If you read comp.sys.next.advocacy, talk to NeXT's customers, or talk to resellers, the talk always comes around to one thing: how bad NeXT's marketing is. NeXT have always claimed that their marketing is carefully targeted, and will only be visible by those to whom it is directed. And customers counterclaim that, if it is so precisely targeted, how come they have never seen it?

This is the famous "stealth marketing" story. It highlights a lesson that all of us trying to earn a living with NeXTSTEP can profit by, which is how to reach the real marketplace on the minimum number of dollars. If your aim is off, no matter by how little, you won't get there.

So this is a story about marketing, so much is obvious. What makes me qualified to preach about this, when I am claiming that so many others have it wrong? First, I have spent 17 years full-time in the software business, 11 of those as a Sales and Marketing Manager for global software companies. Second, all of the little marketing classes in an MBA must count for something, if only a neat theoretical background. And lastly, I am currently running a successful NeXTSTEP-only business (the only one in the UK).

I want to put the case that there is a directly parallel for the market structure that we find ourselves in; and that lessons learnt from pursuing that market will tell us how to reach our market more effectively.

Over the years that the NeXT market has existed, it has been through many changes. We can all remember the machine for education, the publishing system, the Super PC, and now the OO developer workstation for Mission Critical Custom Applications.

In the near future it is likely that things may change as the business shifts towards OpenStep; I can see that most NeXT third party applications will die off quickly unless they can compete effectively in the Sun market. This will place greater emphasis on NeXTSTEP specific development tools and objects/palettes.

It is probably true that the MCCA market is the real NeXT market, at least for now. Most of the money earned, certainly in my business, comes from customers in this position. What we sell may be largely office automation applications and training, but the most compelling reason for purchase, the USP, was MCCA.

The only real income to the NeXT market comes from MCCA customers. And of these, most are in the financial community. There is something interesting that we see when analysing the balance sheet of these companies. Whereas traditional manufacturing or retail businesses have the largest sector of their assets invested in property and plant, NeXT customers have theirs invested in computer equipment.

As a small developer producing shrink wrap applications, NeXT third parties survive by being small, living cheap, and responding very rapidly to customer requests. In this business, a "small" (in PC or Unix market terms) purchaser can expect to influence product direction significantly, and talk directly to the company president and developers for relatively minor matters. It is the special properties of NeXTSTEP that allow us to do this.

In general, we sell directly to our customers and use service as the big differentiator.

The PC marketplace, by which so many NeXT people like to compare themselves, isn't like this at all. For this market, volume is the key. Companies can only exist through venture capital investment, because of the broad marketing that is required to break in to the market. Considerable sums have to be spent on magazine advertising, direct mail, indirect channel building, and other FMCG (Fast Moving Consumer Goods) techniques. This entails considerable rigidity.

PC techniques don't work in the NeXT market. I can't emphasise this enough. Every company, ISV or reseller, that has tried to work this way has gone bust.

Some time ago, the rich place to be was working with mainframes. From the early seventies, when IBM unbundled software, through to the late eighties, more money was made in mainframe software than anywhere else in this industry. Successful salesmen (all the best companies were sales led) made more money than hardware salesmen. PC vendors started hitting the top in the late eighties: but most of the pack were still mainframe.

We had small lists of potential users, which we worked hard. Most selling was a mix of telesales and face to face; supported by direct mail, seminars and other highly focused techniques. Sales cycles were typically 3 months to 18 months, depending upon how close you kept to prospects; the closer you were, the longer the sales cycle seemed, because you saw more of it. You also sold more. Typical sales values ranged from $5,000 to $20,000 (for telesales), to $60,000 to $200,000 and more, for the longer cycle close touch sales.

Most products had annual maintenance fees (15 - 25%); a few had annual leases. Sales commission 10%, but with bonuses.

Evaluations were common, usually in the 30 to 90 day range.

To make sure of sales, support was all. The big differentiator between a winning and losing product was the quality of support given to the customer. Trust was essential. Good support would allow a lesser, more expensive product win over a better product. If the customer can't use the product effectively, then it isn't worth much. This is an expensive strategy, requiring big sales.

My favourite example of this was selling a $135,000 product to one customer over a (probably better and easier to use) $19,000 competitior because our product arrived with an engineer who spent three days on site being useful, and the other came in a mailer.

Internationally, the way it worked was with either small subsidiary offices, paying back 25 - 45% of revenue to the parent, or with exclusive licensees/distributors. If the sub/agent weren't exclusive, they couldn't afford to give the support required; exclusivity maintained prices. Customers would often insist that a foreign supplier established a support office (or good distributor) if they were to buy the product. Good distributors worked with a compatible stable of products, offering increased sales opportunities.

A good NeXT sale is much like mainframe software, and nothing at all like PC sales. High value, high service, and (reasonably) high price.

NeXT ISVs are different as well. Some of the benefits of OO and NeXTSTEP are faster application development, smaller programmer teams, and greater flexibility. This means that it is feasible to bootstrap a software company in this market. Small teams and fast development allow ISVs to track customer requirements very closely. The small market means that there is less overhead involved in reaching the market, and a less crowded marketplace.

In the PC world, the marketing model means that companies have to seek venture capital. The barrier to entry is the cost of marketing required to create a niche, and the strength of competition.

The NeXT market allows developers to choose between self funding and customer funding for a new project. This is a big advantage. The only entry barriers are expertise, both technical and in dealing with the marketplace.

Companies that seek venture capital are too slow to respond to the real marketplace, and (so have) have sought to build to large an operation, one more suited to the PC model than NeXTSTEP.

The implications of this are clear. Companies that offer good support, track customer needs closely, and stay lean are the ones that win. One further element in the mix: you have to go out and sell to the customers to make money. Selling is part of the toolkit that lets you better support the customer, and better identify their needs.

Channel strategies based on PC or Mac models fail. Big channels with unskilled vendors price cutting are doomed to failure. Being "right sized" is a virtue. Too big a company will be too slow to react, and will have too high overheads to live between major sales.

Expenditure on adverts and consumer shows is a total waste of money. Targeted ads and shows can be OK, but this is too often money aimed at impressing your peers. Cooperative marketing, seminars and direct selling are the keys to success.

The best marketing is through your customers. Word of mouth counts for a lot. Imagine a situation that a customer asks for a new feature in a product. You agree to include that feature, and send out a test version inside of two weeks. That kind of response gets talked about.

Most of the NeXT companies that are still in business do this. They don't think of the differences from the PC market, they are just doing what they know works. I don't believe that NeXT recognise this. Their marketing has shifted a long way into this posture, but the model that they try and impose on ISVs is the PC one. They value size over expertise and speed of response or market knowledge. They try to force competition over accounts. This is the notorious "level playing field" strategy, usually invoked every time they choose to skew the pitch.

I'd like to see an acknowledgement that the NeXT market doesn't work like the PC market, and an acceptance of what this means. It's always better when you can carry on doing the same things that you have found work in the past, but understand why they work. And I'd also like to see NeXT recognise this, and follow a channel strategy that acknowledges our reality.