
    4) Customers will ultimately get a cheaper unit.
    5) I like selling my own stuff, I still get a warm fuzzy out of it.

    #1 Speaks for itself. Frankly, whilst I do enjoy doing this kind of thing and it’s part of my hobby (I’ve been selling hobby electronics kits/products for more than 20 years now), I am ultimately doing this to make money. I have to eat and support a family, and this is my day job. So the more profit in my pocket the better. I have no vested interest in helping make the distributors more money than I do.

    #2 This is a big deal. I’ve done all the hard work and now have a very optimised process for shipping. My µCurrents come to me tested and individually packed, all I have to do is operate the shopping cart and eParcel system, print the labels and customs form, slap the stickers on and drive to the shipping depot nearby. It’s easier than it sounds. It takes me less than 2 hours tops to process 100 units and drive to and from the depot. Heck, the post office will even come and collect them if I bother to arrange it.
    Incidentally, it would take about the same time to pack and ship bulk units to the distributors. So the time saving question has been rendered fairly moot.
    This is a lot different to when I was doing the testing/packing, and manual labeling and customs forms like I was before. Back then it made sense to use a distributor to save a lot of time.
    Although I will admit that there is the hassle of returns and inquiries, but this isn’t large. I get many inquiries anyway even when people buy from distributors. It’s not 100% hassle free either way.

    Is it worth is economically for me to spend the time on this? Well, do the math like I will later and see. Yes, on a per hourly basis it is well worth it. And remember, this is part of my day job.

    #3 This is more particular to my case than most, but it’s a simple fact that a niche product like this will not do much “passing trade sales” on the distributors websites. Because a) it’s not particularly cheap, so not at that “impulse” level, and b) it’s a niche use product, it’s not for general everyday use. This equation may certainly change if I have a product that appeals to a general audience.

    #4 To understand why customers may ultimately get a cheaper unit, you have to look at the math I’ll do shortly.

    #5 is a freebie for me. Forget I said it.

    In my particular case, the only major advantage that a distributor can offer, and that’s offer to customers, not me, is faster local shipping within their own country. And potentially less import tax issues etc, but that gets complicated.

    And BTW, using distributors is not a zero hassle solution. Unless you chose the completely hands off method like say Club Jameco. They do absolutely everything and you take home say 5-10% of the net sales.

    The Economics Of Pricing Hardware:

    There is an often thrown around figure of 2.5 times for hardware products. This is the Cost Multiplier. And 2.5 is not bad number to work from as a baseline as you’ll see shortly. Generally you’d want a good reason to go below this number.
    If something costs you $50 in true cost to manufacture, you’ll likely want to sell it for 2.5 times that, or $125. Why? Well, read on…

    First of all, pricing your product. There are two ways to price your product.
    The first is the Bottom Up method.
    This is where you take your base manufacturing cost, and then build a multiplier on top of that to get the retail price. That is the Cost Multiplier 2.5x number we are talking about above, and it is a number you need to pick based on what gross margin (profit) you need to make per unit in order for you to live and make this a viable business worthy of your time and effort.
    This method works well for niche products that have little or no competition, or something new that is not available anywhere else, people will pay what they have to pay to get it, up to a reasonable limit.
    Also, technical people understand this, and accept that you need to make a certain fixed margin. Typically you’ll try and make sure this figure is just above the one-off parts cost of someone making it themselves (customers will do the math!). So most would rather buy it from you assembled and tested instead of dicking around making it themselves.

    The second method is called Top Down pricing.
    This is used when you want to introduce a competing product into an existing market, or you have done your research and you know people will only pay a certain amount.
    Essentially, you price the product “at what people are willing to pay”. If your widget only costs $10 to manufacture, and you know they will pay $100 for it, then great, your gross margin (essentially your profit) is huge. And you’ll be able to sell it through even the most markupidy (that’s a word, really) of distributors. But if that same product costs you $80 to manufacture, then well, congratulations, you are likely going to be working for minimum wage, and oh, by the way, no distributor will touch you with a 10 foot pole. In the OSHW industry, where your customers are cluey, you can’t get away with huge markups.

