Skip Navigation
Back to Navigation

Customer Care with a Distributed Service Model

← Back
Pro tip: Use "title:keyword" or "library:keyword" to limit to that specific field

Customer Care with a Distributed Service Model

New York Public Library, N.Y.

Health & Wellness | 2018

Innovation Synopsis

Ask NYPL helps patrons navigate the New York Public Library and provides information about its collections, programs and services. Using modern customer service technology and a distributed service model, the team transitioned from reference service to a customer care operation with decreased call wait times and fewer responses per case.

Challenge/Opportunity

In the last three years, Ask NYPL has taken on additional responsibilities, such as supporting NYPL’s own shared collections catalog and the e-book app SimplyE — services that require fast interactions with both patrons and engineering staff. This challenge prompted an investigation into solutions that make the team more efficient, improve the experience, integrate multiple channels (including the SimplyE app itself) and allow staff outside the team to participate in helping customers.


Key Elements of Innovation

The key elements to this innovation were a) switching to a Salesforce customer service system that provides efficiency, integrates email, SMS, chat, apps and social media and deflects cases with its knowledge base and b) introducing a new staffing model. The team now automatically assigns questions based on content or channel, quickly tailors reusable macros and scripts for faster responses and distributes the caseload to staff across NYPL, who log in to help when necessary.


Achieved Outcomes

The new software and the distributed staffing model allow Ask NYPL to better manage its complex queue of inquiries and improve experience, convenience and consistency. The number of staff responses necessary to solve a case decreased by 11 percent between FY16 and FY17, which enabled the team to reallocate one person to the phone queue. This in turn has reduced the call wait time by 50 percent to one minute on average.