Getting Real on the Web

by Richard Dean
Associate Director of
New Media Development
Strategic Interactive Group

Richard Dean serves as National Public Radio's Information Age Specialist for "Weekend Edition-Sunday." You can hear his segments broadcast on Sunday mornings.

Hype. Promises that can't be realized. Unrealistic expectations and timeframes for projects. Nope, I'm not talking about a Dilbert cartoon or your latest software product. I'm talking about the Internet. And at Strategic Interactive Group (SIG) in Boston, we deal with issues of hype (and much more) every day. Our clients are all big companies—the kinds of people you do business with every day—and they are all working feverishly to use this medium to their best advantage. These Fortune 50 companies are forming new divisions, integrating their backends, developing new "enterprise systems," and creating new marketing effort—all of which need to be integrated and speak common tongues.

You've probably read in The Wall Street Journal or New York Times about the demise of some of these corporate Web sites (not our clients, mind you). It's not terribly surprising that journalists see these "failures" as a sign of the Web's inability to be translated into dollars. After all, these folks at our major newspapers aren't known for their savvy regarding new media. It's not exactly their fault. They're not used to covering evolution in technology; and that's exactly what the Internet is doing—evolving. And before you go online, you need planning and the right tools.

Rule Number One: The Web is a Long-term investment.
The Web is about long-term investment and a solid business/marketing plan. Think about it in terms everyone can understand: you can make a killing in the stock market through short-term investments. You can also lose your shirt. The same is true on the Web. ROI online isn't just about collecting money—it's about not spending money on your 800 number, reducing catalog requests, buying fewer stamps, and most importantly, laying the groundwork for whatever the Internet becomes in 1998.

It's difficult to predict the Internet for 1998. You may not make money, you may not get more "hits" than your competitor. But one thing is certain: companies that aren't online throughout 1997 are going to be hopelessly behind in the following year.

Rule Number Two: Respect Your Marketers.
At SIG, we're often called upon to help companies design new backend systems. Guess who is at the table with the MIS folks? Ad agencies, marketing directors, customer service directors, and Web developers. The days of unhelpful and isolated MIS are over. And the tools these folks are evaluating have to be much better than the old proprietary systems, and way better than they are right now.

I got a call from a tools vendor here in Boston the other day. The sales guy claimed he had a product that could serve content to Web visitors dynamically. I head up the development group at SIG so I get calls like this every week, if not more often. Sometimes they're from reputable companies, sometimes you've never heard of them. Dynamic publishing systems are all the rage these days, as are HTML and site-creation tools. And most of them miss the mark pretty widely.

To figure out why, consider the market they're talking to. In the old days, there was a lot of selling to traditional MIS groups or software developers. Software was used by people who understood software, and decisions were made based on an internal, corporate-view of information. But now, there are all sorts of folks involved in decision making: marketing people (with little IS knowledge, usually), Webmasters, young Web programmers, and ad agencies. Very often the people driving these decisions are new to technology. And lest you miss the big underlying picture, consider this: The decisions these new Web-enabled folks make today are both re-defining their company's backends and replacing those clunky legacy systems. This is a monumental task, even if the end result is a little HTML page with a "submit" button.

Rule Number Three: Learn the New Criteria for Buying Hardware and Software.
Whether hardware or software, it absolutely cannot be proprietary. Those of you dealing with systems that can't handle TCP/IP know what I mean. We're building the new "legacy" systems now and none of these companies want to be locked-in. Its kind of like the saying in the banking industry: "No one ever got fired for saying 'no,' they only get fired for saying 'yes' when they should have said 'no.'" I don't want your "C-like" language, I want Java (or at least C/C++).

If it's a database, we'd like to speak natively, thanks. ODBC is just fine, but some of the best Internet tools I've seen let me talk natively to the backend database of my choice. Also remember that databases are no longer just the realm of MIS. We have marketing people updating information that gets piped directly online. Ad agencies are part of the publishing chain. So are everyday consumers who may be adding to a company's online store of information. As a result, the administrative interfaces better not be command-line driven. If it's a commerce engine, don't hard-code work flow and process into the product. I've yet to see two companies sell their products the same way, but most commerce/merchant servers think they can create off-the-shelf solutions. Untrue in the extreme. "Off-the-shelf solution" usually means "too much time spent customizing and working around bad design."

One of the most annoying trends is to ignore the collaborative nature of Web site building. If more than one person can't use the tool on a site, it's not of much use. The days of one person building a corporate site are over. In fact, they were over a year ago. Also, since most places already have a site of some kind, please don't force me to re-input my site's code through your tool. At the very least, let me import it automatically.

Make sure the product runs on the right platforms. If it's a server product, don't make me run NT just to use your product. If it's a client, it better have a Mac version (most creative departments I've seen runs Macs). The Mac issue is even more critical if your product is for the end user. The clients I deal with care about this issue. There's a QuicktimeVR-like product on the market that we can't use because it'll lock out 25 percent of this particular Web site's visitors.

Make sure the product fits into your workflow. The Web's introduction to large companies is hard enough without having to train people to do their job differently.

Whatever you do, don't build a product that does everything. We really don't need that. I've seen dozens of tools that try to cover all the bases: HTML editor and dynamic site serving, with CGI generator and asset management system. The more areas you try to cover, the more your product suffers. Most of all—the best advice I can provide—listen. Build what your customers ask for, not what you think they want. Bring your clients into your planning sessions. Map out the workflow with them. I've watched it work, and seen it fail when you leave them out.

Rule Number Four: Don't Panic.
Most likely you'll miss the mark the first time (especially if you fail to solicit input from the marketing and creative groups during the requirements phase). If you are nimble, you can fix things quickly and get the next release out in three or four months. Can't release software that fast? Learn fast folks, your competitors can.

Don't panic about the medium either. By not panicking and keeping an eye on long-term investing, you'll avoid the mistake many folks made in the 1940s. A radio station owner I worked for in Illinois had a shot at a TV license back then, but she turned it down because there wasn't any money in TV; after all, no one owned one then. No one remembers the losses TV station owners incurred back then, and when the Internet turns the corner to mass media, no one will remember the Web "losses" of the mid-90s either.

Richard Dean can be reached at richdean@clark.net

---

Copyright 1997 © Sybase, Inc. All Rights Reserved.