    But in either case, remember, your gross margin will change drastically depending upon whether you sell the unit yourself, or you use a distributor.

    Distributors:

    So what about the economics of distributors? Time for some simple math.
    Assume your widget costs you $50 to manufacture, and this has to include all your costs including parts, assembly, testing, shipping, packaging, consumables, taxes, defective units (allow say 5%), missing shipments (allow a few %) etc etc.
    Using either the Bottom Up or Top down pricing system, you have determined a retail price of $125 for the widget. That x2.5 rule of thumb in this case, you’ll see why this number is important in a minute.
    If you sell it yourself, (assuming the customer pays exact shipping cost) then you make a gross margin profit of $75 per unit.
    This is a gross margin percentage of 60% ((($125-$50)/$125)*100%)
    Assuming you sell 1000 widgets a year, that’s $75K per year profit, not including other business overheads of course.
    That’s pretty darn good! That’s a decent full time living wage, you could conceivably quit your job to sell your one widget.

    Now, what if you use a distributor? Ok, well:

    Any distributor will want at least a 60% markup.

    It’s important to understand what markup actually means.
    Markup is the retail price divided by what it costs the distributor to buy it, minus one. So:

    Markup = (Retail/Cost)-1

    Sounds simple enough, and it is, but here is the kicker – If you want them to retain the same retail price you had in mind (or you are already selling at, because you know, it looks bad if you sell it cheaper than the distributor), i.e. $125 in this example, then they will want to buy it from you for about $78 at best to make their 60% markup

    (formula rearranged) Distributor Sell Price = Retail / (Markup + 1)

    Or $125/(0.6+1)) = $78 in this case.
    Ok, so if you sell it to the distributor for $78/unit, your gross margin has dropped from 60% to 36%!

    Profit = Distributor Sell Price – Cost

    That’s a profit drop from $75/unit to $28/unit! (a 63% drop)

    You now only make $28K/year for that same 1000 units. Still ok as a side business, but not one you’d quit your job for.
    Yet you still have to do all the management work getting the parts purchased, manufactured, tested, packed, shipped, ad chasing your distributors to pay your invoice.
    And worst of all, the distributor earns more than you do for your product! ($47 for them, $28 for you).

    And let’s not even talk about who pays for shipping to the distributor, and any import taxes etc.

    So that’s the hard reality using that Cost Multipler 2.5X rule of thumb. Drop that to 2.0X and the numbers start to look dire:
    $50 profit for you without a distributor (still good), or $12.50 profit through a distributor.

    That’s now 12.5% of the retail price in your pocket, and you are still doing most of the work. At this point you might as well take one of the royalty agreements were the distributor produces the product and just gives you a cut. Otherwise you are likely working for minimum wage or less. That’s no way to run a business.

    Do the math. If you distribute yourself, you get an extra $47 per unit. If you charged your time at say $100/hr, that gives you a whopping 30 minutes to pack and ship each unit. Worth it? – you decide – working for the man, or you having your own business earning $100/hour for shipping alone. All for one single $125 product selling in not very high volume per year.

    Selling your own product also gives you the greatest ability to discount the unit, absorb and offer free shipping etc. Something those down the food chain can’t often do.

    This graph of Gross Margin % vs Cost Multiplier for Direct Selling vs Distributor @ 60% Markup shows that you can’t even break even by using a typical distributor unless you have a 1.6X Price Multiplier.

