Many early computers were sold as self-assembly kits (you get a box of parts and an instruction leaflet and have to solder them together yourself). For example, the Altair was priced at $439 kit, $621 preassembled.

The obvious rationale for offering this option is that if you are willing to do the assembly work, this saves the vendor the cost of labor to do it, so it's natural to pass on the saving to the customer. But it's not clear to me that this stands up so well to closer scrutiny. Some percentage of the time, the assembled computer won't work. The vendor, having no way of knowing whose fault this is, has to take the machine back and try to debug it. This quickly adds up to much more expense than it would've been to just assemble and test the machine before sending it out, particularly since if every machine will be sold preassembled, the vendor can invest in things like wave soldering equipment that can do the job much more efficiently than a user doing it once on the kitchen table. Tandy, for example, considered selling the TRS-80 in kit form, but concluded it would be more trouble than it was worth. But if Tandy didn't think it was worth the trouble, why was it that other vendors did?

Possible explanations:

  • Tradition. The personal computer industry developed from roots in the electronics hobby, where magazines such as Popular Electronics would publish projects, instructions on how to build something interesting; that is exactly how the Altair was first announced to the world. Originally, you would have to source the components yourself; a kit with all the components in a box, was a step up from that. It just takes time for tradition to change.
  • Market segmentation. To maximize overall profit, you want to be able to charge a higher price to customers with more money, e.g. business versus hobbyists. Kit versus preassembled was a way to do that.
  • Add an egg. There is a famous, possibly apocryphal but plausible story, about instant cake mix that sold better when it was made less comprehensive, requiring the customer to add an egg; this was interpreted as a finding that people like a product better when they have to put some work into it and get to feel some pride in creation. By this way of thinking, perhaps some hobbyists actively preferred soldering their computer together themselves, and would be less interested in a preassembled machine.
  • Speed. Yes, it's ultimately more efficient to do all the assembly in the factory. But hiring workers and setting up factories takes time. Just shoving the components in a box lets the business start ramping up quickly, and that was what mattered in the early days of the industry, when companies like MITS were flooded with far more orders for the Altair than they could fulfill.
  • Something else I haven't thought of.

Is there any evidence – ideally something written down at the time, but I would also be interested in any relevant experiences people have, or any logic that I have missed – to indicate which factors were the decisive ones?

  • 25
    “The vendor, having no way of knowing whose fault this is, has to take the machine back and try to debug it.” Was that true in the US back in the late 70s and early 80s? Nov 20, 2020 at 12:19
  • 2
    @StephenKitt A fair question! Tandy behaved as though they believed it was, though I don't know to what extent they were legally obliged to do so, versus chose to do so to protect the value of their brand.
    – rwallace
    Nov 20, 2020 at 12:20
  • 11
    Tandy had a different relationship with its customers, through Radio Shack, which might explain a different approach to handling non-working hardware (and also make it easier to handle queries and returns). Nov 20, 2020 at 12:36
  • 22
    Self-assembly has a long history in home electronics. See Heathkit for example. For the Altair, MITS was already in the 'kits' business, and were not AFAIK equipped to supply fully-built products in volume.
    – dave
    Nov 20, 2020 at 13:14
  • 4
    With the technology at the time (in particular, no surface-mount components) there was no technical or reliability advantage of wave-soldering over hand soldering. (I have some hand-soldered devices that have been in use 24/7/365 for 30 or 40 years and still work perfectly).
    – alephzero
    Nov 20, 2020 at 16:34

9 Answers 9


Back in the 80s, it was an industry and market very different from today's PC world.

The audience were mainly do-it-yourself people who wanted to create things, and to understand how they worked. So, of your suggestions, "Tradition" best captures that aspect.

You'll also find that, in line with that mindset, writing individual programs was a main topic. Nearly every microcomputer came with BASIC pre-installed or at least easily available. That also meant "do-it-yourself", only at software level.

