A New Way to Sell Delphi

In my last article, I wrote about selling development services. And of course, this got me thinking. (What doesn’t get me thinking is probably what you are asking about right now.) Anyway, it got me thinking about how to sell shrink-wrapped software. (A disclaimer first: I’ve never been involved with the production and sale of a shrink-wrapped product, but I have bought a lot of shrink-wrapped software. Plus, I stayed in a Holiday Inn last night). Selling software is an interesting endeavor. I think that the way it is sold needs to change, and that this change is being driven, like so many other things, buy the way the Internet affects everything.

(Another aside – what do I mean by “shrink-wrapped software?” I mean any software that you market and sell to individual customers. It need not actually be delivered in a box; it can be sold purely over the Internet, for instance.)

Selling shrink-wrapped software is hard. It’s tough figuring out how to price your product. Deciding what license to use, how to collect the money, how to deliver the product, what to deliver in the product, how to market it and whom to market it to are all difficult, challenging decisions. Can’t be easy to do.

So I was going to make up some fictional company to show what thoughts I’ve come up with, but I’m not going to do that. I’m going to talk about an idea I have for Borland. Now, I’m no expert on this, but I do see an obscene amount of customer feedback on the newsgroups, and I am a customer myself, so I’m not totally pulling this out of .. uhm, thin air. Yeah, thin air.

In general, Borland sells their products as if they were cars. Every year or so they product a new version, have a big announcement, and go on a marketing blitz to make people aware of the new version. Frankly, I think this an outdated and outmoded way to sell software. I also think it isn’t really what customers of software development tools want.

As a general rule, I think that if you are in the technology business, and you have been doing something the same way that you did before the Internet came to the fore, then you need to rethink the way you are doing it. How Borland sells software is a good example. Borland is selling software the same way they did before the Internet changed things. They are selling Delphi the same way they did before ubiquitous newsgroups made communication between users easy and copious, before eCommerce became the norm, before blogs made putting out information to customers a piece of cake, and before the ClueTrain Manifesto discussed and laid plain the need to change. Or, put another way, I think it is time that Borland rethinks how they sell software.

In addition, I think that Borland faces a unique business dilemma. Selling software development tools isn’t like selling other software. To a large degree, Borland’s customers rely on Borland products for their livelihoods. Whether it’s a consultant, a small development shop, or a large corporation, committing to a development tool is an important decision involving a lot of time and money. It’s fairly painless to change word processors. Changing development tools is a huge commitment. Therefore, Borland needs to sell their software in a way that recognizes this unique relationship that they have with their customers.

Now, let me be clear — I’m not trying to tell Borland what to do; I am merely making some suggestions, offering some food for thought, tossing out some cud for them to chew. I don’t have nearly enough information about Borland’s business to even begin to think that I could run Borland better than it is being run. I’ve frequently said that people shouldn’t try to claim that they know more or better than Borland unless they really do, and I’m not claiming that at all. That said, I do think I have some interesting ideas that they ought to consider. I am a customer, and I know what I want, and I see a lot of comments by customers and I think I know what they want.

Okay, here’s what I think: Selling software like cars is old-fashioned and needs to be changed. (Shoot, the way we sell cars in this country is nuts, but that’s another article…) Anyway, I think that the idea that versions of Delphi ought to be discreet, distinct events separated by time measured in years ought to end. Borland should consider the idea of selling Delphi only as a subscription – sell nothing but Software Assurance. Customers could purchase variable lengths of service, getting discounts for longer commitments. They could renew at anytime. Prices could be adjusted to ensure revenue streams aren’t altered much by this change. Customers could even pay a larger fee for individual updates.

Then, the really big change: Borland should plan and release quarterly updates to the product. These updates should include bug fixes, incremental improvements to existing features, and new features. Quarterly releases could be a goal, and not firm targets. New versions could be released as builds and features stabilize. New features could be implemented one or two at a time. The frenzy of producing, marketing, and selling a single release would be replaced with the task of selling a product as a concept and a commitment.

This is a win/win scenario. Customers would love this. They’d be getting frequent updates with frequent fixes to problems. They’d be getting a steady stream of new features, reducing the learning curve for any individual release. Bugs would be quickly fixed. The company’s commitment to Delphi would be clear, and customers really like clear commitment to products that they buy.

Borland would love it because their revenue streams would be smoother and steadier. The pressure on the R&D team would be lowered, as they would no longer be trapped in the frenetic cycle of pushing for a big release. Smaller, incremental releases allow for more flexibility and a steadier, more deliberate release schedule. The push to finish any particular release in a specific quarter would go away, because there would be a steady revenue stream. Features and fixes could be allowed to “stew in the pot” for the right amount of time because the pressure for any particular feature to be released at a specific point in time would be illuminated.

This change is needed because the Internet makes things move too fast. Long spans of time between releases of a product are not conducive to customer loyalty and satisfaction. The ease of distributing even large software packages has caused the marketplace to be more demanding of such frequent updates. Problems in software are made readily apparent to large swaths of a customer base because of the instantaneous communications possible on the web. Since news spreads quickly on the Internet, Borland needs to be able to respond just as quickly. Features get announced as vaporware, and Borland needs a vehicle to more quickly respond to such announcements. Everything is moving faster, and Borland needs to be able to move as fast as the folks in the left lane.

The time has come for a change in the way that shrink-wrapped software is sold. Making a commitment to a product and to customers by providing a steady, regular update to a tool is what customers desire. This is especially true for customers of development tools.