    Gross Margin % vs Cost Multiplier for Direct Selling vs Distributor @ 60% Markup

    Gross Margin % vs Cost Multiplier for Direct Selling vs Distributor @ 60% Markup

     

    This graph of Dollar Profit vs Cost Multiplier for Direct Selling vs Distributor @ 60% Markup shows (on the Difference line) that for you to earn the same as the distributor per unit, you need to set a Price Multiplier of around that 2.5X – 2.7X mark. That is where that magic number comes from. Go below that multiplier and the distributor earns more than you do! See how the Difference line crosses 1 at around 2.5-2.7. You want to be on the high side of this number if using a distributor. Just for the warm fuzzies of course, because, well, you are doing all the work getting this thing made. It makes sense for you to earn more than they do.

    Dollar Profit vs Cost Multiplier for Direct Selling vs Distributor @ 60% Markup

    Dollar Profit vs Cost Multiplier for Direct Selling vs Distributor @ 60% Markup

     

     In Practice, The Value of Distributors

    At this point there are many people who would be screaming at gaping holes in the above scenario, and well, you’d be right. But make no mistake, the above numbers are real if you fit the scenario.

    The above is assuming that the distributor adds no value apart from taking shipping off your hands. And in reality that’s not the case, because they are a reseller who does much more.

    A huge benefit is that they market your product to their established audience, and also in many cases (e.g. Sparkfun or Adafruit), their distributors as well. And that is why they need this big 60% markup, because they have distributors too, and so the math continues…

    Distributors can have other advantages too, like handling returns and customer “where is my product?” complaints, and they take the cost risk on shipping if it goes missing. But ultimately, all failed goods will come out of your pocket, they usually won’t eat that. But thee are good benefits. And you only have to ship one bulk lot to them (they might buy say 50 as a trial then, then more depending upon sales), that’s a lot less hassle than shipping each unit yourself.

    If you just want someone to take the burden of packing, mailing and distribution off your hands though, then what you want is a “fulfillment service”, “mailing house”, or the many names they go under. They will charge a much much smaller percentage, putting you back into the financial drivers seat. What you are buying with the big name distributors is their brand, their community, and their existing customer base.

    You’ll note that in my situation above, I spent months organising and effectively setting up my own mailing house. So I get the benefit of discounted shipping rates like the big mailing houses, and a minimised workload that allows me to maximise my profit per unit. Not everyone will want to this, so there is great value in using fulfillment services for this.

    So the real value with distributors/resellers is in the extra sales they bring over just having your own shop. Because they have the big ready audience you are looking for! What figure do they need to bring in to make it worthwhile going with them?, well, we can do the math on that too:

    Assuming the original 2.5X scenario ($75-$28=$47/unit difference) and that you think you could sell 1000 units yourself. How many more sales would the distributor have to add on top of that for it to be worthwhile?
    You are forgoing $75K profit, as you’d now make $28 per unit instead of $75, so that comes out to 2678 sales. So the distributor has to sell almost 2.7 times the volume you could on your own for you to earn the same $75K profit.

    That’s not a bad figure, and any of the major distributor should be able to add that value fairly easily for a general purpose product. But the equation changes the more niche and the more expensive your product gets (#3 in my original list of reasons above).

    And likewise, the equation could be drastically in the favor of of using reseller/distributor by even a 1000-1. If you don’t have a name for yourself, don’t think you can get your product mentioned on high profile blogs or whatever for exposure, then the choice is obvious, you go with the distributor with a huge ready sales and distribution channel.

    If you want a bit of exposure that your own website won’t get, but don’t want (or can’t afford) a distributor or reseller to take a big margin, then you might want to use something like Tindie.

    Conclusion

    Don’t get me wrong, I’m not saying what to use either way for your own product, or that selling direct yourself is the best method. Even I may go back to using a big name distributor for more general appeal products in the future, as their customer base can be compelling, indeed, it may be your only or best option. Heck, the likes of Sparkfun have hundreds of distributors around the world, so if each one of those stocks 10 units, there is a thousand or two sales right there.

