Program Approach

Welcome to the Tableau Program

Tableau Program is launched to empower UCSF faculty, staff and students to use data to gain insights and improve performance for patient care, research, education, finance, HR and other practice areas.

Data in Action at UCSF

Data and analytics (D&A) are becoming increasingly important in understanding and improving performance across UCSF, such as:

  • Health Lean Teams: improving the quality, value and safety of patient care
  • Educators: understanding student competency development to improve student performance
  • Researchers: analyzing clinical data to gain insights into disease pathways and ways to inhibit, stop and reverse disease progression
  • Administrative staff: understanding and improving clinical, financial and operational processes

Today, analytics are largely developed by IT staff, both within and outside of central IT. IT staff work with departmental teams to understand and address business problems and questions. The IT staff then identify the “right” data, and work with the departmental team to create reports and dashboards that support the business conversation and help define the business problem.

With Tableau Program, departmental teams take the lead in identifying the data, reports and dashboards that address their business needs.

Rollout Approach

The Tableau Analytics capabilities are rolled out by department, groups of departments or to research teams. 

Each group being onboarded will go through the following process:

  1. Identify a group Lead Developer. This is a very important role as it will serve as the main point of contact between the group and the Tableau Program Team. The Lead Developer will coordinate the group’s onboarding activities and manage the ongoing administration of the group’s data, reports, dashboards and users.
  2. Identify participating users, and define their roles in the program: e.g., Developers, Interactors, and Viewers and train them in the UCSF online, self-paced curriculum.
  3. Once training is complete, Users will be granted access to the Tableau environment.
  4. Identify key data needed to perform analyses, and verify that exists within the Enterprise Tableau environment. If not, develop a plan to onboard any new data.
  5. Identify and onboard any pre-existing Tableau reports and dashboards.
  6. Train the group Lead Developer in how to manage, maintain and administer the group’s Tableau project folders and users.

Key Data & Analytics Insights

In 2017 we conducted approximately 20 Lean catch ball sessions with clinical, administrative and IT leaders to get insight and feedback on the key analytics that support health-system improvement. 

Key findings from these sessions include: 

1. Communication: Communication and understanding gaps exist between departmental and IT staff which hamper the creation and usefulness of D&A that get created. This gap is driven by low literacy between these key players:

  • Departmental staff aren’t proficient in using D&A to solve business problems
  • IT staff don’t have expertise or purview into how business operations work and how data is captured

2. Planning: A lack of fundamental D&A planning (how D&A will be used, by whom) results in underused assets (e.g., dashboards, reports).

3. Transparency: Limited data transparency hinders documentation, definitions, ownership and stewardship, and quality of the data.

4. Skill mastery: The current tools (e.g., QlikView, R) to perform analytics require sophisticated technical skills, which keeps departmental staff from creating their own analytics.

5. Resourcing: Demand for D&A resources exceeds the supply, which generates a severe backlog of analytic requests.

6. Siloed knowledge: Knowledge, resources, and data aren’t shared among D&A professionals in different units, which creates silos of redundant analytic efforts.

Balance Access to Data with Confidentiality & Privacy

The goal of the Tableau analytics program is to provide secure, high-quality data to users along with a tool, Tableau, that makes it easier for users to create their own analyses. This must be achieved while meeting UCSF’s data confidentiality and privacy responsibilities.

Careful consideration has been given to how best to balance these competing objectives. The approach that has been adopted is based on the following tenets:

1. All Tableau objects (e.g., data sources and dashboards) will be classified according to UCSF Policy 650-16 Addendum F, UCSF Data Classification Standard so that they can be protected according to UCSF’s data privacy and security policies.

2. Developers are allowed to connect to their Data Sources, Develop Dashboards and Publish them to Tableau Server (Development and QA only).

3. All development of Tableau reports and dashboards will occur in the Tableau Development environment. User Acceptance Testing should be performed on Tableau QA environment.

4. Tableau dashboards will not be shared with end users in the Tableau Development environment. Sharing will occur either in the Quality Assurance, Innovation or Enterprise Tableau environment (see table, below).

5. The central Tableau Admin Team will be responsible for promoting all Tableau objects from the Tableau QA environment Production environment. 

6. Granting access to Tableau objects is the responsibility of the central Tableau Admin Team.

Tableau Environment Summary

 

Tableau Environment

 

Description

Accessible by Tableau Role

Developer

Interactor

Viewer

Development

Environment where data is originally imported into Tableau and where reports and dashboards are created.

Yes

No

No

Quality Assurance (QA)

Environment where end users test and validate Tableau reports and dashboards before they are migrated to either the Innovation or Enterprise environments.

Yes

Yes

Yes

QA User Dashboards

Environment where departments can share reports and dashboard with departmental users. This environment does not require reports and dashboards be fully-compliant with UCSF Data Governance standards. Metrics and analysis in this environment are not approved for internal and regulatory reporting.

Yes

Yes

No

Production

Environment for sharing reports and dashboards that have been certified as fully compliant with UCSF Enterprise publishing and Data Governance principles. Metrics and analyses in this environment are certified as being approved for internal and regulatory reporting.

No

No

Yes