Service and Reliability Survey 1997

The 1997 PC Magazine Service and Reliability Survey reflects our readers' experiences with over 20,000 pieces of equipment. This survey will enable you to base your desktop PC, portable PC, server and printer buying decisions on something you've never had access to before: almost 43,000 years' of real-world use. Read on to find which brands excelled (and which didn't) in the areas of customer service and product reliability.

According to adverts in the press and the signs in computer superstore windows, we only have one thing in mind when we come to choose a new PC--lowest possible price. And yet, whenever PC Magazine asks computer buyers across Europe what influences their decision, low price comes near the bottom of their list. The price needs to be good, yes, but higher up the list of issues buyers want addressed are specification, performance, customer service and product reliability, and these are harder to measure.

The first two on the list--product specification and performance--are regularly addressed by our comparative review features. Detailed summaries of features and labs-based testing provide comprehensive buying information. Meanwhile, the last two on the list are addressed once a year in this, the UK's biggest survey of customer service and product reliability.

This year our Service and Reliability Survey is even bigger. We've always covered desktop and portable PCs, but now we've also solicited your views and learned from your experiences of using personal printers and of managing servers. The results are fascinating, but before we reveal who the winners are, let's see why we do a readers' survey, and take a look at the methodology we used to rank the brands in each category.

Why a readers' survey?

When evaluating reliability, the biggest problem is getting a big enough sample size. Testing one system and having it fail proves nothing. It's unlikely that every PC made by that vendor fails, so what would the failure rate be? You can only know if you increase the sample size.

Labs testing isn't really the solution. Even if PC Labs only chose to evaluate two brands, filling a test lab with enough systems to give statistical significance would require an immense amount of space, electrical power and human testers, not to mention the cost of leasing or buying the PCs themselves--not very practical.

Asking vendors to supply their own data on failures is unlikely to produce a fair result, because they all categorise failures in different ways. There might also be a little temptation to present the results in a slightly flattering light. We've tried talking to third-party support companies, but so many have support contracts with direct PC manufacturers that they too could gain from presenting the results more favourably for some brands.

So a readers' survey is an excellent method of collecting accurate, unbiased imformation, if used correctly. The computers that you, the readers of PC Magazine, use are clearly subjected to real-world activity, so it's difficult to accuse the survey of being based on irrelevant or false usage scenarios. And generally you don't have any vested interest in a particular brand or product vendor. You buy through a diverse range of channels, and reside in companies of every size. You often have a PC at home too, so this survey covers domestic usage as well.

Thank you

We must give a big thank you to all the PC Magazine subscribers who returned the questionnaire for this year's Service and Reliability Survey. By pooling your experiences and collective knowledge of the market, we've been able to produce our biggest survey yet. We hope that it'll be a guide to your desktop PC, portable, server and printer buying decisions, and that its fairness and rigour will encourage the bottom-placed companies to improve, while giving the top-placed companies opportunity to feel satisfied that their efforts don't go unnoticed by real users in the workplace and at home.

Finally, don't forget that computers will fail. No brand in any of the categories we covered had zero equipment failures. The winners of this survey aren't those who've eradicated failures, but the ones who give you excellent service when the inevitable does happen.

--Andy Redfern

Top of Page

Next