The major issue in the PC Card industry was and continues to be compatibility. This has a direct bearing on the distribution models, especially internationally. Looking at compatibility from an end-user's perspective is somewhat different to a PC Card manufacturer's. Firstly, there is a misconception that "plug-and-play" means stick the card in the slot and it'll work automatically, which isn't the case. Automatically recognising a card on insertion without any requirement for changing interrupt or memory allocations is only part of the story - when it works. The situation gets more complex when multiple cards, differing host systems, varying operating systems and application software are added to the equation.
Failures in the channel
But by far the greatest issue is the lack of knowledge within the channel which has resulted in end-user dissatisfaction with the inability to deliver timely and effective support. It has been overlooked that PC Card and portable computing technology is being purchased directly by end-users whose technical skills can be very limited. This is especially the case where corporate users buy this technology and bypass their Information Technology department. If the user bought a desktop, the LAN or modem card would be installed by technical staff but, with PC Cards, many end-users are left to install products by themselves.
Who should be supporting the end-user? Reseller margins in the highly competitive portable computing industry are between four and 12 per cent, which doesn't allow much support time. Deals of 50+ units are not daily occurrences, so we're talking about a lot of small-volume sales. The cost of a PC Card is low in comparison to the host platform and resellers view these as trivial products, until a problem occurs. In some respects, margin is a moot point as resellers simply don't have the skills to support PC Cards. The training required is as intense as many higher-value products and two days of technical support time are worth significant revenue in the services business. If the dealer is unable to support the technology, it falls back to the distributor.
Most distribution organisations are "time and place" operations, providing a warehousing function with the ability to supply product "quickly and cheaply". It would be unfair to say that these organisations have no support capability but it's very limited. Typically the support function falls back on the manufacturer, which is acceptable if you're in the US or Europe but not so good when you're in a remote market.
The market hasn't matured as quickly as anticipated and support for new users hasn't reduced dramatically. In addition, whenever a new hardware platform is released there seems to be an associated support overhead resolving new PC Card implementation issues. Consequently, the cost of support is becoming a vexing issue. Competitive pressures are driving down margins and this poses the question: how much support can you actually provide?
In real terms, even if margins were the same, the price of cards is falling. Modems are half the price they were last year but the cost of people, especially in the technical support function, is increasing. As an organisation, do you simply tell end-users not to call even when the channel is unable to provide the support? This won't build much customer confidence in the long term. Will Windows 95 lessen the number of compatibility issues? Our experience to date is that PC Card recognition has been improved, but it is more complex to resolve problems when they occur. Other operating systems are slowly migrating onto portables and this will no doubt bring a host of new issues to contend with.
I don't have any brilliant solutions to the problem. It would be great if all PC Cards would work automatically every time. Despite the advances in PC Card technology this is unlikely in the short term. If manufacturers want to compete in international markets in the long term they may need to consider putting in place electronic support systems, such as the Internet or international toll-free numbers to resolve technical problems. Even with the rapid uptake on the Internet, I think it'll take a few more years until users are comfortable depending on purely electronic interaction for problem determination.
Again, this leaves out the people factor which seems so important, especially in an iterative process such as resolving technical issues. If markets are large enough, another solution would be to contract organisations to provide local help desk functions. Staff will need to be trained and information will need to be provided in a timely manner by the manufacturer. Perhaps the best solution, however, is for manufacturers to look at different margin structures for export versus onshore business to allow or provide incentive for distributors to build and maintain a quality support infrastructure. The payback to the manufacturer, in terms of ensuring quality branding and reduction in support costs, should far outweigh any loss in margin.
Whatever the solution may be, there needs to be more focus on the channel by international manufacturers to ensure that a quality product is being delivered to the market. Apart from ensuring that there is adequate support by local distributors, manufacturers need to be much more pro-active in the testing and problem determination process prior to releasing products into the market. Testing needs to be completed across a wide variety of platforms and in mixed company (ie with other suppliers' cards). Problem determination information needs to be relayed to distributors when it is discovered rather than in response to an issue being discovered in the local market. Providing quality service to clients requires more focus on support rather than on sales volume. This is very much a system requiring equal input by manufacturers and distributors alike. If we don't provide the necessary support to the market then the future of the PC Card industry will continue to be plagued by compatibility issues.