    There are also countless other pros and cons and nuances and scenarios I haven’t touched on, and every situation is going to be unique. But at least I’ve hopefully presented some food for thought here and some basic numbers on how it all works. I hope you found it useful.

    And ultimately, what I’d love to see most of all, is more people making full time livings producing useful hardware. Whether you anonymously sell countless different widgets you’ve designed though distributors, or blaze your own trail and make a name for yourself and distribute yourself, more power to you.

     

    Be Sociable, Share!
    • JFA in Montreal

      That was a fantastic post Dave. Thanks !

      I too have a hard time seeing distributors taking the apparent lion’s share of the product. However, for having tried to market my own products, I understand that to build and maintain a large purchaser’s base is tremendous work.

      We live in the era of the salesman. However, with the web, things might be easier for the little guy…

      That is, only if you consider your five+ years of dedicated high quality goodwill-producing activities to be “something easier” :-)

      I think you will start to reap the profits of your relentless production of added value.

      Cheers, and may you make money !

      JFA in Montreal

      P.S. Despite you not wanting to distribute, I plug your name all over the place to retailers. Maybe they’ll accept to buy your µCurrent Gold and retail it for, say, 125$…
      Considering the prices of ancilliary accessories in the Rigol catalog (an example of decently priced hardware), 125$ for something used professionally (like that NASA girl using the µCurrent at work) is a dirt dirt dirt cheap price.

    • http://Www.elproducts.com Chuck Hellebuyck

      The next step you need to cover is drop shipping or affiliate program. This way you form your own distribution network of sellers and give them a royalty for sales. It will be lower cost for you to sell through these distributors as royalties will be far less than your 60% markdown and you still control pricing and shipping while increasing sales beyond your own direct sales.
      And with your audience, I’m sure you could find many quality resellers.

    • http://blog.strobotics.com.au Stephen Eaton

      Fantastic post Dave!!!

    • http://AnomalousMaker.com Mr. Cruz

      Thoughts on Tindie?

      • http://www.eevblog.com EEVblog

        You are basically selling it yourself, but using Tindie as a bit of a promotional tool (people just browse on there and buy random stuff) and shopping cart system. Bit of the best of both worlds I guess.
        But you are stll stuck with all the shipping each unit, returns, lost items etc

    • Robert

      Dave
      well thought out argument.
      One thing that you did not mention explicitly was the Internet.
      The Internet has singly enabled all the things you have mentioned. It’s why bricks and mortar retailers are suffering.
      The old school distribution channels are being knocked over by the internet and cheap shipping.
      Maybe there is still hope for small business.

      • http://www.eevblog.com EEVblog

        Buy “distributors” I am talking about the online retailers. Not traditional shops like Jaycar.

    • DB

      You forgot another reason. Most distributors will only sell products that have regulatory approvals. In general you cannot sell hardware, for example, that is without CE approval in the EU, or without Ctick or RCM in Australia. Selling via eBay or other maens has created an uncontrolled mess for the governments dictating standards. These governments are out of control trying to enforce standards. As a result they won’t do a damn thing unless of course someone gets injured or your product interferes with a commercial or government service. Ignoring the ripoff compliance costs and time does place you in a cost advantage over any competition. But you do so at your own risk. Use lead based solder on anything sent to the EU (except medical for now, and aviation) you risk being arrested and jailed upon entering the EU.

      • http://www.eevblog.com EEVblog

        Tell that to Adafruit, Sparkfun, and a host of other OSWH and other online retailers.

        • pp

          That regulatory stuff is particurarly nasty in the EU. Virtually anything electronic sold here requires issuing a so called “declaration of conformity”. Recently there also is RoHS/WEEE to deal with.

          Small company can possibly sucesfully self-certify in house but for the sole maker trying his luck with his first widget as a side job that kills the deal.

          Sure, if you stick to your niche online shop, you can go under the radar possibly for years although you still risk being heavily fined if caught. Every week products are banned from being sold only for lacking the CE marking. At the same time retail chains are full of chinese made crap, because the whole CE thing is almost entirely about having proper paperwork, at least when we consider what is checked by authorities.

