13 Association of Research Libraries Research Library Issues 291 — 2017 Lessons Learned The 2016 projects employed different work models and had different goals. The primary objectives of the author rights outreach project were relatively finite, with tangible deliverables expected following an intensive one-day working meeting. Although minimal planning and coordination were required in advance, and outreach by way of a public presentation followed the meeting, the bulk of the work was contained within one working day. We will utilize this agile development process again when appropriate, to quickly and efficiently produce collaborative work products from a diverse representation of the library. We will also promote our use of this process more heavily when recruiting future project volunteers, as several participants indicated that their involvement was due largely to the anticipated high impact from a relatively low time commitment. The ORCID@Cornell project was considerably more complex, requiring communication with and training for library staff, public- facing resources, and the outreach campaign, as well as technical work in collaboration with Cornell IT. The team accomplished everything it set out to, but assessing some components of the project was a challenge. In particular, we do not know how effective the outreach campaign was. We do know there were far more Cornell-associated ORCID iDs at the end of the project than when we began, but we do not know if that was a direct result of coordinated outreach, or independent uptake by faculty. A specific assessment plan could have helped us measure the efficacy of our outreach efforts, but we chose to balance the effort required of researchers to obtain an ORCID iD with the likely effort involved in responding to a follow-up survey about their use of ORCID. Similarly, explicit support and buy-in from library directors and other administrators could have helped us track outreach activities more closely. Integration of ORCID iDs into campus systems remains a challenge as researchers are under no obligation to make public their Cornell affiliation, or to authorize Cornell as a trusted party.