So, the customers were typically very tech-affine, and many of them came from hobby electronics, having the necessary soldering skills to get the computer up and running. I did that myself, together with some friends plus some support from the equipment of a university EE department (oscilloscope, EPROM programmer etc.). We really liked to do it ourselves, and of course we were always tight on budget.

Compare that with the typical PC customers today. Have them solder their computer? Write a computer program? Nope. Most of them won't even be able to install an SSD. So, selling computer kits doesn't stand a chance today. Even building your computer from components like motherboard, CPU, RAM and graphics card has become a niche market.

But we currently see a revival of the do-it-yourself spirit in the Maker scene, where people create their own controllers (based on Arduinos and similar processors), doing hardware as well as software.

And, as a funny side-note, an Arduino's capabilities are quite similar to the early microcomputers, so, after more than 30 years, I find myself again thinking in kilobytes instead of the gigabytes that we typically waste in our desktop PCs.

  • 3
    While I like Arduino, and think, for instance, that Raspberry Pi is overkill, running Linux as it does, the BBC:microbit (with one given to each 11 year old UK schoolchild) is more in the classic tradition (more like a model B, which came pre-assembled, rather than a Model A, which the buyer assembled), but I have fallen in love with the ESP32, with WiFi and BLE from about US $3. It makes me feel *cough* years younger :-)
    – Mawg
    Nov 21, 2020 at 20:36
  • 6
    The 1980s were definitely not the era of user-soldered PCs. (Maybe some still in 1980, but by 1981 user-soldered kits were definitely niche. The typical hobbyist assembling his own PC was seating DIPs into sockets, not soldering the sockets.
    – RonJohn
    Nov 21, 2020 at 23:00
  • 3
    Yes I agree customer expectations are key. In the 70s in the UK the target market were people who wanted to build a computer, e.g. those who were into electronics. After all the computers didn't do much, so the fun was in making them. There was an overlap period around 1980 when you could buy a ZX80 as a kit or pre assembled. Only when computers became useful did more people want to buy them ready assembled. Nov 22, 2020 at 12:15
  • 3
    @MawgsaysreinstateMonica If you're referring to the BBC Micro with "Model A" and "Model B", the Model A was not a self-assembly kit. Maybe you're thinking of its predecessor, the Acorn Atom, which was available both in kit form and ready-assembled? Nov 22, 2020 at 13:45
  • 2
    Thanks for that (+1). It has been so long since I bought my model B that I somehow have the impression that A was build yourself & B was pre-assembled. Obviously, I am wrong.
    – Mawg
    Nov 22, 2020 at 14:14

I believe that your presumption is incorrect:

The vendor, having no way of knowing whose fault this is, has to take the machine back and try to debug it.

The vendors did not have to take the machine back and if they did offer a repair service it was not free. I don't recall that MITS (vendor for Altair) offered a repair service but I know that Heathkit did as well as SWTPC (Southwest Technical Products Corp.). You could send your non-functional self-built kit to them and they would fix it for a price depending on how much time and material was needed to get it operational.

While I never purchased an Altair, I did purchase a number of other kits from that era and they were always sold "as-is". Any issues were the problem of the user, not the company. Kits always came with schematics and other relevant documentation so that it was possible for end-users to debug, diagnose, and repair. There were also local user groups which were invaluable for getting help from more experienced kit builders.