          All of those OSHW resellers are US based. Is there any operating from within the EU? Receiving single units for own personal use is a whole different story and doesn’t fall under regulations, not to mention the ability to inspect every incoming product.

      • Jope

        >you risk being arrested and jailed upon entering the EU

        No you don’t. You go way overboard with your hysteria.

        • DB

          Actually, I am not going overboard. I did a course on EMC compliance a few years ago and the lecturer stated clearly that the EU countries have legislation to make exporting non ROHS stuff to the EU a criminal offence and the penalties are severe including imprisonment. How many owners of shonky Chinese companies have been sent to jail? My bet Is few if any.

          The easy solution is don’t use lead, cadmium, mercury etch in your products and print on the PCBs the word ROHS.

          The cost of EMC certification for the likes of Sparkfun, EEVBLOG, DIY and kits etc, is prohibitive. To not have a provision for groups such as these is a violation of human rights. The are too many do gooders in corrupt governments telling decent people how to live their lives. They are the enemies of freedom and are vexations to the spirit of creativity and innovation.

          • SydB

            There has been recent provision for R&D within the RoHS legislation, at least in the UK, that can be applied to e.g. development boards (not sure about kits). In practise, it is mainly a leaded vs non-leaded solder issue. It is a simple fact that leaded is easier/more reliable/neater on a small/hand production scale compared to non-leaded. Non-leaded PCB supply is now easy so it is just an issue with the soldering of components and wires.

            I agree with the need for controls but it should be done sensibly and proportionately based on risk assessment. Clearly a business or person should be penalised for selling unsafe mains powered devices that kill (e.g. many Chinese PSUs) but not for putting 100 grammes per year of lead into the global electronics market.

            Other than for big producers, RoHS anti-lead is an absolute joke when considering, in relation, the amount of e.g. lead flashing used in roofing both on old houses and new builds, where the metal is in direct contact with rain running off into water courses. This is still standard in the building industry in the UK (I cannot speak for the rest of Europe).

            As mentioned above, the killing of innovation is being achieved by legislators allowing no or inadequate de minimis levels for such things as RoHS or WEEE.

            I guess that a) the legislators are lobbied by big business who, of course, do not want new small efficient companies to challenge them, and b) the legislators have not much clue about innovation, small scale business or electronics manufacture, largely because the small businesses cannot afford the resources to make their voices heard.

            As an example, until this year, all producers of electronics in UK had to pay a WEEE ‘compliance’ fee, regardless of quantity. Then a re-write came in this year with a de-minimis threshold of 5tonnes. However, a similar level of administration may still be required and so it is still an administrative burden. 5tonnes is a massive amount for little niche widget producers which may be producing less than 100kg per year. Then there is the question of who should actually be paying for the waste provision e.g. component level, PCB assembly level, encased product level, retailer level?

            The writing/re-writing of legislation also burdens small businesses because of the time required to reconsider the new implications and to decipher the documentation. There are government guidance documents provided but, still, any change requires a business to divert resources in order to assess potential impact.

            Of course, the fact is that enforcers are unlikely to be interested in tiny producers producing quality products that do not directly harm people or are not causing problems. But that is no substitute for keeping to the rules. Also, poorly considered legislation allows big companies with their huge legal teams to damage smaller competing companies.

            How many businesses have actually been fined or prosecuted solely for RoHS or WEEE breaches? Any?

    • Javier Loureiro

      Good question! I am starting my own hardware business, and this is something really important for me. Margins are dropping hard and people wants to pay less and less.

      Adafruit, Sparkfun, etc, they all have lots of distributors. What do you think about their model?

      • http://www.eevblog.com EEVblog

        Their model works exceptionally well for them, and anyone who needs their huge distrutor and community base.

