Summary

  • PointCast replaced an idle PC’s screen saver with personalized news, while its client sent viewer information back to a central broadcast facility.
  • The company’s 1998 prospectus tied that client to targeted advertising and reported low early retention; about two-thirds of respondents who had left in its survey blamed client performance.

PointCast’s 1998 prospectus contains a less glamorous measure than its channel count. The company said retention during the first 90 days after registration was low. In a survey it conducted, about two-thirds of respondents who had left said performance problems with the PointCast Network Client were a specific reason. PointCast planned a new client for later in 1998 to improve performance and keep more viewers. The filing does not state the survey’s sample size, so the figure belongs to the company’s survey, not to every person who registered.

That admission sits beside the product’s most appealing idea: when nobody was using the computer, it could still deliver news. PointCast announced its free service in February 1996. Instead of leaving the screen saver to drift through patterns, the software displayed a stream of personalized headlines and information. The company’s launch release described a broadcast aimed directly at the viewer’s desktop, with channels the viewer could choose. The PC’s idle time had acquired a publisher.

It also acquired a client program. PointCast’s later filing describes a central facility that received, translated and transmitted content, a client installed on the viewer’s PC, and a separate set of tools for corporate intranets. The client did more than receive headlines: it frequently communicated with the central facility, displayed content from that facility or a caching manager, collected viewer information and sent it back. The prospectus does not identify the fields in that information or how long PointCast kept them.

The interface offered a choice. Viewers could select up to 12 of 54 channels, then let the service fetch and display the subjects they wanted while the machine was idle. The corporate version added another hand on the controls. PointCast’s free intranet tools let an organization configure clients on employee desktops, manage network traffic, control content and advertising, and broadcast internal news. A personal screen could carry a company’s own notices as well as the central service’s feed.

Advertising paid for the free download. PointCast’s prospectus describes targeted audiences, detailed advertising statistics and a client/server model intended to provide advertisers with more audience information. Its inventory included animated commercials, ticker and content sponsorships, and banner ads. The company was not selling only a convenient way to read news. It was selling a way to place messages on a screen that appeared when the computer was otherwise idle, with a client that reported information back to the broadcaster.

The distribution plan soon reached into the browser. In December 1996, Microsoft announced that PointCast would become a leading content provider for Internet Explorer 4’s planned Active Desktop. The announcement said PointCast Network 97 would use Microsoft’s Active Platform and that PointCast would distribute Internet Explorer as its preferred browser. It described a partnership and future releases; the announcement alone does not establish how much of the plan shipped. The proposed fit was plain: PointCast supplied the stream, while Microsoft’s desktop software offered another route to the screen.

By the first quarter of 1998, PointCast reported an average of about 1.2 million active viewers worldwide. The same prospectus described low retention in the first 90 days and the company’s own survey result about client performance. Registrations and reach could grow while the software still gave people a reason to leave. The stream needed a fast, tolerable client, reliable delivery and a reason to keep the program installed. Its advertising case depended on attention; its product case depended on the code that took up the screen.

PointCast’s experiment makes the direction of the traffic part of the story. News travelled out to idle machines; viewer information travelled back. The filing connects that return path to an advertising business, but leaves the contents of the data unnamed. It also records a product that could not keep every early registrant. The screen saver was no longer just decoration. It was the client, the channel, the ad slot and the point at which users decided whether the whole arrangement deserved another day.

Sources