We believe in a world in which every single human being can freely share in the sum of all knowledge.
The job of the Wikimedia Foundation (WMF) is to be just that: a foundation. We support the Wikimedia communities. We keep Wikipedia online. We build for the future.
We keep the Wikimedia websites fast, secure, and available. We make free knowledge accessible, bring new knowledge online, lower barriers to access, and make it easier for everyone to share what they know. We defend the Wikimedia communities when they are threatened, legally or otherwise. We stand against censorship, and support open licenses, open access, and open source. We support our contributors, users, and developers. We offer resources to people and organizations that believe in the Wikimedia vision.
We will only achieve our vision when we share it. By supporting communities, building technology, and funding great ideas, the Wikimedia Foundation is a foundation for people everywhere to change their world.
2014 was a year of transition and opportunity for the Wikimedia Foundation. We considered our mission, its role in the world, what we want to accomplish together, and how we will get there.
Throughout the year, WMF supported the Wikimedia projects and communities through its core functions of engineering, product development, funding, community resources, legal protection, and more. This support is critical, constant, and the needs continue to grow. Support for these core functions represent the majority of the WMF's work and resource allocation.
In May 2014, Lila Tretikov joined the Foundation as Executive Director, succeeding Sue Gardner. In the fall of 2014, under Lila's leadership, we initiated a process designed to deliver a new strategy for the organization. This process understands not as a set of goals or objectives, but rather as a direction that will guide the decisions for the organization. This report is one of the outcomes of that process.
The State of the Wikimedia Foundation (WMF) 2015 provides a snapshot view of the Foundation's major initiatives and challenges during the 2014 calendar year. The report offers a baseline assessment of key efforts made by internal WMF departments, with an emphasis on data-based results, project impact, challenges, and how our work ties back to our mission.
The content of this report is based on information shared with us from various departments and teams throughout the Foundation. We asked each team to report on both achievements and challenges from 2014 in order to take an honest look at where the Wikimedia Foundation and movement are currently -- and where we are going.
The report is divided into eight sections, each of which covers a specific department. Each section covers a specific department within the WMF and includes a team overview (About), a summary of the team's work during the year (2014: Overview), a summary of key initiatives (Key Efforts), a summary of challenges identified throughout the year (Considerations), and a look into 2015 work (2015: The year ahead).
The baselines in this report offer a transparent and candid reflection on our accomplishments, opportunities, and challenges. They are intended as a resource for the Wikimedia community and ongoing WMF strategy discussions. We hope they will offer insight into the work of the Foundation, the successes and challenges faced by individual teams, and progress toward our collective Wikimedia vision.
As the WMF initiated the 2015 strategy process, we reviewed our accomplishments and challenges from the prior year. We identified a need for better alignment, improved processes, and greater clarity.
In response the WMF developed a set of actions for 2015, designed to focus the organization around immediate and long-term mission-driven opportunities. These objectives are intended to reinforce the core of the Wikimedia Foundation and prepare us for the future. They reflect our commitment and calibration around excellence, community, and innovation.
We are calling them our 2015Call to Action.
Improve technology & execution
Define our commitments – deliver on-time and on-budget.
Make our decisions based on data.
Improve our process for community input and allocate dedicated technical resources to community requests.
Update legacy architectures and deliver mobile-ready infrastructure and services to support structured data, user security, and a simplified user experience.
Focus on knowledge & community
Integrate across community engagement functions to improve communication and results.
Create a central, multilingual hub for community support.
Have a working plan to support emerging users and communities.
Improve our measures of community health and content quality, and fund effective community and content initiatives.
Support innovation & new knowledge
Integrate, consolidate, and pause or stop stalled initiatives.
Create spaces for future community-led innovations and new knowledge creation.
Facilitate and support new models and structures for knowledge curation.
Strengthen partnerships with organizations that use or contribute free content, or are aligned with the WMF in the free-knowledge movement.
In the prelude to the strategy discussions, we identified knowledge as the primary lens through which we see the world and our work.
As an organization, we are committed to knowledge in the following ways:
Expanding knowledge. We believe in free knowledge for every single human being. We are committed to growing the volume of free knowledge we can supply. We are committed to expanding access to free knowledge for people around the world. To that end, we are continuously exploring and expanding our understanding of what constitutes knowledge.
Empowering people. We see knowledge as a foundation for human potential, freedom, and opportunity. A country or community that has better access to knowledge is more likely to be healthy, more likely to be free, more likely to be happy. Unfortunately, access to knowledge is not equal. In some places, people cannot afford to access it, in others, it is censored or constrained. That's why we seek to put knowledge in the hands of people around the world.
Enabling creation, curation, and sharing. Experiencing the full value of knowledge requires participation, not just consumption. We build tools that help people create, curate, and share knowledge. These tools should be open, fast, secure, reliable, and a delight to use. We are constantly looking for innovative ways to make participation with knowledge accessible to the world.
These objectives, and our focus on knowledge, will guide the WMF through 2015, along with continued consultation with the community.
The following are brief summaries of the work of the individual WMF departments in 2014. This includes major accomplishments, considerations, and intent for 2015. For more detailed examinations of our work, we encourage readers to please review the full report.
The WMF Fundraising team raised $58.5 million in the 2014-2015 fiscal year to date, which put them six months ahead of plan by the end of 2014. In December 2014 alone, they raised $32.3 million. At the same time, changes in global traffic and user behavior presented challenges, as banner-driven revenue is tied to traffic on Wikimedia sites. As traffic continued to move to mobile, and on-site readership decreased in countries where the WMF raises the majority of its funds (especially within Europe and North America), the team experimented with new banner formats and messaging. According to Fundraising projections, revenue from the year-end English fundraiser would have been 43% under target had the team run the same campaign as the previous year.
The WMF Engineering teams worked to ensure that Wikimedia technology was scalable, reliable, secure, and accessible to the world. Key priorities included increased support for mobile web and apps development, partnering with leading engineering organizations on upstream projects, improving developer tooling and processes, supporting the developer community, and ensuring the security and reliability of the projects and underlying infrastructure. Key initiatives included deployment of HipHop Virtual Machine on MediaWiki servers (making editing on Wikipedia twice as fast), improved search functionality through deployment of Elasticsearch, the beginning of the librarization of MediaWiki, improved ease of translation with Content Translation, and steps toward a Service-Oriented Architecture.
The Product teams focused on building and testing engaging mobile experiences that support new types of contributions on a wide array of devices, delivering delightful reader experience on tablets, delivering new editing capabilities, developing new user experience standards, and promoting editor engagement and user acquisition. Key initiatives included the release of a rewritten native Wikipedia mobile app, testing of new engagement features, a new mobile Web interface, a more accessible system for wiki talk pages (Flow), foundational improvements to VisualEditor, the introduction of Media Viewer, and testing of a new contribution stream (internally known as "WikiGrok").
Considerations for Engineering and Product centered on the increase of mobile readership without a commensurate increase in mobile editing, change management and change aversion on desktop (for example, in the roll-out of Media Viewer), third-party reader interfaces that develop more quickly than Wikipedia's reader experience can change, and a lack of structured data on project knowledge. These teams also cited the obscurity of MediaWiki as a deterrent for further volunteer participation.
The Wikipedia Zero team within Product and Engineering worked to expand the Wikipedia Zero program, launching 32 new operator partners in 2014. At the end of 2014, the team was working with a total of 56 operators in 48 countries, covering an estimated 450 million mobile subscribers. Of the new partnerships, 80% emerged from carrier requests or existing group partnerships. The team coordinated with partners on local marketing to drive awareness and usage, added product features to ensure full user experience, and developed guidelines to ensure partnerships align with movement values. However, challenges arose around a lack of evidence that the program was reaching the target audience. Other considerations included the fact that Wikipedia was not originally designed for Global South users and that potential users have a lack of awareness about Wikipedia.
The Grantmaking team's grants and project implementations reached more than 28,000 people across 61 countries and 33 language Wikimedia projects. The focus was on supporting community projects that expand knowledge and represent diversity in geography, language, and gender. Priorities included expansion of content in multiple languages through grants and other resources, empowering Wikimedia movement leaders to support community growth, addressing key diversity gaps in Wikimedia content, and supporting research and experiments that increase the quality of content and our understanding of the movement. The team awarded 236 grants totaling $6,299,733 to 61 countries, including 124 grants to Global South applicants totaling $1,190,872. More than 400 program implementations were reported by grantees in 2014. Considerations included the need for caution, humility, and data required in working with emerging communities, a need for better communication with Wikimedia communities, and Wikipedia's gender gap.