    • Jason

      I could talk hours and hours about my experience starting up a hardware company, and also about my nightmare experience by outsourcing hardware design to big companies.

      Excellent article, I’m following EEVBLOG for years already even before I had any Hardware knowledge but by time I jumped into cold water and had a hard time to get started but now I’m able to swim and I’m getting better and better.

      I exactly end up with the rates mentioned in this article and distributing the items by myself.

      • http://www.eevblog.com EEVblog

        IMO if you have to outsource hardware design, then you shouldn’t be in the harwdare design business!

        • http://www.geppettoelectronics.com/ Nick

          I’m struggling to understand the need to outsource hardware design, unless you’re simply not skilled at all in electronics.

          Parts manufacturers fall all over themselves trying to make it easier for designers to choose their products by putting helpful example application schematics – and even entire BOMs in some cases – in their data sheets. Heck, TI even has an entire AJAX web site dedicated to designing DC-DC converter circuits FOR YOU even down to giving you DigiKey BOM exports!

          The Internet (data sheets as PDFs on the web), google, and cheap PCB CAD software and fab services have made the modern Maker movement happen. Anyone can prototype and productize damn near anything all by themselves.

          • Jason

            The quotation for the hardware was 27K USD (without even getting one prototype), I thought well if it is that expensive I can buy the equipment and software which is needed for being able to design such hardware. At the end it turned out the initial cost was 40k USD to be set up with good equipment (which also requires some special testing equipment which I was already able to use when I was doing the software for those devices in the past)
            The software for it took 5 years, hardware took 2 years (both starting from zero).

        • Jason

          It’s hardware and software both are complex, well the hardware is easier than the software but still takes time.

    • http://truthspew.wordpress.com Truthspew

      Dave – Very interesting write up! Of course the distribution houses are going to stack the deck in their own favor. They’re trying to make a living too.

      That said, it’s not hard to do your own mailings particularly if you have a scale, a way to print the postage, etc.

      • http://www.eevblog.com EEVblog

        Of course. I have nothing against the distributors, they run an essential business. I have used them before with great success, and will likely use them again at some point.

      • http://www.eevblog.com EEVblog

        The key is efficiency. I’ve now got mine down to being about as efficient as it gets without actually using a mailing house. But it wasn’t easy, and it costs money to operate.

    • Darren Grant

      There is one disadvantage to your customers is that some of them outside AUS will get hit with handling charges by their local postal service for the privilege of collecting local sales tax. Here in the UK are charged £8 (almost AU$15) + the estimated VAT due before they will hand over the package.

      Now that might not be a big issue on a £200 item but a £20 item that means almost a 50% surcharge. That is one advantage to having a local agent sending out products in a particular territory. I imagine also that the cost to send 100 international parcels is significantly more than sending one large parcel that would then be sent using local post.

      At the end of the day though it probably still works out cheaper for the end user if the big distributors are taking such big margins. Sounds crazy to me.

      • Queue

        Wow! I live in the US and have done plenty of shipping to Europe. I was well aware that buyers are on the hook for VAT but had no idea that the postal service charged a fee to collect VAT. Is this unique to the UK or is it common in the EU?

        • Darren Grant

          It really depends on the postal service but if you use something like UPS DHL or FedEx then yes they charge a handling fee for collecting the tax. I can see from their point of view that it is extra admin but the charges are often excessively large.

        • http://hackedgadgets.com Alan Parekh

          Here in Canada all packages with a value of $20CAD or under just get delivered (when using Canada Post). If the value is greater they charge you tax on it and tack on a collection charge. It was $5 for the longest time but now I think it is around $10.

      • http://www.eevblog.com EEVblog

        Yes, that sucks. But the problem is organising to have a distrubutor in every county that has this issue (and there are lot of them) is a big job.

        • Darren Grant

          Within Europe you have the advantage that if a package is sent from any member state sales tax is paid by the sender rather than having to have it collected and incur fees.

