How does one put metrics to marketing efforts for FOSS projects ?

I was doing some random reading on Marketing Efforts for FOSS Projects. This was from a completely tangential investigation into the “difference” real or perceived between traditional marketing (ie Marketing for non FOSS and even non Software instances) and FOSS Marketing (umm…I guess that term doesn’t exist). For better or for worse I decided to look across a sample set of what the Marketing teams for some projects project on to the public space.

I borrow the definition from the Dictionary of Marketing Terms and talk from the perspective of marketing being “the process of planning and executing the conception, pricing, promotion, and distribution of ideas, goods, and services to create exchanges that satisfy individual and organizational goals.” In effect, traditionally, the marketing department is expected to provide guidance and inputs to organizational units engaged in Product Development or Product Servicing. This would be based on the fact that the department is the one with the deepest insights into the existing and potential customer base and is better placed to communicate with them towards generating loyalty and/or retention.

So on this afternoon I browsed through a few of the FOSS Marketing Projects and here’s a selection of “mission statements”

The Fedora Marketing Project develops marketing strategy to promote the usage and support of Fedora worldwide. Through the development of processes and content, this project aims to support the efforts of other Fedora projects to spread Fedora and to provide a central repository of ideas and information that can be used to deliver Fedora to new audiences.

SpreadKDE is a team dedicated to promote KDE to the world. Our ambition is to increase KDE’s visibility throughout the Internet population by delivering quality content to the main technological events worldwide and by increasing the exposure of KDE inside technological magazines. By doing so, we want to increase the population of the KDE community and the contribution to the KDE project. We also want to make the industry realize how linux and KDE can be a great solution for them.

There is no “official” GNOME Marketing mission right now. It is being discussed though. A general statement might be something like “improving the experience of the GNOME community by facilitating communication and interaction”.

We’re here to promote Ubuntu’s uptake in various ways, all in essence being ways of telling people who want to hear what we have to offer, listening to what they need and trying to identify just that in Ubuntu. Remember: we market Ubuntu, we do not make it more marketable as a priority.

The Marketing Project has two aims: to promote the use of the productivity suite; and to grow the community.

A common thread across all the above statements is that the (sub-)Projects are more focused on [i] promotion of usage and [ii] increasing of visibility ie awareness. What makes this different from “traditional” marketing efforts is the lack of metrics that can perhaps answer the question “what would be the impact if the (sub-)Projects are scrapped ?”

Traditional Marketing Models where there are products or services that are attached with a value, the core question that gets people to ponder is “all things being equal, what would be the expected quantum of sales” – this in general translates into what is a Baseline Number. Baseline Sales figures are usually the result of various specific models applied upon a large set of historical data. Plainly put, baseline numbers would be the set of consumers if there were no promotion of the product. Taking this further, the impact of a promotion or marketing activities would be to result in Incremental Sales. This in turn allows one to calculate the percentage of “Lift” as Lift (in %) = Incremental Sales / Baseline Sales. Plainly, a large positive value of the Lift percentage does indicate a healthy outcome of marketing involvement. These are very simplistic analysis, but what it also allows to calculate is the Cost of Incremental Sales = Expenses incurred for Marketing / Incremental Sales. A higher positive value of Incremental Sales (which would be indicated by a positive Lift %) would also perhaps lead to a lower value for Cost of Incremental Sales. This again allows one to calculate the Profitability of a Promotion = Profits achieved with Promotion – Estimated Profits without Promotion.

A traditional avenue for marketing activities is advertising. Traditional advertising uses a number of metrics among which few of the popular ones are: Impressions (which don’t actually measure the quality of viewers), Gross Rating Points, Cost Per Thousand Impressions (CPM), Net Reach (which is used to estimate the unique touch-points for the campaign in a population), Cost per Customer Acquired, Rate of Abandonment (or, the purchases which were initiated but never completed) Advertising campaigns generally aim to ensure that within the specified time window of the campaign an individual is exposed to a pre-determined number of exposures to the brand.

The components of a Marketing (sub-)Project for a FOSS Project are Websites, Artwork, Project Tours, Articles, Promotions and Events, Advocacy, Branding, Public Relations, Market Research and Market Segmentation. The methods used to cover the components include Mailing lists, IRC channels, Newsletters, Support Groups, Forums, Wikis etc. However, the singular point of concern would be related to quantifying the long term impact of “marketing” ie. how does one actually measure how successful the marketing effort is/was. Once that can be estimated, the learnings would automatically translate into better priority for tasks and perhaps a quicker focus on what the consumer market segmentation and target would bring home. A simple objection to this entire exercise could be the fact that coming up with a Baseline Number for a FOSS project is absurd if not insanely difficult. For example, if (as a control hypothesis) there was no Marketing Team for GNOME (say), how would one come up with a Baseline Number ? The expected number of consumers for GNOME components might be limited to the small core set of developers or, it might not have since FOSS projects by their very nature spread over word of mouth and peer trust systems. Add to that the random question whether consumption patterns for FOSS projects are “skewed/lumpy” by nature and whether tailoring the access to project bits through infrastructure tweaks creates significant impact on consumption and hence the Baseline Number. Do market segments with varying expectations impact the success of a marketing effort for a FOSS project ?

The current trend of assessing marketing effort involves the implied base unit of communication. How a simple, credible, concrete, reliable story about the FOSS project is conveyed to the existing and potential consumers. This includes communication of feature sets, roadmaps, FAQs among other things. However, as the underlying technologies that make up various distributions and various projects converge, there would be a flattening of expectations from the marketing team. The real challenge then would be to come up with some metrics that allow measurement of effort and provide a realistic assessment of what is being done right vis-a-vis what is required to be reworked.