In 2014, the Communications team focused on growing core team capabilities to support strategic initiatives, driving special branding projects to tell Wikimedia's story more effectively, promoting key events through public relations, responding promptly and accurately to incoming requests from press and other influencers, managing key communications channels for the movement including the official Wikipedia and Wikimedia social properties and the Wikimedia blog. Key efforts included the WMF's first-ever year-in-review video, the first microsite version of the Wikimedia Annual Report, a relaunch of the Wikimedia blog reflecting its role as a global channel, and major media launches around the Wikipedia iOS app, Transparency Report, Right to Be Forgotten, and more. Considerations included potential brand "identity" confusion, media perception that Wikipedia may be in decline, and the absence of a unified content strategy.
In 2014, the Talent and Culture team focused on employee and contractor relations, organizational and employee development, making key hires at all levels, supporting board and executive development work, improving recruiting processes, and the ongoing work of compensation and benefits, payroll, and compliance. The team hired 58 new staff and contractors and processed 25 departures throughout the year. These new hires included an Executive Director, Chief Communications Officer, and Vice President of Engineering. Other key efforts in 2014 included launching a Learning and Organizational Development program, implementing a new applicant tracking system and other efficiencies to drop the days to hire average from 87 to 43, and adding team capacity with a Benefits Coordinator, Recruiter, and Recruiting Coordinator. Considerations included retention and recruiting challenges as a technology-driven non-profit organization in one of the most competitive job markets in the world, and opportunities to increase staff diversity.
In 2014, the Finance and Administration teams focused on continued transparent reporting for donors and the Wikimedia community, new investment measures to extent the value of funds, and improvements to the WMF headquarters to support evolving staff needs. This team also delivered ongoing support for the organization through financial management and planning, operating internal information technology (IT), and managing and allocating WMF facilities and equipment. Key efforts included creating a long-term reserve fund, improving the 6th floor of the WMF headquarters, scoping and beginning renovations for an expansion to the 5th floor of our office building, increasing building security to prevent theft and protect employees and contractors, and managing Wikimania 2014. A balance sheet as of June 30, 2014 is included in this section. Considerations included the need to protect contractors and volunteers as we continue to work internationally, and the challenges associated with limited resources to support all WMF staff.
Each section in this report includes a "2015: The year ahead" component, which gives a brief glimpse into the teams' plans for the year ahead. In keeping with the 2015 Call to Action, these plans fall into at least one of three categories: improve technology and execution, focus on knowledge and community, and support innovation and new knowledge.
Fundraising will continue to diversify how they raise funds, explore new payment options, and improve outreach messaging, while working with other departments to support the WMF and Wikimedia income model for the year.
Engineering and Product will focus on delivering reliable, quality software and products that meet the needs of our readers and editors, supporting our ability to enable creation, curation, and sharing of knowledge through new contribution streams on mobile, and strengthening community input into product priorities and develop standardized update and feedback channels.
The Wikipedia Zero team will work to define the program's target more specifically and its strategic focus and align internally about strategic focus, resources, and commitment, with the goal of ensuring that we reach our target audience.
Grantmaking, now integrated into the larger Community Engagement team, will focus on understanding the needs, concerns and priorities of volunteers, proactively sourcing and funding new projects that address key Wikimedia challenges (for example, gender diversity), promoting content development and community health through education programs, strategically allocating and supporting Annual Plan grants, improving content by enabling "super-users" via the Wikipedia Library, and continuing to gain and process insights about emerging communities to inform organizational activities. The team will also have a continued focus on impact and non-monetary support, in addition to funding.
The Legal team will continue to defend Wikimedia and the community by using all tools available, from litigation to policies. The team will also build on its public policy strategy to continue combating the global uptick in censorship and address other crucial issues. The Community Advocacy team, now part of Community Engagement, will continue community support, including by keeping the more than 15 regular workflows fast and efficient. They will also launch the 2015 strategy consultation to help the WMF maintain a strategic direction consistent with the Wikimedia vision.
The Communications team will focus on driving major campaigns, messaging, and initiatives that support our mission and story; developing strategy for brand management, development, and communication; managing public relations and issues management; developing product marketing and communications; and improving transparency and coordination around activities of the WMF.
The Talent and Culture team will focus on hiring great people, training and developing staff, and designing an internal culture shift. Specific initiatives include building a training roadmap for key skills and implement to grow second-level leadership, and building out the recruiting team to hire "the right fit."
The Finance and Administration teams will continue to explore ways to extend the value of our funds and cooperate on more effective reporting for decision-making while continuing to support the organization's core functions. This group will also drive the 5th floor expansion of the WMF headquarters.
The WMF Product & Engineering teams power, shape, and deliver the technology behind the Wikimedia projects, making knowledge accessible to people around the world in formats that promote contribution, curation, and learning.
In 2014, these teams approached this mission in the following ways:
Ensured that the underlying platform behind Wikimedia – MediaWiki – remains scalable, reliable, and secure for users, and accessible in order to encourage third-party contributions.
Developed new products and features that make knowledge more accessible on mobile devices, where readers increasingly consume content.
Improved site performance and search, making wikis more reliable and effective for users.
Continued development of a user-friendly editing experience for all devices, including easier methods for editing tables and citations.
Began to experiment with micro-contributions beyond editing.
Developed and began to implement a consistent, platform-appropriate experience and interface for users.
The WMF Engineering teams ensure that Wikimedia technology is scalable, reliable, secure, and accessible to the world in order to support the more than half a billion readers who rely on Wikipedia and its sister projects on a monthly basis.
Priorities for 2014 included:
Strengthening support for mobile web and apps development.
Built capabilities for mobile web and apps development.
Shifted towards a service-oriented architecture (SOA), building out a Services Engineering team, and more cleanly separating concerns in MediaWiki's aging PHP codebase (more detail on SOA approach below).
Rewrote our mobile apps in native code and hiring dedicated teams for Android and iOS development.
Deployed HipHop Virtual Machine (HHVM) on MediaWiki servers.
HHVM is an open-source virtual machine developed by Facebook for executing programs written in PHP and Hack.
The deployment decreased loading time for any page viewed while a user is logged in and for any page saved that a user has edited whether logged in or not.
HHVM also reduced median page-saving time for editors from about 7.5 seconds to 2.5 seconds and reduced mean page-saving time from about 6 to 3 seconds.
The team leveraged existing upstream open source performance improvement projects and resources, including Facebook developer time, for positive partnership and performance outcomes.
HHVM was fully enabled on all web servers by December 2014.
Despite improvements in page loading and saving times, research has yielded no evidence that HHVM has increased the number of edits saved or reduced the time spent completing the same number of edits for new editors.
More research should be done in 2015 to determine what impact, if any, HHVM has had on editor engagement.
Improved search functionality through deployment of Elasticsearch.
Beginning in Q4 of FY 2012-2013 and continuing throughout 2013 and 2014, the MediaWiki Core team worked to replace MediaWiki's legacy, home-grown search system with a new system based on Elasticsearch.
Previous to this initiative, very little work had been done by the WMF on the Wikimedia search infrastructure, which was outdated and no longer serving the needs of our projects.
The new system provides scalable search, has near real-time indexing, and supports multitenancy.
After an extensive testing phase as a beta feature, Elasticsearch became the primary search on all wikis in Q2 of FY 2014-15, serving 872 million full text searches and 3.1 billion find-as-you-type searches in the month of December.
Initiated librarization of MediaWiki to improve platform flexibility and performance.
Beginning in Q2 FY14 the MediaWiki Core team began breaking MediaWiki into reusable libraries which can be integrated into any PHP application.
Goals for the librarization project include making MediaWiki more accessible to developers and easier introduction of code from third parties.
The first two libraries released were CSSJanus and CDB.
Librarization has become a standard operating procedure in 2015, as new libraries are introduced and old functionality continually gets rolled into new libraries.
Improved ease of translation of Wikipedia articles with Content Translation.
The Language Engineering team released the third version of the Content Translation tool in Q2 FY14 to facilitate the translation of Wikipedia articles from one language to another.
Content Translation allows editors to quickly create a new language version of an article from an already existing article (with presumably validated references and citations).
This tool was enabled in mid-January 2014 as a beta feature in Catalan, Danish, Esperanto, Indonesian, Malay, Norwegian (Bokmål), Portuguese, and Spanish Wikipedia. The team aims to roll the feature out to another 10-15 Wikipedias by end of Q3 FY14-15.
This new feature has the potential to increase the number of articles across various language versions of Wikipedias without compromising on the standard that is expected of the content.
The Engineering Community team works to engage and promote contributions from volunteer developers throughout the Wikimedia movement. These active technical volunteers help develop and support many areas of MediaWiki development and maintenance, including bots, gadgets, user testing, or institutional memory.