          • Queue

            Not to delve too deep into taxation but this is one of the things that makes the US and Europe so different with regard to VAT/sales tax. In the US, generally if someone mail orders a product from an entity out of state (or internationally) the onus is on the BUYER to pay sales tax. Many (most) Americans don’t realize this and consequently don’t pay tax. Even though we don’t pay tax when we take possession of our uCurrent, come April (when we pay income tax), we are supposed to pay what is called “use tax” on any items we have purchased that we haven’t already paid sales tax on. A few states don’t have sales tax so this doesn’t apply there.

            • http://www.eevblog.com EEVblog

              Yes, taxation in different countries is mind boggling. I might get asked to do something special for someone’s particular country, not realising I’d typically sell into 70+ countries for a big campaign like I had on Kickstarter.

      • SydB

        It is hit and miss. My uCurrent arrived this morning (thanks Dave) with no taxes or duties being asked for. In my experience, CN22 customs items (i.e. value below aboutu £250GBP) in regular mail often slip through the net but CN23 items through a courier are charged.

    • http://www.geppettoelectronics.com/ Nick

      Here’s what I struggle with.

      I have designed some gizmos that I really believe in. But because I’m nobody, marketing them on the Internet is like pissing in the wind.

      Because I’m nobody, and because I’ve only sold a handful of anything, I’ve been having boards fabbed at a basically single-unit level and manufacturing them by hand (manual pick-and-place and reflow), and buying small quantities of parts and so on. So my costs are quite high.

      If I had some product that I had a reasonable expectation would sell in quantities of a hundred at a time, I’d instead buy T&R quantities of things like ATMeta328Ps and outsource the SMT assembly. Getting, say, 500 of one of my widgets made would probably reduce my unit cost by 20%. But if I do that and sell exactly 1, then I lose my shirt.

      • TomC

        Why didn’t you provide a link to a product page in your post, Nick? Do you have a sig line with such a link in your forum posts, across all forums you participate in? If you believe in your product, have you sent one to Dave for a mailbag show?

        I’m perhaps preaching to the choir, but one really has to sell, sell, sell. What have your marketing efforts been so far?

        • http://www.geppettoelectronics.com/ Nick

          I actually have sent one of my widgets to Dave. I believe it’s in the mailbag queue as I write this. I have a blog and a store and I am moderately active on twitter. The main project I have is the J1772 Hydra, and I am very active on several EV related blogs and the OpenEVSE project.

          I am sort of reluctant to just outright spam the world crowing about my (meager) achievements. I probably could do more, but the bootstrap for a product is very painful – low production runs mean high costs, which translates to higher pricing, which pinches demand.

    • Pingback: ACRONET – Economics & Open Hardware

    • TomC

      Excellent post. Adafruit has a more detailed treatment at https://learn.adafruit.com/how-to-build-a-hardware-startup.

    • tom

      I am humbled by how dedicated and open you are about the process. Wait to go boss.

    • http://www.myrocontrol.com Danny

      Great post Dave!

      I’ve been dealing with this issue for years now with my very own products the most recent is an AirPlay/UPnP audio media streamer that has the flagship Wolfson WM8741 DAC (http://myroair.com). Manufactured in small quantities and made in the USA with US sourced components (e.g. not through some no-name supplier that typically provides counterfeit ICs).

      The balance between selling direct or through distribution is something I have been debating at lot lately — what you mention as pros and cons is spot on. I’ve chosen to sell direct because I can maintain the relationship with the customer and hopefully provide a better experience (purchase and support) as well.

      The challenge is marketing and moving product — word of mouth is much slower…

      Best,
      Danny

    • Robert

      I reiterate that Dave’s discussion what spot on.

      For people who have never been involved in commerce it’s hard to understand why the distributors take such a large cut.
      Now having worked as a marketing manager (my first career was an EE though) I can say there are good distributors and bad ones.

