Building Machines vs. Buying Them

Building Machines vs. Buying Them

 - Lessons Learned, Episode 54

 

When I started my first IT business in 1995, enterprise businesses (really big companies) bought pre-built machines, primarily from big companies like HP and IBM while small companies built their own machines or hired consultants who built machines. So, at the time, it was very easy to move into the business of making machines. These “IBM Clones” were everywhere and really were the standard.

This affected my business in two primary ways. First, I had to buy parts and keep an inventory. Back in the day, motherboards did not include integrated video cards, sound cards, or even network cards. So there were lots of peripherals that had to be bought, kept in inventory, and sold before the prices dropped again and we lost money on our investment.

The second affect was that almost every business we went into was filled with “homemade” machines, including the servers. Every one was unique. Very often, we came across brands we’d never heard of for all the components: Motherboards, SCSI cards, hard drives, and so forth. In many cases, this meant that we did not have drivers, instructions, warranty information, etc. In other words, if anything went wrong, it was likely that we had to sell the client a new one – the brand we carried.

Clients had every right to like something was wrong with this business model. Every tech they talked to wanted to get rid of “the other guy’s stuff” and install their own. That just felt like an unnecessary expense to the client. And in some sense it was unnecessary, except that there really weren’t a lot of options, given the way things worked.

There was a massive, worldwide industry of making and selling IBM clones. So much so that Microsoft included questions about OEM installation of their operating systems in their certification exams. In addition to the parts, you could also buy customized tags so you could easily create your brand of machines.

This model had a few big disadvantages. Inventory was one. Another was that we had to maintain a place to build and test all these machines. We frequently had tens of thousands of dollars’ worth of machines and equipment in our office, which always made me nervous. It also meant that we had to train technicians to build machines properly, which is pretty far from the calling of being a consultant.

Worst of all, the margins on these machines were quite thin and the prices were changing every day. It was extremely common for advertisements to list “CALL” as the price for many components. It was like trading Frozen Concentrated Orange Juice Futures. And if you bought the right components on the wrong day, you were guaranteed to lose money. The worst was memory modules – a circumstance that’s repeating itself today.

Finally, the big problem with these clones was that we might lose all of our profit for a machine if one component (such as a fan) went out and we had to pay a technician by the hour to replace it. Not only were we out for that job, but that tech was not doing something else that did bring in money.

Even at the beginning, I wanted to sell pre-built brand name machines. I had come from a shop full of Compaq/HP machines and that brand was extremely reliable at the time, with spectacular support. But I wasn’t an authorized reseller and they required huge quarterly sales to even consider becoming one.

Eventually, HP brought out a series of pre-configured but customizable machines that we could buy if our distributor was willing to work with us on that relationship. So we started offering these pre-built machines. Yes, they cost more. But nothing ever went wrong. Nothing. When we bought and sold business class computers with a three-year warranty, nothing ever broke. It was like a whole new world.

We made up for the increased cost by not having to carry inventory, not having to train technicians to build machines, and not paying technicians to build machines. And the “nothing ever broke” piece of the equation really allowed us to focus our business back where it belonged: solving client problems rather than selling them stuff.

From time to time, I also sold Lenovo, Toshiba, and other brands. But we really had great experiences with HP and stuck to that, especially for servers.

I honestly believe that the decision to buy rather than build helped us take on bigger clients and bigger projects because we could put a strong focus on client problems and solutions. Our technicians had to learn to dig a little an find the “real” problem when I client said they had a problem. Our technicians became better consultants because they could focus on the problems and the solutions rather than on replacing parts.

Other people have had other experiences, but that was ours.

One time, I had a big (for us) installation of about fifty machines. I certainly didn’t have the staff to do this in a reasonable time frame, so I engaged members of my local SMB IT Professionals group. I gave them machines and checklists and they helped us install dozens of computers. One of these IT Pros was a good friend of mine who had always built his own machines. But that day he saw that installing lots of high quality pre-built machines was incredibly fast and straight forward – primarily because the experience was predictably identical from desktop to desktop.

When that job was over, he told me he’d decided to sell prebuilt machines from then on.

Today, almost no one builds their own machines. The parts are still often a hassle to buy, store, and try to make a profit on. Parts makers (including big names like Intel and Western Digital) change their policies, promotions, and partner programs on a regular basis. Some of them actually swing back and forth between wanting to engage partners and not wanting to engage them. The result is that finding good quality parts that are consistently available at a reasonable prices is just plain difficult. It’s a layer of expense most consultants don’t want to deal with.

Advocates of building their own machines always go on in great detail about how easy it is and how much money they make. But this always sounds like when people start talking about their gambling. Gamblers often tell you that they never lose (an actual impossibility) or that they are ahead for life (a virtual impossibility). In other words, they remember and report the wins. They don’t remember and don’t report the downside.

I’m glad we made the switch pretty early on and I firmly believe that it made our technicians better consultants.

 

What’s your experience?

 

Feedback always welcome.

-- -- --

 

Episode 54

This Episode is part of the ongoing Lessons Learned series. For all the information, and an index of Lessons Learned episodes, go to the Lessons Learned Page. 

Leave comments and questions below. And join me next week, right here.

Subscribe to the blog so you don't miss a thing.

:-)