Engaged more than 400 active individual volunteer contributors.
Two professional engineers volunteered full-time on quality assurance, multimedia, and HHVM.
Co-organized the Wikimedia Hackathon EU, the Wikimania Hackathon, and 25 Tech Talks.
Participated in outreach programs such as Google Summer of Code, FOSS Outreach Program for Women, Facebook Open Academy, and Google Code-in.
Helped bootstrapping the MediaWiki Stakeholders Group.
Launched a Wikimedia instance of Phabricator.
The Engineering Community team coordinated the deployment of Phabricator, a third-party open-source software development platform, in Q2 FY14.
Phabricator now supports Wikimedia project management, bug reporting, and new feature design, replacing Bugzilla, the Wikimedia bug management tool used for the previous decade.
Phabricator also replaced RT, Mingle, and Trello, tools used by various Wikimedia Foundation teams.
The Wikimedia instance of Phabricator includes more than 84,000 current tasks – 18,000 of which are "open" as of February 2015 – and more than 1,500 users.
This is the largest known instance of Phabricator for task management.
As the engineering organization has grown, there has been an increased need for clear and streamlined development processes. To that end, the Team Practices Group was created in July 2014 to facilitate improvements around predictability, collaboration, and software delivery in WMF engineering.
The team takes a consultative and case-by-case approach to help all engineering teams at the Foundation with fulfilling their commitments and delivering quality software, for example by helping a team adopt Scrum methodology. In the six months that the Team Practices group has been in existence, key efforts have included:
Provided dedicated resources to four teams (Analytics Research, Analytics Engineering, Mobile Web, and Mobile Apps).
Conducted a team development workshop with one team (MediaWiki core).
Provided ad hoc coaching on process improvements to seven people since Q1 FY14.
Currently, the Team Practices Group is developing a team health check survey to identify how different groups are performing against qualitative focus areas.
The Product team drives the conceptual development and implementation of software features from idea to release. The team works across the organization and movement to guide the product development lifecycle.
In 2014, the Product team's focus was the development of foundations: for consistent use of data to drive engineering decisions, the development of a user experience vocabulary for our products, the maturation of community engagement processes throughout a product's lifecycle, and a consistent approach to development in a multi-device world.
Priorities for 2014 included:
Creating engaging mobile experiences that help users access, create, and share knowledge, including new types of contribution appropriate to a wider array of devices.
Delivering a delightful reader experience to tablet users, replacing the traditional desktop experience.
Delivering foundational new capabilities for VisualEditor, including source-specific citation dialogs, table editing, a mobile-appropriate UI, media dialogs, character insertion, language-specific improvements, and many more.
Developing a new user experience for discussions and piloting it in the French Wikipedia's help forum for new users.
Developing standards for iconography, controls, styles, and other aspects of the user experience and creating the first prototypes of a living style guide.
Promoting editor engagement and facilitated user acquisition.
Throughout the year, the Product teams worked to create a more compelling, accessible experience for both readers and editors on mobile. Much of this work was foundational and prepared us to do more transformational work in 2015.
Released new, entirely rewritten native Wikipedia mobile apps.
Released Android in Q4 FY13 and iOS in early Q1 FY14.
These native apps improved speed and responsiveness over mobile web experience, and introduced a number of new features including: in-app editing, reading history of recent pages, geolocation of nearby articles, offline reading, short title descriptors, collapsible infoboxes, article sharing, and improved readability.
The apps also integrated support for Wikipedia Zero on participating carriers and networks.
Tested new engagement features.
A number of new engagement features were developed for release in the Android app in early Q3 FY14, and for a future release on iOS. These included:
Prominent, contextually-relevant images at the top of articles, with parallax-scrolling, for improved reader engagement and a beautiful reading experience;
"Read More" feature including links to as many as three related articles at the end of each article, for continued reader exploration;
Improved search functionality, including more defined and higher contrast search bar and a list of recently searched topics; and
Improved image viewer that enables users to view a larger version of any image without distractions (image appears unobscured if tapped on) and swipe left or right to view the previous or next image.
Integration with Wikidata search results.
Solid download numbers, app store exposure, and positive reviews.
The new native mobile app was featured in the Apple Store on its initial release.
While neither Apple nor Google provide a raw "total downloads" number, we do have the following information:
iOS: Units shipped in 2014: 3.34 million (via iTunes Connect)
4/5 star user rating
Android: Total installs by user in 2014: 5.67 million (Google Play Developer Console)
4.5/5 star user rating
Launched a new mobile Web interface.
This new interface added a beta of VisualEditor for tablet users, with the goal of making it easier to read and edit on a mid-sized screen.
Product and engineering worked throughout the year to improve the mobile Web experience, focusing on:
User experience improvements to the editing workflow, particularly for first-time editors;
Creating a lightweight mobile VisualEditor experience for tablet users, with the ability to quickly and easily format text and add links and references; and
Experimenting with new ways to highlight the living nature of Wikipedia content.
We saw a major increase in mobile editing following the launch of capability in July 2013.
Since July 2014, 20% of new active editors on all our projects have made their first edits on the mobile site.
Roughly 40,000 editors (new and existing) made at least one mobile edit a month.
The Collaboration team developed Flow.
Flow is a more accessible communication system for wiki talk pages, reducing technical barriers to user participation by making it easier to ask questions and get involved in community discussions.
Contributors can subscribe to specific conversations, in order to help users stay connected and keep track of the discussions they're involved in.
The Collaboration team's focus at the beginning of 2015 is to expand Flow's reach in various language Wikipedias. There are currently Wikipedia project pages that use Flow in French and Catalan, and that will expand in early 2015 to include Portuguese, Hebrew, Russian and others.
The team is working on decommissioning LiquidThreads, a previous discussion tool, and converting LQT threads into Flow threads. LQT pages will start to be converted to Flow in early 2015.
The Editing team improved VisualEditor further, introducing new workflows.
Substantially improved key workflows for editors, including adding citations and links, following extensive user testing and iteration.
Introduced table editing without wiki markup (thanks in part to code contributed back by a team working for the Public Library of Sciences (PLoS), the first to reuse VisualEditor outside a MediaWiki context).
Passed the five millionth VisualEditor edit on Wikipedia.
While VisualEditor remains opt-in on some large Wikipedias where editors had objected to its introduction in 2013, it is now the part of the default editing interface in more than 160 Wikipedias.
The Multimedia team introduced Media Viewer, a multimedia browser designed to:
Improve the viewing experience for readers and casual editors.
Make it easier to preview and browse images, right where you are.
Display images in larger sizes, with basic information – and links to more details.
Help enlarge, share, download, or embed images on other web pages.
Lessons Learned from Media Viewer:
The Media Viewer project ran from July 2013 to November 2014 and was more challenging than expected. While the product received positive or neutral feedback on some projects, it was met with negative reactions from many on the English and German Wikipedias, as well as on Wikimedia Commons. This required the team to work longer than planned, to improve features based on user feedback.
Some key findings from WMF research about this product
Media Viewer serves a lot more images than before (17M intentional views/day).
Most users keep Media Viewer enabled (99.5% enabled).
Media Viewer key features were found easy to use.
Media Viewer is more useful for readers than active editors.
What worked well
Detailed activity and performance metrics.
Design research – before and after implementing a feature.
Working with community champions in different projects.
Agile development process and tools.
Unit tests to improve the code.
What didn't work well
Many community discussions did not effectively inform product development.
Surveys were not representative, because they were optional.
We lacked the tools to get productive feedback from different user groups.
Juggling feature and platform development at the same time was hard.
Scope creep: the workload kept growing beyond available resources.
No clear success metric: we couldn't tell if we had met our goal.
Pilots and Experimentation
Throughout 2014, the Product team led pilots and experimentation to identify new ways to engage people in the Wikimedia projects. The most significant of these is known internally as WikiGrok. A new contribution stream intended to enable a wider array of users, including readers to contribute to Wikipedia.
Suggests answers to simple questions about a Wikipedia page, asks readers to verify those suggestions.
Introduces simple and easy "tagging" of critical information in Wikipedia for the creation of structured datasets in Wikidata.
Inspired by "Wikidata – The Game" created by Magnus Manske, a longtime Wikimedian and volunteer software developer.
Initial testing showed:
1,200 responses were submitted in the 1-week test period.
90% of users who began the workflow (tapped on "okay!" on the first screen asking if they want to help Wikipedia) successfully submitted a response.
Based on a sample, the quality of the responses is high: over 80% were correct.