      When you go with a distributor, you potentially get access to their customers indirectly. Dave, over many years, has built his own customer base or list of people who potentially can but his products. It’s taken him years – how much money (ie time) has he spent doing this? With a distributor you get this straight away.
      The distributors also do the advertising. Sure your products is going to go on a website or a few square inches in a catalogue. What would that cost you to get to so many people in this form?

      Dave has calculated where his transition point is from using distributors to going it alone.
      Not that many businesses ever reach that transition point and that’s why they never transition from being more than a backyard enterprises

      Last comment which Dave touched upon slightly
      At the end of the day in many instances its about making money and turning it into a full time business
      Your retained earnings are gross profit per unit sold times units sold
      Going through a distributor if you are not an established name usually means more units sold than going it alone, but at a lower gross profit.
      Do the sums and work out whether using a distributor or not results in more retained earnings. That’s really what the core of marketing is (not selling)
      Robert

      Robert

    • Pingback: The Economics Of Selling Your Hardware Project – EEVBLOG [Reblogged] | Protoneer.co.nz

    • http://www.electrodacus.com electrodacus

      Great post Dave. In order to use that 2.5x multiplier you will need to know how many units will you expect to sell.
      Using Kickstarter and similar crowd funding sites has the advantage that you only need to deliver if there is enough interest. Still you need a prototype and if your product is something complex that will take time and money so you can still have a loss if project not successful.
      I’m just working on my Kickstarter project that was successful but a the limit is really hard to advertise and Kickstarter is no help in that for me only around 20% where directly from Kicksatrter browsing the rest where from my direct advertising.
      And considering their commission not sure there is any benefit in using those websites unless it helps a bit with credibility (not sure that is the case).

      Dacian.

      • http://www.eevblog.com EEVblog

        Kickstarter and other crowd funding sites, whilst offering extra marketing, also offer a psychological factor that is hard to quantify.
        People seems willing to throw money at crowd funding campaigns they ordinarily wouldn’t touch if you were just taking pre-orders on your website.

    • http://hackedgadgets.com Alan Parekh

      Hey Dave.

      Have you looked into Fulfillment by Amazon? Might be a cost effective way to allow US customers to get their product in a few days. In the end with potentially reduced shipping rates it might work out to be cheaper for all.

    • Stefan

      Dave, one thing that I was always wondering about: How much do you account for returns, lost shippings, broken arrivals at customers? Or if this is included in the 2.5 cost multiplier: How much does this reduce the margin?

      I was wondering about that since I received the eevblog uRulers in this special (probably expensive) tear-resistant envelope, after you noticed that quite some rulers got lost during shipping.

    • Electroid

      Hi Dave,

      Great post, however I am not agreeing with you about the 2,5 multiplier. I think you should never use a percentage (because thats what the multiplier is) of the manufacturing cost for deciding about the selling price. What if you have little hardware but a large microcontroller with a lot of software in it? Then maybe your hardware price is as low as $5, but you can never sell it for $12,5 as you make $7500 a year then (assuming you sell 1000 a year like in your example).

      I think it is really hard to say what margin to add on the initial costs if the bottom up method is used (that is the manufacturing, shipping etc. etc.). It depends on a lot of factors like the estimated selling amount, the amount of development hours to engineer the product and maybe even the most important part: what someone wants to earn :)

      • http://www.eevblog.com EEVblog

        Of course there are a lot of factors as I said. The X2.5 factor is a minimum value you should be looking at to produce a viable business.

    • Pingback: The Economics Of Selling Your Hardware Project | HudsonWerks

    • Jane Martin

      nicely done, thanks for share
      For more information visit our website

    • http://www.nrdcentre.com Ravi Saini

      Electronics & Electrical Module & Final Year Projects in india

      website : http://www.nrdcentre.com
      Like Us at : http://www.facebook.com/nakshdesign