Hacking SkillFinder: How The World Bank’s Talent Network Can Reduce Poverty

C0-authored by Maria Paz Hermosilla

In the mid-eighteenth century, Denis Diderot and Jean-Baptiste D’Alembert began publishing their twenty-eight volume encyclopedia with contributions from Enlightenment greats such as Rousseau, Montesquieu and Voltaire. The Encyclopédie endeavored to categorize and widely disseminate secular knowledge systematically in order to shift people’s thinking toward a secular Enlightenment worldview. It is perhaps fitting, therefore, that Pascal Saura, a former scholar of Diderot and professor of philosophy has been among those leading the charge at the World Bank to catalog and make searchable the know how of its 27,000 employees, consultants, and alumni.

Begun in 2014, SkillFinder is an online expert and talent network. Originally intended to help employees find internal project peer reviewers across the organization, like many tools however, SkillFinder has been “hacked.” That is to say, in its first year of use, World Bankers are relying on the software more to find those they know to help with team building than to find those they don’t know to conduct peer reviews. Managers and employees want to understand better the skills and skills gaps of those who sit across from them in order to inform their work plans.

In public institutions, especially, it is all too common for individual know-how to be masked by vague titles like manager and director. But, as Saura says, “titles don’t speak, expertise does.” Using software to give organizations greater access to insights about what their employees know — an accelerating practice in the private sector where expert networking has taken off — has the potential to improve effectiveness and efficiency for public institutions as well.

What remains to be seen, however, is whether the Bank — one of the first major public institutions to deploy a company-wide expert network — can also leverage such data-driven tools to further its work of ending extreme poverty and promoting shared prosperity.

How SkillFinder Works

The World Bank’s SkillFinder tracks expertise and experience across three dimensions: technical expertise, geography and clients, and business processes.

Technical expertise includes items such as primary and secondary specialization, languages spoken, and publications.

Geography and clients shows where someone has worked, with whom, and toward what end.

Business processes includes products, project lifecycle, activities and roles.

Skill Finder

World Bank Skill Finder Specialization, Skills and Languages

In an effort to yield data about people that is both comprehensive and trustworthy, SkillFinder’s personal expertise profiles are culled from three sources. The tool scrapes data from institutional records, such as human resources data about projects and credentials. This provides a source of authenticated, “official” data.  It also asks people to fill out free-form narratives and apply short tags that endeavor to give additional color and flavor to descriptions of what people know. Finally, it asks colleagues to endorse their peers and bosses and adopts recommender and badging techniques common to such platforms in the private sector. These third-party endorsements, à la LinkedIn, corroborate self-assessments.

SkillFinder

World Bank SkillFinder Third Party Endorsements

One of the challenges of building such an expert network is knowing how to organize information using a vernacular familiar to the institutional culture. The World Bank already possessed a shared vocabulary comprised of a taxonomy of development topics at the core of the Bank’s mission on one hand, and a taxonomy of business functions common throughout most organizations on the other. It was able to build on these dictionaries to ensure that expertise would be more searchable using terms common within the Bank.

Using SkillFinder to Address the World Bank’s Core Mission

Now that SkillFinder has been successfully deployed and tested, the next step will be to use this “technology of expertise” to address the World Bank’s core mission: fighting poverty.

It is not enough to assume that having a skills directory will lead to the alleviation of the challenges of the global poor. By itself, no tool accomplishes much. But if the ability to target and pinpoint expertise and to match people to problems based on their skills is leveraged well, SkillFinder could be a game changer for the Bank.

Rather than change business practices and assume this is so, however, SkillFinder presents an ideal opportunity to test if and how having more searchable data about what people know makes a difference.

Does the ability to pinpoint what people know make it possible to assign people more effectively to projects and find more comprehensive and innovative solutions?

At its core, an expertise locator should help unpack tacit knowledge and therefore boost the “random access memory” of the organization; it provides the organization with the power to tap quickly into its deep and wide reserve of human capital. At the same time, a comprehensive mapping of specializations and skills makes structural needs and gaps more visible and subject to agile analysis. It has the potential to shape what the organization should urgently learn, and how it will learn it.

