Marketing FOSS with donor-paid ads

50

Author: Bruce Byfield

On December 16, 2004, the appearance of a donor-sponsored ad for Firefox in The New York Times became a landmark in the history of free and open source software. The first of its kind, the campaign received widespread attention, while the ad itself was seen as a sign that FOSS had entered the mainstream consciousness. Since then, similar campaigns have been organized both in the FOSS communities and in online social actions groups. While not suitable for every FOSS project, such campaigns remain an important marketing tool. However, they require careful consideration of their suitability, a high degree of organization, and the recognition of donors’ contributions if they are to be fully effective.

As Gregory Heller, manager of the Defective By Design campaign, points out, donor-paid ads have been used for decades in newspapers. Typically, they consist of a message addressed to the public or a government official that takes a strong stand on an issue, as well as the names of those who helped to pay for the ad. Since the arrival of the Internet, the names of contributors are sometimes omitted and are named on a Web page mentioned in the ad.

The specific idea for the Firefox campaign came from Rob Davis, a public relations expert from Minnesota. Mozilla was already experimenting with marketing that would be true to the spirit of FOSS, and, according to Rafe Ebron, former product manager for the Mozilla Foundation, the ad idea seemed a perfect chance “to see if this open source marketing could really work.” Announced on October 29, 2004, the campaign quickly received so many donations that the ad grew from one to two pages. Despite complaints from some contributors that their names were misspelled or that they were not informed of the publication date, Ebron says, “We absolutely were successful. It generated more buzz than we could imagine.” Henri Poole of CivicActions, who is also a director of the Free Software Foundation and an organizer for Defective By Design, remembers it as “a first for free software.”

Although repetitions cannot hope for the same novelty or impact, the tactic has become widely recognized as an effective tactic for promoting the goals of online communities. Volunteer Benjamin Horst recently organized a pledge-paid ad for OpenOffice.org that ran in the New York Metro on July 31 and did not include contributors’ names; he says he is currently considering a promotion of OpenOffice.org in the newspapers of major American universities. Similarly, Poole says that Defective By Design is contemplating variations of the same idea in its struggle against digital rights management. In one interesting twist on the original idea, Poole reports that CivicActions may be assisting Public Interest Pictures to produce a new movie tentatively entitled Votergate. For a donation of $50, contributors will have their name added to the movie’s credits. In one form or the other, the idea continues to thrive.

Planning the campaign

According to Poole, organizers of such campaigns should begin by assessing the likelihood of raising the cost of the ad. “The size of your community has a big impact on how effective you are,” Poole says. Citing the example of fund drives in newsletters distributed by Move On, an online grassroots organization, during the 2004 American primaries, Poole suggests that pledges generally amount to an averaged 20¢ per community member. In other words, if your project has a community of a few hundred, then a full-page ad in a major newspaper is probably an unrealistic goal unless you can find unusually generous donors or receive a special rate.

Poole also stresses the need to set concrete goals. “It’s really important to know what you want.” Talking about his work at CivicActions and Defective By Design, he says, “We look at: What are the intended outcomes of the campaign? What are the key messages we want to get across? How are we going to measure success?” For example, for the Firefox campaign, Ross states that the three goals were to celebrate Firefox’s launch, to promote Firefox, and to provide an example of FOSS marketing for other projects. Having such concrete goals allows you to plan more carefully, and evaluate your success afterwards.

For Ebron, the organization of the campaign is key. “Any marketing campaign takes a lot of work,” he says, “and the Firefox Ad campaign was no different. Behind the scenes, there were plenty of conversations with lawyers and accountants, putting the team together, [and] several review processes. Doing an ad campaign like this is a big undertaking, and people need to understand that. Having a budget of time and money and really treating the ad campaign as a project with schedules and timelines will make it successful.”

All this should be done in maybe three to six weeks. Otherwise, Poole notes, the media tend to lose interest in the campaign, and donors may feel that they are being pressured to make multiple donations.

At some point, Horst says, “the effectiveness of advertising open source software will shrink.” For now, though, he believes that ads can still be effective, particularly if the medium for them is chosen carefully. “Programs like Firefox or OpenOffice.org, which can be used by almost anyone, are the most appropriate project types for general media ads. Others might work if a more specific target is chosen.”

The message needs to be considered, too, notes Blake Ross, one of Firefox’s original developers. For an ad in the general media, he advises, “Focus on what the consumers of the ad want to know. Tell them why your software is better, not that it’s open source.”

All these practical considerations are important, but Poole suggests that campaign planners should not forget that they are partly — possibly primarily — involved in an exercise in community-building. For this reason, one of the most important considerations is the recognition of contributors.

Donating to such ad, he says, “is like philanthropy. You’re buying yourself karma.” Just as the contributors to a piece of free software code expect to be paid in credit, so philanthropists and campaign donors are more interested in being involved if they know they will receive recognition for their generosity.

“When people give money,” Poole says, “it’s within the context of a social network that they’re doing it. Even when people are anonymous donors, there’s a large group of people who know who they are.” Moreover, just as free software spreads by users introducing friends and family to it, so pledge-paid ads receive wider circulation through word of mouth among the contributors. The more contributors feel their donations are recognized, the more each of them is likely to become involved in promoting the campaign — a situation not very different from most FOSS projects in general.

Evaluating the results

The success of these campaigns should not be measured in downloads alone. Following the release of the OpenOffice.org ad, Louis Suárez-Potts, chair of OpenOffice.org’s community council and community manager, observed “a slight bump” upward in downloads, but he says, “I think more important than seeing blips in downloads is inserting the name/concept of OpenOffice.org in the minds of readers.”

In many cases, the campaign itself, and the chances for community involvement it brings, may be more important than the publication of the ad. “Contrary to popular belief,” Ross says of the Firefox campaign, “we found the ad itself to be less effective than the buzz surrounding it before and after it ran.”

Horst found a similar result in his OpenOffice.org ad. “I tossed up a first design to get conversations started,” he says, “and, wow, did it ever! Then, when our professional designer’s version was put up, there was so little to complain about that it was rather anti-climactic.” Looking back, Horst reflects, “Good thing for us that we didn’t start with that one, or we might not have gotten any attention.”

The best way to judge results, Poole suggests, is to establish the criteria for evaluation early on. For example, if the point of the ad is to attract new members to the project’s community, you could keep track of new users of the mail forum. Similarly, if the goal is general publicity, you could watch for the number of links to the project site from blogs and media reports in the weeks following the ad.

For the original Firefox campaign, Ross says, one of the goals was to “demonstrate to other open sources projects that not all marketing is the devil’s work, which seems to be a widely held belief.” Judging by the widespread interest in adapting Firefox’s tactics, the campaign was obviously as successful in this goal as it was in celebrating and promoting the project. Probably, projects need to be reasonably large before they can follow Firefox’s example, but, for those who can, it remains one of the most effective marketing techniques available.

Bruce Byfield is a course designer and instructor, and a computer journalist who writes regularly for NewsForge, Linux.com and IT Manager’s Journal.