The Analytics team supports and empowers the community and the WMF with critical data and builds systems for measurement and sharing to enable people to create and share knowledge.
Launched Vital Signs, a dashboard displaying key editor metrics for all 882 projects.
Made improvements to Wikimetrics system tracking editor activity.
Deployed Hadoop cluster (longterm project underway since 2011), offering support for fundraising efforts and generation of pageview data available to the community and other research projects.
Added mobile views to our public pageview data sets.
Integrated Python and R utilities for processing Wikipedia data.
Mobile use is increasing, but without a commensurate increase in mobile editing.
Our data indicates that fewer than 10% of edits started on mobile apps are finished, the size of mobile edits is smaller, and long-term retention of mobile editors is very low.
We need to create new ways for growing mobile community to participate in our projects that recognize the opportunities and constraints of mobile devices.
Change management and change aversion on desktop.
The Media Viewer experience on desktop demonstrated that even a seemingly simple change like a lightbox image viewer can quickly balloon in complexity as users discover edge cases related to Wikimedia's massive taxonomy of templates and user- or project-specific customizations.
Combined with a high investment of the core volunteer community in the current experience, this means that changes that impact desktop users are expensive and slow.
Our immediate response to this has been to treat mobile web and apps as proving grounds for new user experiences, but we need to find better ways to bring key improvements to all platforms and devices, and staff efforts that target this user segment in full recognition of the complexities involved.
Third-party reader interfaces develop more quickly than Wikipedia's reader experience can change.
In 2014, WikiWand received $600,000 in investment to provide a "beautiful new interface" to Wikipedia content.
Recent updates to our mobile web experience (which scales responsively to screen size) feature many of the kinds of usability improvements redesigns tend to target.
Design changes can be difficult and slow to bring to the desktop (see above) when they clash with edge cases or with editor-focused functionality.
We risk being left behind when it comes to offering the best experience reading Wikipedia.
An overwhelming percentage of project knowledge is not available as structured data. Our content styling is inconsistent with many edge cases.
Wikipedia content lacks consistency in styling, and is often not available in structured formats.
Content is in long-form text that is styled for large screens and difficult recombine in new ways for new platforms.
Lack of clear structure makes reader and editor engagement features more difficult to build.
It isn't always obvious that Wikimedia projects are living and thriving ecosystems of interaction.
MediaWiki doesn't surface the evidence of users and edits.
The evidence of user contributions and interactions are often obscured by complexity and limited styling.
We suspect many people don't realize they can participate in the projects.
The Grantmaking team works to increase the quantity, quality, diversity, and reach of free knowledge by supporting people and organizations aligned with the Wikimedia Foundation mission. They make grants to individuals, groups, and organizations working on building community and growing content on Wikimedia projects and sites as well as related open knowledge projects. They are committed to supporting under-resourced and emerging regions, languages, and communities, particularly in the Global South.
In 2014, the Grantmaking team worked to support community projects that expand knowledge and embody diversity in terms of geography, language, and gender. As always, this required unique, tailored approaches in order to empower different communities.
Money is only one of many resources that matters to different communities, and is often not the most important one. In 2014, Grantmaking team worked across different Wikimedia Foundation departments to identify and provide community resources and improve coordination at many levels.
Priorities for 2014 included:
Facilitating the expansion of quality and quantity of content across multiple language Wikipedias and sister projects through grants and other resources for individuals, groups, and organizations.
Empowering Wikimedia movement leaders by working with online and offline contributors, from readers to super-users, to support the growth of communities and address their specific needs.
Addressing some of the key gaps in diversity of Wikimedia content and communities in partnership with Wikimedia communities around the world. Currently, 80% of the world's population is represented by 20% of total Wikimedia contributors, and only 1 in 10 contributors worldwide is projected to be female.
Supporting research and experiments that increase the quality of content and our understanding of the movement around the world, based on an understanding that knowledge is differently understood, constructed, used, and expanded in different contexts.
Supported a wide range of community ideas and global initiatives through strategic grantmaking.
Awarded 236 grants totalling $6,299,733 to individuals and groups in 61 countries, including 124 in Global South totalling $1,190,872.
Supported work in more than 30 language Wikipedias and sister projects.
More than doubled support to Global South in number of grants and total funds awarded year over year: more than half of all grants awarded to the Global South.
Significantly increased monetary and non-monetary support to individual contributors across the world: 75% of the total number of grants were awarded to individual contributors.
Grantees used funds awarded by Grantmaking to conduct community outreach and capacity building; identify, engage, and mentor new editors; partner with institutions; and increase, improve, and curate content, among other things. In 2014, more than 400 program implementations were reported by grantees including at least:
109 Introductory Editing Workshops
103 Photo Contests and Events (62 Wiki Loves Monuments)
60 GLAM Content Release Partnerships
41 Wikipedia Education Programs
22 Writing Contests
Collaborated with advisory groups and community members to shift program design, execution, and evaluation towards measurable impact and strategic goals.
Worked with the Funds Dissemination Committee (FDC) and larger Wikimedia organizations to improve correlation between funding and conversion of offline interest into online engagement, including 16% year-over-year reduction in overall FDC allocations.
Worked with Individual Engagement Grants (IEG) and Project and Event Grants (GAC) committees to identify and engage community members in more than 50 proposals for experimental tools and projects that improve community health and contribution.
Piloted a collection of baseline project and engagement impact data.
Developed metrics for measurement of content acquisition, expansion, and community participation, including first round of impact analyses of grants and programs, data on global education programs.
Developed and executed first-ever Global South reader and contributor survey, which involved 47,000 respondents.
Introduced Global Metrics as a consistent (necessary but not sufficient) movement reporting mechanism for WMF-supported grants and programs.
Initiated research into current and future forms of Wikimedia movement organizations, including organizational effectiveness tools and research.
Matured support for community from ‘grants only' to multiple forms of resources.
Supported experimentation and growth of communities and content through money, program guidance, mentorship, tools, research and impact analysis; for example, through learning patterns.
Winding down what doesn't work
Wound down the "catalyst" program in Brazil, in recognition of the need for more gradual building of community capacity, and the complex history of WMF's involvement in Brazil.
A Closer Look: GLAM Partnerships 2014
WMF GLAM grantees created partnerships with and supported Wikipedians-in-Residence at GLAM institutions, leveraging Wikimedia brand recognition to generate significant content contributions and donations.
In 2014, Wikimedia Sverige (Sweden) supported the upload of 20,000 images to Commons from the Royal Armoury, Skokloster Castle, and the Hallwyl Museum.
Images donated through a Wikipedian-in-Residence program at Cancer Research UK have been used more than 200 times in seven language Wikipedias. Of the 400 images released, more than 43% are now in use on Wikimedia sites and elsewhere – a very high percentage.
Funded projects to increase diversity of knowledge in terms of geography, language, and background.
Funded an experimental project that explored the use of oral citations for documenting and curating indigenous and traditional knowledge from underrepresented regions (much of the Global South). Much of this knowledge is not included on most major wikis due to current Reliability policies.
Funded projects to increase female-related content by building toolkits, networks, and events to grow hundreds (and lay groundwork for thousands) of new biographies on women scientists and artists.
Funded a medical translation community organizing initiative that created improved systems allowing translated versions of the Ebola Virus Disease article to be integrated into over 60 languages within a few weeks in 2014.
The New York Times reported that the article received 17 million pageviews as of October 2014, on par with the World Health Organization's Ebola fact sheet and the Centers for Disease Control and Prevention's Ebola portal.
Significantly increased online content through work with education programs worldwide, as exemplified by the Arabic Wikipedia Education Program.
3.2% of new Arabic Wikipedia content over the last three years has been added by education program students in Egypt, Jordan, and Algeria, including:
49 of the 670 peer-reviewed Good and Featured articles
7.3% of total peer-reviewed content
More than 80% of Arabic Education Program students are female.
Increased focus on women-specific topics by some student groups, supported by a collaboration between the education program and the Egyptian user group.
Emerging communities require caution, creativity, humility, and data, especially in the Global South.
We need additional research with and into our current and future communities in order to formulate good hypotheses and make good decisions about engagement strategies.
We need more focused pilots (rather than to try and work at scale from the start) in order to increase awareness of our projects to readers and potential contributors, and support to existing contributors.
Our work may challenge some definitions of what constitutes ‘core' ‘encyclopaedic' work on Wikipedia.
Progress toward the sum of all human knowledge requires questioning what constitutes knowledge and thinking critically about Wikipedia tenets such as notability and verifiability.
New and underrepresented knowledge often comes in unique formats, e.g.: oral or written, offline or online, by multiple voices or just a few. Future work in this area could be met with resistance from the community or other groups, but is at the core of fulfilling the Wikimedia vision.