The hypothesis that having tools to match people to challenges befitting their skills and experience will enable leaders to reach outside their usual networks and find new people with fresh and innovative approaches to hard problems. Whether matching enables the Bank to increase the responsiveness, coverage and innovation of the organization can be empirically tested. We can imagine three such experiments.

First, given a development challenge, imagine using SkillFinder to see how many teams the leadership of the World Bank could assemble, at least on paper, to tackle the challenge? How diverse would those teams be? Would the tool surface people who are more junior? With members who are not known to each other?  In parallel to using SkillFinder to form new teams, an A/B test might be run to assign the same challenge to an existing team. Such a test might not be perfectly scientific but it might yield meaningful insight into the value of technologies of expertise for problem solving.

Second, an expert network like SkillFinder makes it possible to test whether targeting people on the basis of their experience makes more of a difference than staffing people on the basis of their credentials? This hypothesis that people will be more effective doing what they have done before (and not simply studied in school) can also be tested by forming teams on the basis of credentials, on the one hand, and on the basis of experience on the other. The SkillFinder deployment offers an ideal laboratory for testing these dynamics.

Finally, a similar experiment can be run to test whether assigning people to work on that which they are passionate about trump the relevance of credentials and experience altogether? Like minds can reasonably differ on this hypothesis, which would also be interesting to test.

In the case of each of these three questions, the Bank could advertise an internal prize-backed challenge relating to decreasing poverty in a particular area by a quantifiable metric, opening the challenge to all comers at the Bank. At the same time, it could pinpoint and compose teams based on different criteria and see what happens.

Ultimately, to rationalize the significant investment in SkillFinder, the World Bank needs to explore — and test empirically — how this talent bank can help the institutions solve its core problems differently. Dr. Jim Yong Kim, President of the World Bank, might consider articulating a grand challenge to his staff to tackle a particularly vexing problem in global poverty. Then SkillFinder might be brought to bear to form more diverse and skilled teams to tackle it.

Diderot, the father of the Encyclopedia, famously said “only passions, great passions, can elevate the soul to great things.” Too true. But tools like SkillFinder that offer a map of institutional expertise help find and match people to public problems they can be passionate about and solve in more innovative ways.

Cross-posted at Forbes.com

 

 

One Response to “Hacking SkillFinder: How The World Bank’s Talent Network Can Reduce Poverty”

  1. Varun October 1, 2015 at 7:50 pm #

    Hi – this is interesting that the world bank is pushing this idea in a big way. We at ARGO Labs built something similar albeit in a more local context – NYC. Using profile data from Linkedin, learnr.nyc matches tech workers in NYC concentrated mostly in Manhattan & edges of Brooklyn with after school programs to deliver structured hour long STEM programming.

    The matching idea itself is not unique – the PeaceCorps designed an app a while back (http://bit.ly/argo_peaceCorps) that matches its cadre of volunteers with its international program offerings.

    The innovation and real impact, we think, lies in how you “deliver the content” in a meaningful manner that i.e. its what you do after the match. Google’s Civic Innovation team conducted a pretty exhaustive research study and conceptualized the “Interesting Bystander” demographic that revealed some great behavioral aspects of half the country that is civically aware but not civically engaged and we operationalized it through Learnr. The Interested Bystander is our audience.

    Our hypothesis is simple. In the localized context of NYC where volunteering time & skills is often hard and nebulous – why dont we use modern digital tools to minimize the barriers to just volunteer your tech skills. NYC ranks 50/51 in volunteering time across large American cities according to the Corporation for National and Community service. its time for a digital way to mash up people and offer personalized learning experiences.

    Fred wilson, a pretty visible investor articulated it perfectly here (http://bit.ly/argo_fredvid)

    The world bank is onto something for sure but what is concerning is that initiatives like this are prime candidates for the sunk cost fallacy. A localized deployment similar to Learnr could very well facilitate testing some of these impact hypothesis through rapid prototyping and could bring an order of magnitude more volunteers into the system.

    Do check out learnr.nyc and happy to hear feedback.

    Varun & Team ARGO

Leave a Reply