  • 8
    Heathkit of course is a different case since their entire business was selling electronic kits to novices at building electronic kits - ham radios, shortwave radios, entire TV sets (in the CRT days!), entire electronic organs (!!), etc. It was a key selling point that they wouldn't let you fail - otherwise how could they get you - someone who had never picked up a soldering iron - to buy their gear? (And thus, the costs of fulfilling their promise to some customers must have been baked into the cost of their kits for all customers.)
    – davidbak
    Nov 20, 2020 at 14:49
  • 5
    P.S. A Google search for their Wikipedia page left me with the astonishing surprise that the company has been revived and is selling kits! Go here for modern-yet-retro electronic kit goodness! (Including kits to teach you how to DIY with SMT! I'm gonna try that now ... always been hesitant ...)
    – davidbak
    Nov 20, 2020 at 14:50
  • 5
    @davidbak It was certainly not "free" from Heathkit. An individual hobbyist could have bought the parts from electronics retailers much cheaper than the cost of the kit. You were paying insurance up front to cover the cost of fixing your mistakes.
    – alephzero
    Nov 20, 2020 at 16:31
  • 2
    @alphazero - Oh, well, if you're talking economics or something like that ... I can assure you I had no knowledge of anything like that when I was slobbering over Heathkit catalogs when I was 12 ... saving up and eventually buying and building a DX-100 ...
    – davidbak
    Nov 20, 2020 at 16:54
  • 1
    @alphazero - Oh and by the way ... I used to get my catalogs from Allied and Layfayte (sp?) electronics distributors ... their prices for small quantities (ones or twos) of anything were of course horrible - even 10s of resistors or caps of a certain kind and value - not to mention there wasn't free shipping then ... - and let's not even talk about Radio Shack prices ... - you couldn't actually buy components for one Heathkit kit for cheaper than the kit - I think kids trying to build their own gear then had adults (in clubs) to connect with who'd give them stuff out of their stock ...
    – davidbak
    Nov 20, 2020 at 16:57

if every machine will be sold preassembled, the vendor can invest in things like wave soldering equipment that can do the job much more efficiently than a user doing it once on the kitchen table

I think this is a key point. Back in the 1970s, you did not have the super-cheap build-it-in-China and ship extremely inexpensively that you have today. Container shipping. Electronic transmission of specifications. Large job shops set up for electronics. A lot of different pieces have evolved over the last 40 years, that make a huge difference in producing an assembled consumer electronics product at a low price.

This also fits with a huge factor, particularly in the days of Altair & Heathkit (less so with Heathkit as time went on and they had experience/market knowledge): Nobody knew how many products they would sell: 100, 1,000, 10,000? If your only bespoke piece is a motherboard (and I mean only the board, not populated/assembled) then your up-front costs are pretty low. Nearly everything else is buy 'x' quantity of each chip, transistor, LED, etc., and put them in bags, make more copies of the user manual (couldn't just laser print them, but copy machines were available), etc.

If you are really confident that you will sell 10,000 of a product, it paid to get it done in a big factory (whether US or overseas, depending on timing and other factors) that has big equipment, or you actually invest in some of that equipment for your own factory. But if you have no idea how many you will sell, "kit form at a discount" makes a lot of sense. Plus charge enough extra for "assembled & tested" that you can afford to do that at a profit, even if you have to hire hourly workers on short notice to do the work.


Many early computers were sold as self-assembly kits

Not really. There is a visibility bias. Kits were simply more visible than ready made machines.

Kits were simply the best way to fill pages of magazines - and that's what most people saw and still find on the net - thus they got more 'screen time' than ready units. Especially early on, when there were no (semi-) professional publications for computers, but a well sorted assembly of electronic magazines which featured computer kits as just another kit to be build - among electronic dice and water level sensors for rain collection.

In fact, when reading thru electronic magazines of the mid 70s, one can find quite a lot of different, ready made computers - most of them complete forgotten by now. Those ready made machines simply didn't need lengthy articles about how to solder a transistor and how to add some interface from scratch.

It is these magazines we see (and some remember) when looking back, thus adding bias toward kits, as they came with a lot of magazines and articles to support them. Browsing thru professional publications of the same time gives a complete different picture of microcomputers.

It's as so often: Looking back gives us a biased image that does not fully reflect history. That's why serious history work needs more time than skimming a Wiki article or opening a single issue of Popular Electronics (Ask me why I know :))


At the time the numbers of unit made were was smaller than today, and technology for assembly electronics was way less advanced than today. Was a labour-intensive process so shifting the work to the end user made sense, especially if the end user was an amateur without a lot of money, but some spare time and a love for electronics.