We need better communications from the WMF to community on key decisions.
Opaque or inconsistent communications on key issues may frustrate or alienate volunteers.
The Grantmaking team often feels these effects more directly and for a longer time because of its ongoing, interdependent work with community leaders.
Wikipedia's gender gap is persistent, and few projects emerge organically at scale.
Despite some grassroots efforts, much work remains to be done to increase gender diversity in terms of editorship and content on Wikipedia and its sister projects.
New focus, experiments, and project leaders are needed to reduce systemic bias.
We need a more unified approach to community engagement.
During 2014, it became clear that the WMF needed a new structure for engagement with Wikimedia editors and volunteers.
This consideration and the Call to Action served as the basis for the creation of the Community Engagement department in February 2015.
In the year ahead, the organization formerly known as Grantmaking, now part of the newly created Community Engagement department, will focus on the following priorities:
Understand the needs, concerns and priorities of volunteers in order to support the community and develop recommendations for future programs.
As of March 2015, the Community Engagement team has launched in internal working group to begin this process
The Community Advocacy team (see Legal and Community Advocacy section), also now part of Community Engagement, will play a key role in this effort with the former Grantmaking team.
Proactively source and fund new projects that address key challenges of the Wikimedia community, for example the Gender Gap.
One example of this is the Inspire Campaign, led by the Community Resources group, which aims to source and support 20 new gender-focused projects, targeting 200+ currently active women editors, 1000+ newly registered women, and 1,500+ women involved.
This project also aims to increase non-male classified gendered content and ensure that at least 50% of project leaders supported are women (increase from <30%).
Promote content development and community health around the world through education programs.
For example, by growing Arabic and Spanish Wikipedia through education programs that promote quality content. This will require solidifying community support in Jordan, targeting mentoring in Egypt, Mexico, and Argentina, and investigating expansion into Saudi Arabia.
Support and develop education programs worldwide by developing program materials, mentoring program leaders, and tracking and reporting results.
Strategically allocate and support Annual Plan grants, with a shift in focus from money and process to impact and non-monetary support.
Increase impact by providing more non-monetary support while decreasing size of grants and offer capacity building for Wikimedia organizations.
Non-monetary support includes peer and WMF-led knowledge sharing and mentorship, for example through trainings and providing tools.
Conduct community consultation to solicit input from FDC, grantees, and others on the process and goals of Annual Plan grants.
Build a more impact-focused culture by socializing benchmarking research.
Improve content by enabling "super-users" via the Wikipedia Library.
Add more journal partners, including non-English language partners, to increase quality content in library.
Roll-out global satellites to support program expansion and train volunteers to run these branches.
Continue to gain and process insights about emerging communities to inform organizational activities.
The team will continue to surface results from the Global South survey and other research efforts to the organization and conduct additional surveys to track progress and gain further insight.
The team will also conduct a series of qualitative interviews in key countries to complement quantitative data from these surveys.
Wikipedia Zero aims to reduce the affordability barrier to access to knowledge for people in the Global South. The Wikipedia Zero team partners with wireless carriers to make Wikipedia available to people – even if they can't afford the cost of data charges. This is a process known as zero-rating. Today, an estimated 450 million people in 48 countries can access Wikipedia through Wikipedia Zero.
In 2014, the team worked to expand the Wikipedia Zero program by partnering with more carriers, expanding existing partnerships, leveraging partner promotion to drive adoption, and making changes that improve the overall integrity of the program. At the same time, there remain major challenges with Wikipedia Zero, especially since it may not be reaching the target audience at scale.
Total mobile usage nearly doubled in both Nepal and the Philippines following partner advertisements in those countries.
Operators piloted direct community support for local language content development.
In Bangladesh, Grameenphone is collaborating closely with WMBD, and sponsoring a major training program in seven regions.
Smart Philippines is providing space and wifi for WMPH editathons; Smart has also donated prizes for the WMPH heritage project. WMPH has conducted teacher trainings as part of Smart's Tap & Learn program.
In Thailand, Telenor's dtac, ran a contest to get employees contributing to Wikipedia as part of their volunteer commitment, under the supervision of Thai Wikimedians.
Added product features to ensure full user experience.
Began migration of partners to IP-based zero-rating, enabling free editing and browsing of all Wikimedia projects.
Developed guidelines to ensure partnerships align with movement values.
2014 saw continued growth of the Wikipedia Zero program in terms of visibility and partner launches. This continued growth and maturation has allowed the Zero team to begin to assess its efficacy and impact across a broader data set.
The biggest concern for Wikipedia Zero is that we do not yet see evidence that it is reaching the target audience – the world's poorest people who cannot afford mobile data charges – at scale. We still do not see organic growth in usage. And our own data on pageviews by language version show roughly 90% usage in English throughout South Asia, indicating the program is actually reaching more privileged segments of society. The Wikipedia Zero team has identified a number of possible explanations for this:
The target audience is not yet actively using the Internet.
Carriers care about growth markets. They are actively trying to engage new users in the mobile internet, leading with discounted bundles of well-known services like Facebook and whatsapp. Our mission is relevant, but our main tactic is less so.
This raises the question of whether Wikipedia Zero is premature, and if there are other ways we could help bring people online.
Our product was not designed to serve Global South users.
In some Wikipedia Zero countries, locally-relevant language content is not well-developed.
The Foundation has not dedicated resources to optimizing the product for Global South users.
In countries where knowing two or more languages is the norm, the product does not support easy toggling between languages.
Not all cultures are familiar with or accustomed to using "encyclopedic" knowledge – this requires investigation of product-market fit. For example, target users may prefer shorter, more accessible bits of information rather than long-form articles.
Potential users have a lack of awareness about the product, and we are not marketing to them.
Many potential users may not know about the availability of Wikipedia Zero – and many are not familiar with Wikipedia or encyclopedias in general.
Because of constraints in our trademark policy and historical resistance to marketing, all awareness efforts are borne by partner operators.
Operators tend to be unwilling to heavily market a service that doesn't contribute to their ARPU goals.
Our experience indicates pageviews do not grow except when the operator advertises.
Making Wikipedia free of data charges is not driving usage in underserved segments.
Scaling pageviews is not the most critical metric.
The Wikipedia Zero team needs to understanding potential alternative metrics, and how to achieve them, with attention to awareness-building initiatives and Global South user needs.
Zero-rating is controversial among some policy and advocacy audiences.
Zero-rating and Wikipedia Zero have received negative attention by some critics who believe it is at odds with net neutrality.
Critics argue that this is because zero-rating favors incumbents with the ability to pay for preferential access to users.
We believe that since our program involves no exchange of money and is in support of our mission to make free knowledge available with the world, it is not at odds with net neutrality.
Zero-rating is falling out of favor with carriers.
Zero-rating is also becoming a bit passe among carriers, and it is typically used as a 90-day promos or first month trial, not a persistent program. Wikipedia and Internet.org are exceptions.
We have committed to operating principles that protect the integrity of the program with respect to net neutrality, for example "Wikipedia Zero cannot be sold as part of a bundle," "no exclusive rights," and "no exchange of payment." However, as we have started negotiating contract renewals under the new terms there is a risk we will lose old partners. Two are currently up for renewal and pushing back on these principles.
In 2015, we need to rethink priorities and timing of the Wikipedia Zero program, and allocate resources for the biggest impact. We also need to rethink the program strategy around key challenges, for example how we successfully enter new markets.
In the year ahead we will aim to:
Define the target.
Set specific goals and understand exactly who we aim to reach with Wikipedia Zero, for example sub-segments within the poorest segments in particular countries.
Identify key markets where we will focus resources in partnership with other groups including Communications, Community Engagement, and Product.
Align internally about strategic focus, resources, and commitment.
Clarify the role and strategic focus of Wikipedia Zero within the Foundation's Global South strategy.
Identify resources that can be devoted to Global South growth, for example through designing product with Global South users in mind.
Explore the "open library" as a policy initiative to align stakeholders to the mission and create an umbrella to facilitate our ongoing work.
The Wikimedia Foundation (WMF) raises the funds for its operating budget every year, primarily from individual reader contributions. The fundraising department is responsible for ensuring that the organization and Wikimedia movement have funds to fulfill our mission.
WMF fundraising operates on the principle that donors are Wikimedia contributors too – just in a different way. The team is dedicated to understanding why readers and contributors value Wikipedia, as everything from a practical service to an abstract ideal. They evolve the fundraising model to respond to these user needs and mission values, while remaining ahead of new opportunities.
In the fiscal year 2013-2014 (July 1, 2013 – June 30, 2014), the Fundraising team raised $52.6 million, exceeding the annual plan goal by 5%. So far in the 2014-2015 fiscal year, (between July 1, 2014 – January 7, 2015), the team has raised $58.5 million from 4.2 million total donations. By reaching this number, we met our goal for the fiscal year six months ahead of our revenue plan.
During the 2014-2015 fiscal year, WMF and chapter online campaigns accounted for $52 million, and foundations and major gifts have accounted for $6.5 million of funds raised. In December 2014, during the primarily English-language campaign, the team raised $32.3 million from 2.5 million donations.
Also in 2014, Fundraising saw a 178% increase in matching donations, from $282,389 in 2013 to $786,296 in 2014. We believe this is a result of improved communication about corporate matching gift donations and more staff time allocated to processing requests.
The Fundraising team continued to diversify income sources in 2014 within mobile fundraising, international campaigns, and email campaigns. Early figures show significant gains were realized in funds raised in these areas, however desktop remains the leading source by a large margin, accounting for roughly 73% of donations during our December fundraiser alone.
Due to changing trends in readership [see Considerations section, below] Fundraising intentionally exceeded our goal in the 2014-2015 fiscal year in order to prepare for future challenges. We have also explored new banner formats and messaging to increase the effectiveness of our appeals. We will continue to do so in the months and years ahead.
Ran a full campaign for the first time on mobile and tablet to address readership shifts.
Improved mobile donations per impression. Initial testing (final results pending) indicated that average mobile donation size grew from $6.86 in 2013 to $10.42.
Initial testing also indicated that donations (or attempts to donate) per impression grew from $0.000574 to $0.002956, a 415% increase.
Optimized donation flow for people on the mobile site and for people reading email on mobile devices.
Mobile full screen banner increased donations by 250% over the 2013 banner format.
Expanded email campaigns.
More than doubled the number of email A/B tests from 2013 to 2014, resulting in donations gains based on iterations in email message copy, design, deliverability, donation options and flow.
Raised 20% more in the December email campaign in 2014 than in 2013.
Raised 73% more in email donations in 2014 ($7.1) than 2013 ($4.1) partially due to launching the email campaign one month earlier.
Launched the WMF's first multilingual direct email campaigns: Japanese, Dutch, German, Italian, French.
More to come in the next 6 months: Hebrew, Swedish, Hungarian, Norwegian, Spanish, Catalan, Chinese, Portuguese, Brazilian Portuguese, and Ukrainian.
Refined international campaigns.
Generated a roughly 40% increase in revenue in each international campaign in the first half of the 2014 fiscal year compared to 2013.
Focused on only one or two countries at a time, allowing an in-depth analysis of each country and language (by dividing the year into two to four week-long sections).
In addition to the multilingual desktop banner campaigns, started to run multilingual mobile banners and send emails in donors' native languages.
Added France as an international campaign.
Explored and improved payment methods to make it easier for people to donate around the world.
Began accepting Bitcoin as a payment method and added local credit cards in France.
Integrated a new payment provider (WorldPay), that will eventually give us a holistic redundancy system for most of our international and domestic cards.
Explored local methods for LATAM (Brazil, Mexico, Chile, Peru, Colombia, Uruguay and Argentina) and E-wallet for China (Alipay) for the 2014-2015 fiscal year.
Continued year-round continuous research, testing, and feedback.
The Fundraising team conducts a great deal of research before, during, and after every major online campaign.
Prior to the December English campaign, the fundraising team conducted several focus groups with past donors, a 1,000 person randomized reader survey, over 20 user tests, and over 100 A/B tests.
During the course of the campaign, the team received feedback through a donor survey which was completed by over 250,000 respondents. Our donor services team replied to and tagged over 24,000 emails from readers and donors. They also monitor social media and feedback from staff and volunteers.
Post-campaign, the team is in the process of conducting five follow-up focus groups and two separate 1,000 sample readers surveys – one replicating the November baseline survey and a second that digs in deeper to concerns raised by volunteers. We will share the results of all of this research once it is complete.
~45,000 email addresses signed up to receive information on how to edit Wikipedia.
Banner-driven revenue is tied to traffic. Our data indicates that all traffic is trending down in the countries that have traditionally been our core financial contributors. If this trend continues, our model will be challenged in 2015.
According to our projections, our revenue from our year-end English fundraiser would have decreased by 43% had we run the same campaign as last year.
90% of funds come from North America and Europe.
Traffic has declined in Europe and North America
In the U.S.: Between 2013 and 2014, total desktop and mobile pageviews in the U.S. dropped 8.6%
Traffic has declined in many key European countries, for example:
Belgium: 30% decline in total pageviews
Netherlands: 31% decline in pageviews
Traffic is up in many other countries, however these are not countries where we raise significant funds – and in some cases we are unable to fundraise due to local or national laws.
India: 13% increase in total pageviews
Iran: 168% increase in total pageviews
Readership is shifting from desktop to mobile, including in the United States and Europe where we raise the overwhelming majority of our funds.
The vast majority of our donations come from desktop banners.
In 2014 mobile traffic increased by more than 60%, while desktop traffic remained flat or declined.
With this shift from desktop to mobile in addition to the traffic decline, our banners are on average less productive. We need more effective messages to produce the same result.
The overall decline in traffic may be linked to readers accessing Wikipedia content off-site, for example via Google's Knowledge Graph or Facebook.
While these readers still may be accessing knowledge, they are not met with our appeals for an annual donation unless they visit the Wikipedia sites.
This offsite engagement fulfills elements of our mission in the near-term, but may detract from long-term sustainability.
This trend, along with the others outlined above, may impact the ability to meet future budget targets.
As we face these challenges and explore new approaches for our banners, we will need to continue to innovate. Our responsibility is to meet the growing financial needs of the organization, while balancing often conflicting feedback from donors, readers, and the editor community.
In November 2014, following a review of a proposed new banner format, the team incorporated design and technical elements suggested by community members and staff.
In February, the WMF commissioned a professional, randomized survey of Wikipedia readers in the target countries for the December 2014 English fundraiser – the United States, Canada, the United Kingdom, Australia, and New Zealand. The survey focused largely on understanding reader opinions of the banners and messaging used during the 2014 campaign. The results of the survey are available here.
The fundraising and executive teams are committed to engaging the organization for input and partnership as we explore these challenges in 2015 for Wikimedia to continue to be stewards of knowledge for the world. Fundraising views the following as organizational priorities to support the WMF and Wikimedia income model for 2015:
Continue to diversify how we raise our funds, explore new payment options, and improve our messaging.
Work with other WMF departments, prioritize readership and traffic growth, including in North America and Europe, through new avenues including partnerships.
Turn up the other variables in online fundraising performance, such as banner effectiveness, improved payment methods, and diversified countries for fundraising.
Grow other income channels, including mobile, email, foundations and major gifts.
Develop new channels for reaching existing and potential donors.
Hire Vice President of Partnerships to develop and oversee strategic collaborations with key partners in the free knowledge movement.
The Legal and Community Advocacy (LCA) team oversees all legal matters for the Foundation and focuses on key initiatives to help support the community, consistent with the Foundation's goals and values. This team deals with a wide range of issues and projects, including policy drafting, trademark and copyright law, international law, employment law, litigation, fundraising and grant law, domestic and international contracts, privacy law, ethics, internet law, and non-profit corporate governance.
Key focus areas include:
Carrying forward the Foundation's goals of advocating for the community in new ways, including fighting for content online and facilitating community discussions about critical WMF initiatives that affect the community.
Better supporting Wikimedia administrators and functionaries.
Becoming a center of knowledge for strategic change management.
Providing information about legal and legislative issues that impact online content.
In 2014, LCA focused on protecting the projects and users by developing proactive community policies, leveraging existing policies, taking legal action (litigation), operating key trust and safety functions, and supporting community groups and processes. We expect the need for these activities to continue and grow, with rising global censorship, internet regulation and copyright "reform," and increased interest from third parties to obtain user information.
Developed proactive community policies to further our ability to protect users and projects.
Solicited extensive community discussion – over 500,000 words in aggregate – for several major policy changes that address and confirm values of our movement and mission – including privacy, open trademarks, and fair editing.
Supported freedom for users to use neutral hosting platforms, including Wikipedia, to speak and share freely.
In July with Wikimedia Italia, successfully dismissed a defamation lawsuit initiated by a rich and powerful Italian politician – the Italian judge ruled that the Wikimedia Foundation, as a hosting provider, is not subject to stricter defamation standards than the Italian press.
Supported community's right to share knowledge freely without fear of speech-chilling lawsuits by privileged people.
In February, used the WMF Legal Fees Assistance Program program to help defend a Greek Wikipedia administrator against a defamation suit from a powerful Greek politician and businessman for well-sourced, but unflattering, statements in politician's Wikipedia article. The user won the preliminary hearing. The trial is set to begin in 2016.
Fought undisclosed paid advocacy to protect the truth and neutrality of user contributions.
Operated trust and safety functions to protect users from harm.
Community Advocacy enabled the community to quickly deal with threats by handling emergency email systems which deal with difficult topics like suicide threats.
Throughout the year, the Community Advocacy team performed trust and safety functions, processing on average one threat of harm to self or others every three days.
Community Advocacy was available 24/7 to process emergency calls.
Supported community groups and processes, directly liaising with and supporting a number of committees created to facilitate volunteer efforts.
These include the Arbitration Committee, stewards, checkusers, oversighters, and other global functionaries.
In September, Community Advocacy hosted the first OTRS administrators meeting in San Francisco to help develop better operating practices for that team.
The team solicited, evaluated, and offered candidates for the Ombudsmen, and assisted them throughout the year.
The team supported the 2014 English Wikipedia Arbitration Committee elections, including leading a redesign of the SecurePoll voting system.
Supported both internal and external sharing of information.
Community Advocacy actively created and disseminated knowledge about the Wikimedia community, including to staff, community, article subjects, and readers who do not yet contribute.
Staff communications started with teaching new staff how the community works and continued into advising staff of developments on the projects, evaluating specific content, and serving as a sounding board for community concerns.
Externally, Community Advocacy responded to many regular communications via both snail mail and email. During 2014, the team improved processes for handling these sorts of communications, consolidating two channels into a streamlined response system with an answer wizard.
Community Advocacy provided information and assistance to members of the editing community, helping them directly, putting them in touch with other staff, or referring them to appropriate community processes.
Legal processed 166 trademark requests, granting licenses for 38 requests. Many of the usage requests did not require require a license because of the new trademark policy, which allows for more open use of WMF trademarks by community members.
Conducted strategy consultations in late 2014.
Began laying the groundwork for the 2015 strategy consultations so the community can be empowered to give input into determining the best direction for our movement.
While copyright reform presents an opportunity for the movement, it also represents potential risks, as various interest groups around the world will press for SOPA-like changes and term extensions, both directly through copyright law and indirectly, through mechanisms like trade agreements.
Continue to defend Wikimedia and the community by using all the tools available, from litigation to robust privacy and DMCA policies.
Continue to support WMF from product development, to board meetings, to contracts review.
Build on our public policy strategy to continue our active engagement with the community and develop strategies to combat the global uptick in censorship and other crucial policy issues.
As of February 2015, the Community Advocacy arm of LCA has been integrated into the newly formed Community Engagement department. The Community Advocacy team will aim to:
Continue community support, including by keeping its more than 15 regular workflows fast and efficient.
Respond quickly to address obstacles and support the work of our communities and staff.
Continue to support legal and other departments in major community consultations.
Also in 2015, the Community Advocacy team will drive the 2015 strategy consultation. Community Advocacy is coordinating with other key staff to launch an agile new strategy process to help Wikimedia maintain a strategic direction consistent with the Wikimedia vision and to support the Wikimedia projects, while remaining sensitive to the changing global environment. This process began in February with a "Strategic Visioning" Community Consultation.
The Communications team leads the Foundation's efforts to openly and effectively share information – about the Wikimedia movement, the Wikimedia projects, and the Wikimedia Foundation's work itself – with a global audience including Wikimedia users, the general public, Wikimedia community members, and other stakeholders.
In 2014, the Communications department grew quickly. Beginning with the addition of the organization's first Chief Communications Officer, the team added and reorganized staff to meet current and future needs. Throughout the year, the team worked to shift its focus from reactive to proactive communications in order to ensure that the Wikimedia story was being told broadly, accurately, and with compelling messages.
Key priorities for 2014 included:
Growing core team capabilities to support more strategic and far-reaching communications activities.
Driving special branding projects to tell the Wikimedia stories in impactful, new ways.
Promoting key events with public relations to build awareness among key audiences, for example the iOS app launch and Transparency Report.
Responding promptly and accurately to incoming requests from the press and other influencers.
Managing key communications channels for the movement, including official Wikipedia and Wikimedia social properties and the Wikimedia blog.
Created first microsite version of the Annual Report to help contributors connect to our mission and explain financial realities of a top 10 web property. The new format created a more accessible experience with potential to increase our reach.
Proactively pitched organizational milestones with media, resulting in coverage to support WMF goals.
Wikipedia iOS app launch (July 31st): resulted in 42 media hits including in mainstream tech press and coincided with X % increase in new mobile account registrations.
Transparency Report (August 6th): resulted in 27 media hits from 19 of the world's top 40 publications. Most of these articles focused on the low number of takedown requests and requests for user data that were granted, as well as WMF's fight against censorship. 80% of coverage carried quotes from Wikipedia leadership, including: The New York Times, The Guardian, BBC, Daily Telegraph, The Washington Post, and more.
Right to be Forgotten (August 6th): resulted in 25 media hits from 19 of the world's top 40 publications. Of these stories came, 24 from media attending press briefing or from a WMF/Minassian embargoed pre-brief, and 100% of stories ran quotes from WMF and/or Jimmy Wales.
Year in Review Video (December 2014): within the first week of the launch, the video received a total of 130 media hits in 16 languages. The most hits came from Spanish, Russian, and English language outlets.
Communications also responded to more than 250 inquiries from the press during 2014, ensuring that the story we told about the community, projects, and the WMF was accurate and aligned with movement values.
Amplified the voice of the Wikimedia movement and communities.
The Wikimedia Blog was relaunched with a new design in July 2014, reflecting its role as a global channel for the Wikimedia movement, with posts ranging from tech updates to human-interest profiles.
Important feature changes include the ability to feature posts and surface comments, and better support for multilingual posts.
The Communications team intends to support increasing its use as a movement communications hub in 2015.
Conducted more than 50 community member interviews in 2014 (including 10 on camera and 41 audio only).
8 repurposed as blog posts (with more yet-to be published in 2015)
2 produced into videos
Produced 9 videos showcasing the power of the movement and our mission.
Blog was viewed more than 8.6 million times in 2014, or an average of 23k views per day. (Note this high number is not typical, and was due to a huge traffic bump from CentralNotice banners in May-July 2014: in contrast, average daily traffic reported by Wordpress in December 2014 was about 4k views/day, which is more typical of current blog performance.)
The top referring sites in the last quarter of 2014 were: Facebook, Wikimedia Foundation, Hacker News, Twitter and Reddit – as much as a third of blog traffic was referred from our social media channels.
The high referrals from the Wikimedia Foundation website are likely to originate from Wikipedia and our largest sites. This suggests that tighter integration with these projects could help increase traffic to the blog in the future.
2014 Social Media Performance
4,398,896 total page likes from 1,369,144 over 2014, or 221% growth (net 3,029,752 likes).
559 posts in 2014 (about 1.5 per day)
Tweeted 206 times from @wikipedia in 2014, or ~ 0.56 tweets per day
Grew followers by 10.3%, reaching more than 293K
Tweeted 495 times from @wikimedia, or ~1.36 tweets per day
Grew followers by ~40%, from 17,492 Followers (1/5/2014) to 24,482 (12/28/2014)
Traffic and engagement for content and brands is increasingly social, but we lack a unified strategy and clear goals for our social presence.
We have not thoroughly analyzed the world's use of social media and what that means for us, for example: there may be opportunities on platforms where we do not currently focus, like Instagram, WhatsApp, and more.
There is a significant opportunity to increase frequency across platforms. However, we do not yet have a single person dedicated to social media full-time. We continue to crowdsource internally our social media and aim to add strategic direction with data and oversight.
Like Wikimedia as a whole, our social platforms have an "identity crisis" – the team consolidated Wikipedia and Wikimedia into "Wikipedia Facebook" but we have separate Wikipedia and Wikimedia accounts for Twitter and Google+. We have a Wikimedia Foundation account for LinkedIn with no clear strategy.
Our reach is limited by policies of social platforms – for example, Facebook's restricting organic reach to about 1–2% of our followers. This creates an obstacle to reaching supporters, unless we pay for exposure.
Like on our projects, there is a distinct gender gap on our social media channels:
Likes: 67% men 32% women (Facebook average is 54% men 46% women).
People engaged: 73% men 27% women (wider spread than our fans).
Followers: 203,000 (71%) men 83,100 (29%) women
Followers: 18,300 (79%) men 4,900 (21%) women
Followers: 35,388 (74.28%) men 10,928 (22.94%) women 1,327 (2.79%) other
Followers: 2,424 (80.96%) men 466 (15.56%) women 104 (3.47%) other
Drive major campaigns, messaging, and initiatives that support our mission and story. This includes preparation for Wikipedia's 15th anniversary in 2016.
Conceptualize, plan, and execute forward-looking campaigns, messaging, storytelling, advocacy, and initiatives in support of WMF high-priority strategic initiatives.
Develop strategy for brand management, development, and communication.
Coordinate brand research, positioning, and management.
Initiate global market research, analysis, planning; brand positioning and partnerships; marketing and other materials; design, collateral, and products; social media; and trademarks support.
Manage public relations and issues management.
Oversee and execute on media and public relations, external positioning, crisis response, statements and releases for increasing positive coverage and distribution of priority initiatives and campaigns.
Develop product marketing and communications.
Drive positioning and awareness of WMF-developed products and technologies, in support of the Product and Engineering teams and Wikimedia strategic innovation objectives.
Support research, analysis, and identification of personas and segmentation to guide messaging, positioning, collateral, and public relations.
Improving transparency and coordination around activities of the WMF.
Produce a State of the WMF report.
Hold a Product/Communications workshop to improve processes and objectives around key communications releases.
The Talent and Culture team acts as a strategic partner with organizational leadership to impact and develop the culture, leadership, and personnel administration in service of the ongoing maturation of Wikimedia Foundation.
Focus areas include employee and contractor relations, organizational and employee development, compensation and benefits, payroll, recruiting, and compliance.
In 2014, the Talent and Culture team focused on enabling strategic advancement of the organization by making key leadership hires, supporting strategic planning, improving benefits processes and coordination, and expanding Learning and Development. At the same time, recruiting has improved over the last few months, with the average days to hire dropping from an average of 87 to 43 . There is still more work to be done and a new recruiting leader to be hired. Overall, Talent and Culture is accountable for approximately 4% of the Wikimedia Foundation's overall budget.
Key 2014 personnel updates include:
58 employees hired for new roles or to replace departures in collaboration with hiring managers and, for certain roles, search agencies
202 full-time employees (FTE) as of 12/31/14
77% are US-based in 19 states
23% are based abroad in 19 countries
Staff speaks 45 languages
38 active cases with immigration
New Hires (FTE)
Temp, Contractors, Interns
Product & Strategy
Talent & Culture
Finance & Administration
2014 New Hires, Departures, Current Employees, Temps/Contractors/Interns. All numbers as of 12/31/2014.
Retention and recruiting challenges as a technology-driven non-profit operating in one of the most competitive job markets in the world.
The team's biggest challenge is to build up internal recruiting capability in order to improve candidate experience while continuing to increase hiring quality and speed.
We are, however, on target to hire all requisition number openings by end of fiscal year.
We still need a dedicated leader to further evolve our recruiting practices.
As we continue to align the organizational structure to our evolving strategy, including reorganizing Engineering, WMF staff may experience increased strain (at least in the short-term), posing a risk to retention and recruiting efforts.
As an organization, we have significant progress to make in terms of diversity.
While the split between women and men in leadership is nearly even, women make up only 35% of the total staff and 17% of our Product/Engineering teams.
We support a global movement but more than 74% of our staff is Caucasian.
Build recruiting team to hire "the right fit" – a diverse body of capable, mission-aligned people to support the organization on executing on their mission, particularly the leadership level – at an increased velocity.
The focus will be on defining key characteristics, ongoing candidate experience research, support fellowship program, gender and diversity outreach, hiring manager training (including on "Overcoming Bias"), and recruiting.
Train and develop staff.
Build training roadmap from January 2015 to June 2016 for key skills and implement, growing second-level leadership.
Key activities will include WikiLead 2.0, conducting targeted skills-based trainings (for example, community communications), and running a coaching program.
Design the culture shift.
Use process levers to support culture with the dual purpose of supporting the WMF unique identity and mission, and increase capability to execute.
The team will work to build desired culture attributes more robustly into our leveling and organizational structure through recruiting, onboarding (Orientation 2.0), recognition, and performance assessment.
Because our operating budget comes primarily from individual donors, the WMF has a strong commitment to stewardship of contributor funds. We consider all expenditures against a framework of our strategy, values, and mission. We carry financial reserves in accordance with non-profit best practices to ensure the future and health of the Wikimedia projects.
The Finance and Administration team's core responsibilities include:
Financial management, planning, and reporting for the WMF
Operating internal information technology (IT) to enable organization
Managing and allocating WMF facilities and equipment
For 2014, in addition to Administration and Governance and Fundraising, the WMF's financial activities and performance can be broken down into three categories: how donor contributions support Wikimedia project websites, our communities, and free knowledge overall.
As of June 30, 2014
In thousands, in US dollars
Cash & cash equivalents
Current portion of contributions receivable
Prepaid expenses & other current assets
Total current assets
Property, plant, & equipment, net
Noncurrent portion of contributions receivable
Liabilities and Net Assets
Total current liabilities
Total net asset
Total liabilities and net assets
Statement of Activities
For the period from July 1, 2013 through June 30, 2014
In thousands, in US dollars
Unrestricted net assets
Support and revenue
Donations and contributions
In-kind service revenue
Other income, net
Release of restrictions on temporarily restricted net assets
Total support and revenue
Salaries and wages
Award and grants
In-kind service expenses
Other operating expenses
Travel and conferences
Special event expense, net
Increase in unrestricted net assets
Temporarily restricted net assets
Release of restrictions on temporarily restricted net assets
Increase in temporarily restricted net assets
Increase in net assets
Net assets at beginning of year
Net assets at end of year
Departmental allocation of expenses
For the period from July 1, 2013 through June 30, 2014
in thousands, in US dollars
Instituted new measures that extend the value of WMF funds.
Started new purchase card program for all department admins to incur rebates back to the WMF.
Created a long-term reserve fund to invest for long-term value through means such as stocks and bonds.
Made improvements to WMF office and began planning for 5th floor expansion to support productivity and the growing organization.
Completed 6th floor improvement project within budget, balancing sunk costs of base building repairs and upgrades with our current tenant space needs.
Built give new, small, conducive conversation hangout rooms for open-office function.
Heating, ventilation, and air conditioning (HVAC) disruption diminished in south-west corner of system to build out four new small, usable spaces, however we did not get the overall HVAC improvements anticipated with system changes.
Scoped and began renovations on 5th floor expansion (available for use beginning in late March).
Budget decisions have been made based on function and with minimal sunk costs into this leased space and building. We will limit spending to things we can take with us.
Increased building security to prevent theft and protect employees.
Due to a number of external security issues in 2014, the Finance and Administration team worked to extend hours that building has a guard present and ensured that cameras were installed on every floor.
The team also worked with building management to install key access for floors via elevators and add a guard for 2nd floor vulnerability until building management adds a door.
Produced annual international gathering Wikimania 2014, London, August 6-10, 2014.
The event was produced and organized by community volunteers with support by WMF conference coordinator, Ellie Young, and WMF teams from Communications, Engineering and other departments.
Final Attendance: 1,520
172 of the above were volunteers
62 of the above were performers
72 of the above were press
139 scholarship registrations (includes WMF-funded scholarship recipients, IEG, PEG, and 2014 bid team members)
157 WMF registrations
59 different countries represented
1,076 Wikimedia project usernames given out of 1,520 total attendees (71% Wikimedians)
Community Village: 35 conference booths for information by groups and chapters.
18 Non-Wikimedia organizations.
10,138 people attended Digital Revolutions exhibit at the Barbican Centre during the conference and most will have been through the Community Village
Finance and Administration priorities for 2015 include:
Manage office expansion to 5th floor of headquarters.
Create collaborative meeting space for larger metrics style meetings with upgraded A/V system.
Target project cost per square foot is an estimated 40% below market rate for similar expansions.
Install 24-28 dedicated sit/stand seats on half of the floor.
Increase conference room availability through two new rooms, and additional collaborative areas for non-sensitive meetings and create more communal, lounge areas for eating together and small group collaborations.
Re-purpose space on 3rd and 6th floor to accommodate organization needs like recording studio space, design work space for teams.
Base budget decisions on function and with minimal sunk costs into leased space and building.
Renovation in process; target April 2015 to make space available.
Continue to explore ways to extend the value of our funds.
Manage existing projects to make our funds last and explore further opportunities to add value.
Cooperate on more effective reporting for decision-making.
Work to bring consistency around the various reporting structures within the organization.
Develop processes to support on time, on budget Call to Action initiative.