Will the bubble burst? The Freeserve flotation

25 June 1999

[Note from the author (Feb 2000): Freeserve are currently doing their best to prove my concerns wrong by thinking as radically as I suggest they need to. Their current incursions are into broadband. I leave my article intact for historical interest!]

Edd Dumbill edd@usefulinc.com

The free UK ISP Freeserve is shortly to be floated at somewhere between £1.3 and £2.5 billion. Much has already been said about the inflated value of Internet stocks, the premise being that the high valuations are in expectation of e-commerce opportunities to come in the future. As Paul Farrelly of The Observer puts it [1], the most widely used valuation method "is pure hope: the revenues each subscriber is expected to generate if Internet retailing takes off."

Is this hope a sure one for Freeserve?

How does Freeserve make money?

Freeserve has three main revenue streams: a share of the telephone income, advertising revenue and a share of revenue generated from e-commerce sites accessed through the Freeserve site. The main difficulty for Freeserve is that none of these streams are secure.

Freeserve receives a share of the income from the terminating telecoms operator for calls to Freeserve's POPs, Energis. The state of telecoms regulation in the UK is far from stable at the moment, and it is by no means certain what the future will be for this income stream. Already one provider, screaming.net, offers free call charges.

A default install of the Freeserve software sets the Internet Explorer home page to be Freeserve's web site. This has led to Freeserve's home page receiving a large amount of traffic. The advertising income depends on users continuing to use and stay within the Freeserve site when they come online.

There are some flaws in this model, the most obvious being that it is relatively easy to alter the default home page; the counter-argument to this being that, as Freeserve is a portal to a wide variety of content, the user won't want to.

As I have written before, users on the Internet tend to use it to accomplish a specific task, and their activities are more goal-oriented online than in other media. If the Freeserve home page doesn't satisfy their goal then they won't stay there.

I would suggest strongly that the vast majority of page impressions on Freeserve occur on their front page and that little traffic filters through to the rest of the site. This limits the opportunities for advertising revenue. It is clear that Freeserve are trading a lot on being the starting point for web use. Freeserve's chief executive, John Pluthero, stated recently "We will be the UK's home on the Internet", adding "Internet advertising is experiencing dramatic growth" [2].

Freeserve does indeed have a lot of potential for advertisers, providing advertisements can be used to supplement the user-experience, adding functionality that users want from the Freeserve site. Banners alone won't work.

This third component, and perhaps the most significant one for Freeserve, trades on their site being an effective portal for users. That is to say, depending on the users to always come through Freeserve in order to visit other sites or make a purchase.

Other web sites either pay up-front to feature in Freeserve as a co-branded subsite (such as Popcorn, which paid a significant sum for tenure in Freeserve) or agree to give Freeserve a share of each transaction originating from their site (such as BOL, an online bookseller).

The absolute crux of the matter here is, do portals work?

As an "introducer", then yes, portals do work. To expose a user to their first experience of a particular site, portals can be quite effective. This is because a site which has access through a portal often syndicates some amount of content or functionality to that portal, therefore offering more utility than a basic advert; or it is because the portal has some credibility as a starting-point for the user.

My question is: once a user has found site they like, will they always navigate through Freeserve's home page to get there? You can't depend on the fact that users won't figure out how to use the "bookmarks" feature on their browser, or figure out how to change their home page. You may liken a portal to a shopping mall, where shops are conveniently gathered to make it easy to visit each one. This convenience is a myth on the Internet -- once you're aware where a site is, it's easier to go straight to it.

Recent research from Jupiter Communications suggests that portals are performing poorly in sending traffic through to sites which have paid for placement.

The user population is incredibly diverse. There is content available on the web to cater to each part of the diversity. So why will users choose to go through a one-size-fits-all, cramped, portal every time, when it may only help them occasionally?

The portal strategy for Freeserve seems misconceived. The site that does all things just-about-okay will lose out to the site that does what the user is interested in well.

Gambling with the future

The Internet stocks game is a gamble on future e-commerce opportunities. Freeserve have neatly filled a gap to become the UK's No. 1 ISP in a short time by taking the cost out of connecting, and using retail outlets to reach the population, neatly usurping AOL in the process.

However, the face of telecommunications and the Internet is changing rapidly in the UK. As regulatory action opens up local call provision to more operators, and more and more homes get cable access, Freeserve cannot be confident of its current advantage. Localtel have already used free-call Internet access as a lure for switching to using them as a telecoms provider.

Freeserve has aided the commoditization of Internet access, and may well become the victim of it. Internet access is likely to become a normal part of any telephone or TV service offering, and available to all: this would leave Freeserve with no claim on its most valuable asset -- users.

In order to stay ahead of the ever-changing industry, Freeserve needs to be thinking as radically now as it did when it started out.

I don't think I'll be buying any shares for now.

If you would like to respond to this article, then please email me at edd@usefulinc.com. I will post responses here with their author's permission.

1: The Observer, 4 July 1999

2: Sunday Business, 27 June 1999

Updates

ZDNet asks Are portal sites losing their magic?. It appears that the growth in traffic to portal sites is levelling off: the power of user-choice is showing itself again! (7 July 1999)