Add to this that was a totally new market, so nobody know exactly what product should have be done. Selling an electronic organ was a more streamlined process, music shops already sold Hammonds and Rhodes, so having a Farfisa or a Vox wasn't a big deal because they were similar things with similar customer base.

A microprocessor based home computer was a pretty new thing, event if at the time there were some "personal computer" from Olivetti and HP but they were basically programmable calculators and were designed for business and scientific use. Olivetti made in 1976 the P6060 that was based on 8080 CPU, had a printer and a 8 inch floppy disk integrated but was really expensive and designed for banks and the like.

Next, with Commodore PET, Apple II and TRS-80 home computers and office computer converged, due the decrease of costs, the units sold were more, there was a market for them, so a less labour intensive manufacturing was possible.


The difference is the 2 years from 1975 to 1977 (when the Apple and TRS-80 came out). A 1975 Altair was programmed in opcodes using toggle-switches, and the output was flashing lights. Whee. By 1977 we had video, BASIC, keyboards and the ability to save programs on tape.

The only reason to get an early 1975 Altair was to mess around with it as a technical artifact. Programming it with real hardware codes was cool (I saw one once as a kid -- it was). But you probably won't get your family interested. But as other excellent answers have noted, electronics skill was cooler and more common then. Putting cutting-edge chips together was the rest of what made it cool. My engineer dad built an electronic organ around then and he didn't even play.

By 1977 the software side was much more friendly, accessible and entertaining. Neighbors who might watch the Altair's lights for a few minutes would wait for a turn to play TRS-80 Hangman, loaded from a tape once typed-in from a book. Your kid could realistically learn to program it. That market was about the end-product being fun. Like all products designed to be used, "some assembly required" would have been a drawback.


Predictable volume was an unknown. This stuff was so new, nobody knew how many would sell. And for many who dared make them, market estimates were way off.

Mass manufacturing requires knowing the volume so that the tooling investments can be amortized. Tooling for manufacturing also adds lead time, by which time you might be "scooped" by some competition. Whereas kits, you could advertise as soon as you could verify small quantity parts availability. Then just kit up the parts and ship them out as the orders come in. Very little inventory, and no underutilized tooling or finished inventory risks (bank loans, interest, obsolete product, etc.). New designs could use much of the same old inventory, so not a lost investment.

As for repairs, many kit vendors would only do that for a fee, if at all, e.g. another revenue center.

Whereas currently, computer companies have huge marketing departments, with tons of people working on sales estimates, dynamic cost of sales spreadsheets (which didn't exist back then!), just-in-time inventory management and operations, and etc.


Another reason—which applies today (but probably also back then) and only has a retrocomputing angle in that it applies to reproduction kits like the Minstrel or Harlequin—is that finished consumer products have all sorts of regulations (such as CE-marking and WEEE) whereas a kit sidesteps all of that because the recipient isn't considered an end-user. The idea is that B2B transactions within a supply aren't covered by consumer red tape, and it's only the final sale which needs it.

In this case, it's a fiction that the kit-buyer is a business who will assemble it and deal with the paperwork, but hobbyists playing with electronics are such a tiny market that this obvious loophole doesn't need to be closed.

Consider that back in the day, the Apple 2 omitted a TV modulator because it did not want to faff around with FCC certification for it. A kit is just the same, but with more soldering and swearing.


In many cases, you can maximise profit by having one price for many people, and another price for a few - you just have to somehow achieve this. If you have one price, then the number of customers gets less as you increase the price, but the profit per item grows, and there is an optimal point. If you manage to offer a different price point at less convenience, then you will get more customers and more profits, without everyone going for the lower price, which costs profits.

That's why restaurants have "early bird" offers, shops have vouchers, because they get customers they would otherwise not get, without losing profits for the majority of customers. Same here; they sold more computers this way, without losing on customers who were willing to pay the higher price anyway.

Today you couldnt do this, because very very few people would be able to assemble say a MacBook from parts, and even fewer could assemble it and get a working computer.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .