Page 1 of 3   next page »

From a commercial open source company's point-of-view, open source is ideally the ultimate in "grass roots" marketing where people learn about the project by word-of-mouth and where they volunteer their time and effort, resulting in a vibrant community that benefits the company in many ways.

While this ideal may apply to some open source projects, for the vast majority of open source companies, it is not a case of "build it and they will come." Instead, most open source companies need to understand who comprises their community so they can formulate a viable business model. In particular, they need to understand that communities consist of heterogeneous types of people, with their own interests, motivation, needs, and ability.

Open source companies need to identify the groups in their community, decide which ones to focus on, and choose the best way to work with them. This is de rigueur for determining how best to monetize the interest in their software, ideally without disrupting the community spirit that differentiates their software from proprietary offerings. This is where "old school" marketing can help. Obviously these techniques need to be adapted for open source, requiring the blending of traditional marketing techniques and community relations.

Treating a community as an undifferentiated blob can result in failure to generate sustaining revenue as well as alienation of a community. As a community is perhaps the most distinctive component of an open source company, losing its community is tantamount to failure. If the community is not properly nurtured, an open source company's potential won't be realized.

This article suggests a general approach for segmenting a community to aid in formulating a business model and marketing plan to reach a project's potential. Moreover, it uses the example of an open source company, Funambol, Inc., to provide concrete examples of how open source community programs can be effectively used.

Does Open Source Need Marketing & Segmentation?
For many (Cherkoff 2005), open source is meant to be a disruptive business model where the conventional laws of business and marketing don't apply, e.g., it's ideally meant to be the ultimate in "viral marketing" (Rushkoff 1994, Helm 2000, Skrob 2005) or "guerrilla marketing" (Levinson 1984), where only interested parties raise their hands and volunteers give their time and/or money to the cause, and where you don't need to spend much time or money on traditional marketing.

While that may be the case for some open source projects, especially those that don't need to make money, and while it may work for start-ups and early stage projects (Rosenberg 2005), established commercial open source companies need to determine how they can monetize the interest in their software, ideally without disturbing the open source philosophy that differentiates their software from commercial offerings.

This is where time-honored marketing principles and best practices apply (among the others, Baker 2000, Doyle 1998, Kotler 1999 and 2002). This means understanding your community and what makes users tick, determining their needs and interests, and formulating products and services that your organization can offer and that the community is likely to want and buy.

The most successful open source companies such as MySQL, Red Hat, SugarCRM, and Zimbra all do this to a certain extent, albeit in a stylistically different way than pure commercial companies. The sooner an open source company comes to grip with the reality that it needs to practice standard marketing techniques such as segmentation, target marketing, and direct marketing, the better it will be.

Obviously, these techniques need to be adapted and adjusted to take into account the appropriate ways to communicate and interact with open source community members, so we're talking about the blending of two disciplines, marketing and community relations.

A widespread stereotype about open source is that communities mainly consist of hardcode hackers who only contribute code. In reality, communities are comprised of many different types of people, each of whom has their own interests, motivation, needs, and ability to contribute. As a project experiences success and crosses the "chasm" between the early and mainstream stages (Moore 2002), the community is composed not only of "techies" (developers and IT people), the typical early adopters, but increasingly by non-technical people, i.e., end users looking for products and solutions. Moreover, communities are increasingly composed not of individuals but professionals who work in companies such as system integrators, OEMs, and service providers and are involved to further the business objectives of their corporations. The fact that communities have corporate members is relevant because business-to-business marketing requires a different approach than marketing to individuals/consumers.

Communities are really heterogeneous groups that need to be addressed differently: various people and segments require communication through different channels with different messages. An open source company should identify the different groups in the community, decide which ones it's going to address, and choose the best way to leverage the target groups. Community segmentation and marketing are essential for designing effective business models and actions.

The risk of unfocused actions is to de-energize the community, which could result in losing the unique competitive advantage distinguishing open source companies. The community is an open source company's primary distinctive asset and if not properly leveraged and nurtured, the "disruptive" potential (Christensen 1997) can stagnate (Onetti and Capobianco 2005). Just being open source is not itself a guarantee of success; there are plenty of companies without active communities. Anyone can go to Sourceforge and see projects that languish because of minimal community involvement; furthermore, projects are like stars, they can shine for periods of time but if they don't continuously renew their energy, they can burn out. Consider the case of a highly popular and publicized open source project such as Evolution, a one-time alternative to Outlook that has languished.

Recognizing that your community consists of several kinds of people and grouping them according to their needs and profiles has broad implications for the business model in terms of how to best work within each of these groups. This article aims to describe, through case study research, a generic approach for how commercial open source companies can segment their communities to aid in formulating a business model and marketing plan to reach their potential. It's for anyone who works in an open source company or project who's trying to determine a viable business model.

The article is structured in two parts. The first part proposes how an open source company can segment its community. In the second part we will present the experience of Funambol, a provider of open source consumer push e-mail and personal information management (PIM) synchronization. We'll describe how it segmented its community and created nurturing and leveraging open source programs.

Segmenting an Open Source Community
An open source community often consists of an assortment of people, developers, end users, IT people, ISVs, SIs, ODMs, and partners. Segmentation helps by identifying these distinct groups. One question we'll try to answer is how it's possible to segment an open source community.

Note that for this article, we broadly define the use of the term "community" to include everyone who participates in an open source community. This includes people who download project software and documentation, read or post messages to community mailing lists, visit the project Web site looking for project information, and participate in project events such as webinars. This definition of community may be different than the more narrowly defined group of hardcore developer enthusiasts but in our experience, the broader use of the term is more relevant to an open source community. Note that it doesn't necessarily include commercial customers and partners though there's likely to be some natural overlap.

A community can be segmented according to several characteristics. We could refer to demographic variables such as age and technical skills (dividing community members into "techies" and "unskilled" individuals), or psychographic characteristics such as the purpose of community involvement (separating people who operate for business purposes from individuals moved by hobby/volunteering aims). Moreover, for corporate community members, we could segment using typical B2B criteria, e.g., dividing them by company size (discerning SMEs from large corporate), location, industry and/or business type (e.g., distinguishing among service providers, system integrators, independent software vendors, device manufacturers, and so on).

Page 1 of 3   next page »