You hear a lot about Sinclair and Amstrad and Acorn computers. But when it comes to British brands, it seems like we don’t hear a lot about Apricot. But thanks to a television program that aired in early 1990, we know a fair bit about Apricot’s triumphs and struggles in the highly competitive PC market of the 1980s and 90s. It was on June 30, 1999 that Apricot closed its factory in Scotland and wound down its brand.
I have to give credit to longtime reader Gary Osborne for alerting me to the existence of this TV show. In the early 1990s, the BBC had a TV series called Trouble Shooter. In it, Sir John Harvey-Jones visits troubled manufacturing businesses and helps uncover and solve their problems. It reminds me in some ways of the modern TV series Kitchen Nightmares, where Gordon Ramsay visits restaurants and tries to turn them around in a week, except these weren’t restaurants. In episode 3 of Trouble Shooter, Harvey-Jones visited Apricot Computers in mid-1989. The episode aired in 1990. Thanks to that TV episode, we know more about Apricot’s inner workings than we do about the typical 1980s and 90s PC clone manufacturer.
What was Apricot Computers?
Apricot was an interesting computer company. They were the first company to ship a 486-based PC in 1989. Arguably that’s not quite as noteworthy as being the first to ship a 386-based PC, but they still beat all the big names in shipping a 486, including Compaq and IBM. Apricot was also noteworthy in licensing Microchannel from IBM. While Tandy and Dell announced Microchannel PCs, Apricot actually shipped them in quantity and stuck with the architecture for several years, selling them alongside conventional AT designs rather than choosing one or the other. And Apricot was interesting in that they bucked outsourcing design and manufacturing to Asia longer than many PC makers did. They designed their PCs in Birmingham, England, and manufactured them in Glenrothes, Scotland.
The challenges Apricot faced were not at all unlike the challenges companies attempting to design and build computers in the United States circa 1989 faced. It was much cheaper to build computers in Asia. And it was increasingly cheaper to outsource the design too. In that sense, Apricot’s struggles designing and building computers in the UK didn’t seem at all unfamiliar to me.
Apricot on Trouble Shooter
Harvey-Jones described Apricot’s problem as a minnow trying to swim with large multinational whales and sharks.
Apricot’s founder, Roger Foster, said the maintenance was the most profitable side of the business, but that it existed because Apricot sold hardware. The hardware side was not profitable. Harvey-Jones then turned to see if the software side of Apricot’s business, which was also profitable, relied on the hardware manufacturing. Of course, he found it didn’t. Apricot’s software ran on other IBM PC-compatible systems too.
The problems with UK-based design and manufacturing
What Harvey-Jones found was that research and development was expensive, and it fed into the hardware manufacturing, but not into the software or maintenance sides of the business. Harvey-Jones’s idea was to raise prices to cover the overhead and raise profits. When that proved a nonstarter, he visited the factory to see if cost reduction might be possible.
As he entered the factory, Harvey-Jones encountered a display of Apricot computers. He asked if that was the historical progression of the product line. An unnamed Apricot employee said no, it was the current product line.
He was audibly taken aback at the large number of models they produced. There were seven machines visible in the camera frame and the framing implied there were more machines out of frame. The employee said they produced both MCA and AT lines of computers. That subtle point is easy to miss. Apricot was making both Microchannel PCs, like IBM, and ISA-based AT machines like Compaq and everyone else.
Letting go
Harvey-Jones visited trade shows and talked to experts outside the company to try to get an understanding of the industry. Guy Kewney, a computer journalist, said you can’t compete with IBM by offering more at the same price. You have to offer more at a lower price, because of compatibility. He said he’d been talking to five different users of clone machines who were running into compatibility issues, so you have to save a significant amount of money to be willing to go through that.
The show ended with Foster concluding that Apricot needed to shift its emphasis to software and maintenance, and making an acquisition to shore up that side of the business and setting up a new company, ACT, toward that end. And he also said Apricot needed to find a strategic partner for the manufacturing side of the business. The episode closed by saying Apricot was close to selling its hardware division, most likely to Mitsubishi.
In April 1990, Apricot did indeed sell its hardware business to Mitsubishi Electric, giving them access to overseas design and manufacturing.
The rest of the company, known as ACT Group Plc, was acquired by Misys Plc in July 1995 at a cost of 212 million British Pounds, forming the largest all-British software and services company.
Even hindsight doesn’t make it analyzing Apricot all that easy
It’s easy to say today that Apricot could have reduced costs by jettisoning Microchannel. But in 1990, it wasn’t exactly clear Microchannel was going to lose. I knew someone in 1990 who owned an IBM PS/2 Model 30-286 and he said it was a compromise. Microchannel was faster, but some expansion cards weren’t available as Microchannel, so he went with the ISA-based Model 30-286 instead of a Model 50 or Model 50Z and he felt like he’d made the right decision about half the time.
One of the changes Steve Jobs made when he took over at Apple was reducing the number of models Apple sold, so a reduction in models seems obvious in hindsight. But there is an argument for the large number of models that made Harvey-Jones cringe on sight. Most likely Apricot had a 286, 386, and 486 in both ISA/AT and Microchannel architecture. That’s six models right there. They needed a base model, a midrange model, and a high-end model to compete. If they were going to offer two architectures, they needed six models. I couldn’t tell from the video if there were eight machines in the montage at the front of the factory or 12 or more. Offering multiple high-end machines, where the profits were higher, wasn’t terribly unusual either then or now.
There was no easy or obvious solution for Apricot in 1989. Their management said near the start of the episode they hoped an outsider could find something obvious they were overlooking. If there was, I can’t see it either, even with 35 years of hindsight.
David Farquhar is a computer security professional, entrepreneur, and author. He has written professionally about computers since 1991, so he was writing about retro computers when they were still new. He has been working in IT professionally since 1994 and has specialized in vulnerability management since 2013. He holds Security+ and CISSP certifications. Today he blogs five times a week, mostly about retro computers and retro gaming covering the time period from 1975 to 2000.