Are all portals the same?

The fact that the portals are so indistinct makes it difficult for investors to predict which ones ultimately will come out on top.

3 min read
I know that at this stage you're probably sick of hearing about portal sites, but have you had a look at them recently? Having spent a couple of hours looking at them--comparing and contrasting their features--I came to the realization that there really is very little difference between them. In fact, I would go even further to say that they are identical.

If you look at the top left hand corner of Yahoo and Snap, you will see that the first category is "Arts and Humanities." Similarly, if you look at the top left hand corner of Netscape's NetCenter, Infoseek, and Lycos, you notice that the first category is either "Autos" or "Automobiles." One could assume that alphabetical order is the simple reason for this, but I don't think so. Rather, I think these companies all are getting similar user feedback and as a result are coming to similar placement decisions.

If they look the same, do they provide similar services? Is there any differentiation between the features offered on the various portals? The answers to these questions are a bit of a disappointment. Not only do the portals provide nearly identical services, but even when there is some innovation, it typically is only a matter of days--if not hours--before the other portals copy it.

The fact of the matter is that these portals primarily are marketing companies. They compete in a fashion similar to the way Nike and Reebok compete in the brick-and-mortar world--by essentially out-branding one another.

This analogy can be taken a step further. Just as Nike and Reebok do not actually manufacture their products, opting instead to outsource that aspect of their businesses, three of the top portals--namely Yahoo, Netcenter, and Snap--don't even have their own search technology, the primary service of a portal site.

The fact that the portals are so indistinct makes it difficult for investors to predict which ones ultimately will come out on top. As I have argued for some time now, I don't necessarily think that there will be only one winner, because there's certainly enough room for a multitude of portals. I've also argued, however, that the portals of today will not necessarily be the portals of tomorrow--that today's corporate Web sites eventually will become portals in their own rights. This is because the services offered today at portal and content destination sites will be backed into corporate Web sites tomorrow. Once enough traffic begins flowing through corporate home pages, portals almost certainly will seek out this prime real estate.

Given that technological innovation does not help distinguish between portals, I predict that, once competitors have aggregated enough critical mass to gather similar market intelligence, the battle for portal dominance ultimately will be fought on two fronts. The first front will be marketing. Portals undoubtedly will use every medium available to them to try to sink their competitors and capture mind share. Radio, television, print, and Web campaigns will be aggressively pursued.

The second front will be perhaps more effective and, over time, will serve to distinguish between portals. It's one thing to provide a service similar to those provided by your competitors, it's another thing to have similar quality of service. What will be interesting to observe going forward is whether all of these portals will maintain similar quality of service as they scale up. Will they all be able to support millions of users? I don't think so. Some will try build services internally in hopes of assuming more control over their destinies, while others will outsource services in hopes that a strong focus on core competencies will lead to superior service and product. My money says the outsourcing model will win out.

Danny Rimer is an equity analyst at Hambrecht & Quist, which covers CNET: The Computer Network, Lycos, and Yahoo. He writes regularly about the Internet in Marketwise.