IMPROVING ENTERPRISE GIS OPERATIONS VIA STAFF USAGE ANALYSIS AND SURVEYS
Download as DOC, PDF0 likes11 views
This document outlines an analysis of the King County GIS Center's staff usage and competencies to improve enterprise GIS operations. It utilizes data from a time recording system to assess the efficiency of GIS roles and to inform budgeting, staffing plans, and professional development. Additionally, a survey of regional enterprises provides comparative insights into GIS staff allocation and operational efficiency.
IMPROVING ENTERPRISE GIS OPERATIONS VIA STAFF USAGE ANALYSIS AND SURVEYS
- 1. Greg Babinski, GISP Finance & Marketing Manager King County GIS Center 201 South Jackson Street, MS: KSC-NR-0706 Seattle, WA 98104 Voice: 206-263-3753; Fax: 206-263-3145; Email: greg.babinski@metrokc.gov IMPROVING ENTERPRISE GIS OPERATIONS VIA STAFF USAGE ANALYSIS AND SURVEYS Abstract: Effective enterprise GIS requires a team performing various distinct roles (management, programming, analysis, etc.). Gaudet, Annulis & Carr proposed a ‘Geospatial Technology Competency Model’ based on typical knowledge, skills, and abilities (KSA’s) associated with each GIS role. Each role includes typical outputs (maps, data, applications, etc.) that enterprise GIS customers expect. But what is the ideal mix of enterprise GIS- related roles and competencies? Within enterprise GIS, what KSA’s are being used and to what degree? How efficiently do GIS staff members apply KSA’s to deliver the outputs demanded? This paper outlines data from the King County GIS (KCGIS) Center’s staff activity time recording system (TRS). KCGIS TRS data will be analyzed against similar data collected from a 2004 survey of four- dozen city, county, and regional enterprise GIS operations in the Pacific Northwest. Examples will be presented to show how GIS activity statistics can be used to improve operational efficiency, develop meaningful budgets and long term staffing plans, improve hiring and professional development, and support individual career development. INTRODUCTION An effective enterprise GIS requires a team performing various distinct roles (management, programming, analysis, etc.). Gaudet, Annulis & Carr proposed a ‘Geospatial Technology Competency Model’ based on typical knowledge, skills, and abilities (KSA’s) associated with each GIS role. Each role includes typical outputs (maps, data, applications, etc.) that enterprise GIS customers expect. But what is the ideal mix of enterprise GIS-related roles and competencies? Within enterprise GIS, what KSA’s are being used and to what degree? How efficiently do GIS staff
- 2. members apply KSA’s to deliver the outputs demanded? This paper outlines data from the King County GIS (KCGIS) Center’s staff activity time recording system (TRS). KCGIS TRS data will be analyzed against similar data collected from a 2004 survey of four-dozen city, county, and regional enterprise GIS operations in the Pacific Northwest. The King County Geographic Information System (KCGIS) is a consolidated and coordinated regional geographic information resource, organized to meet the business needs of King County, local agencies, and the general public. KCGIS is comprised of both the King County GIS Center (an internal service fund, responsible for core GIS resources and enterprise GIS services for the entire County) and business specific activity in various GIS programs distributed across other County agencies. GIS in King County is used for a wide variety of business applications, including mapping, public information delivery, growth management & planning, property assessment, land development permitting, site selection, simulating environmental conditions, emergency response planning, crime analysis, transportation planning, bus & van routing, road maintenance management, public health service delivery, E911 operations, airport sound abatement, and boundary management (legislative districts, voter precincts, tax unit boundaries, etc.). WHY TRACK GIS STAFF ACTIVITY? Enterprise GIS operation is a complex activity. The Geospatial Technology Competency Model identifies 12 distinct ‘roles’ required for enterprise GIS operations. Each of these GIS ‘role’ requires mastery of a set of knowledge, skills, and abilities (KSA’s), which Gaudet, Annulis, and Carr refer to as competencies. They identify 39 separate GIS-related competencies. Each GIS ‘role’ is responsible for delivery of ‘outputs’ to customers or co-workers. 138 separate types of GIS outputs are identified. The KCGIS Center and many other enterprise GIS programs catalogue the competencies their staff members possess, either formally of informally. Most GIS managers know by experience that each of the various roles and competencies are important for successful operations. An analysis of actual GIS staff activity can provide additional valuable insight into enterprise GIS operations. Information about the actual time spent by all the staff members of an enterprise GIS operation can illuminate the actual level of effort for each GIS staff person, the level of effort by GIS ‘role’ for enterprise GIS functions, and the level of effort to deliver GIS ‘outputs.’ GIS managers can use this information to aid work
- 3. load balancing, improve hiring decisions, facilitate development of backup capability and succession planning, and support individual staff career development. A time recording and tracking system that can provide the information needed for GIS staff activity analysis can also meet other management goals. These can include management ‘due-diligence,’ customer billing, setting costs or rates for services, and meeting grant reporting and audit requirement. The information derived from a detailed time tracking system can also facilitate comparisons with other agencies (bench- marking), comparisons with previous years activity, and support ‘activity- based cost analysis.’ TIME TRACKING SYSTEM REQUIREMENTS & IMPLEMENTATION The KCGIS Center decided to enhance its staff time recording and reporting capability in 2002. The primary business driver for this decision was the need to record accurately the activity of 29 staff members performing a variety of GIS roles for three ‘GIS business lines’ and report actual labor costs for hundreds of internal and external customers. Time tracking system requirements included a web-based design that would support both activity recording and project reporting. It had to be flexible and simple to use, easy to administer, support invoicing, and be cheap to acquire, manage and use. Options considered by the KCGIS Center included adapting the Countywide time reporting system (rejected because of rigidity and cost of modifications), an old custom Oracle GIS application (Time Card), two commercial applications (Replicon Web TimeSheet and ClickTime), and a custom MS Access based Time Recording System (TRS) developed by King County’s permitting department. The TRS system was chosen because the functionality needed to record and report permit technician and building inspector activity was similar in complexity to GIS. The system is web based and had proven to be very robust and reliable. GIS did require some application customizing, but was able to share these costs with King County’s Parks Division, which adopted TRS at the same time. Some lessons KCGIS learned from implementing the new time recording system included the need to ‘normalize’ entries by staff during start up. Analysis of time records after the first few months of use made it apparent that some staff were interpreting time codes inconsistently, resulting in questionable activity time records. This problem could have been minimized by defining expected entries for each employee, then working one on one with staff to verify any surprising entries. Foot- draggers need to be identified quickly also and coached to keep entries up
- 4. to date. A campaign to look at TRS as the first and last activity each day has helped improve accuracy and timeliness. PRELIMINARY GIS STAFF ACTIVITY LESSONS LEARNED The KCGIS Center utilized initial TRS data for normal staff activity verification, project reporting, and customer billing purposes. After about six months of data was compiled, KCGIS Center management also performed a comparison of our expected level of activity by GIS ‘roles’ with actual reported levels. While actual KCGIS activity for the majority of the roles came very close to our expected levels, several categories provided surprises. Total hours spent working on some roles was higher than expected. Actual management hours were twice the expected level, while program management hours were four times the expected level. Our time devoted to system management work was about 15% higher than expected. Work performing other roles was lower than anticipated. For example, actual mapping and analysis work was about three quarters of the expected level, while both training/end-user support and GIS application development were about half what we expected. More detailed analysis of this data allowed us to understand these discrepancies and adjust expectations. For example, the higher level of management hours was traced to management, supervisory, and lead staff, who are expected to be working managers, but who worked fewer project related hours than expected. The result of this insight was an increase in the hours allocated to management overhead for our 2005 budget plus a plan to closely monitor supervisory project work charges. It appears that the higher level of hours charged to program management and the lower level charged to both training and application development are related. Our training program was undergoing a redesign and enhanced marketing effort during the initial six month data period, resulting in a large number of hours being spent on program management, rather than actual training activity. It also appears that demand for GIS help desk services has seen an actual decline in use. The lower level of application development work also appears to have been related to a major ArcGIS software migration planning effort underway during the six month period, resulting in a significant quantity of budgeted application development hours allocated to program management instead. KCGIS Center management and staff have now learned that our TRS data is a valuable tool for understanding the level of effort required for each of
- 5. the GIS roles needed to support our enterprise GIS program. Managers and supervisors use staff activity time data on a monthly basis. The administrative effort needed to produce invoices for work billed on an hourly basis has declined and the accuracy and completeness of the invoices has improved. Actual project time data also allows us to generate more accurate estimates for proposed projects as well. The KCGIS Center will begin performing semi-annual analysis of this data to look for general trends and opportunities to refine our operations and overall staffing model. EXTERNAL GIS STAFF ACTIVITY ANALYSIS As the KCGIS Center was implementing its TRS GIS activity time recording system, it became apparent that a comparison of our budgeted time allocations with other enterprise GIS programs might be useful to help analyze our operations and aid other agency GIS managers as well. A simple survey was sent to 96 enterprise GIS managers in the Pacific Northwest. One section of the survey asked for the estimated allocation of the agency’s total GIS staff time to each of 12 defined GIS activities. These activities basically correspond to the GIS roles defined by Gaudet, Annulis, and Carr. Two questions focused how enterprise GIS services were provided to the agency – by a single central service provider, or by some other means (distributed, outside service provider, other). Two questions were designed to determine the total number of individual staff members who provide enterprise GIS services within the agency and the total number of ‘full time equivalents’ (FTE’s) who provide those services. Total number of individual staff captures the number of people who spend at least 10% of their time on GIS duties. Total number of FTE’s reflects total level of effort. For instance, two people who each work on GIS 100% of their time represents a total of 2.00 FTE’s, while three people who each work 50% of their time on GIS represents 1.50 FTE’s. Of the 96 agencies that were sent surveys, 45 responded. Responses were received from agencies in Washington, Oregon, Idaho and British Columbia. Responses came from 19 cities, 14 counties, five utilities, and three tribes. In addition, various federal, state, provincial, and regional agencies also responded to the survey. Table 1 illustrates the responses from all agencies showing both total GIS staff members and GIS FTE’s. Agencies surveyed have an average of 11.76 total staff who work on GIS, with an average FTE total of 7.64, which represents about 65% of total GIS staff.
- 6. TABLE 1: ALL AGENCIES RAW AVERAGE VS. FTE-WEIGHTED AVERAGE Total staff/GIS FTE’s: 11.76 7.64 GIS Activity: Raw Average Weighted Average Application Development: 7.9% 10.8% Data Acquisition: 5.3% 5.4% Data Development: 11.3% 10.4% Data Maintenance: 17.0% 14.8% Data Management: 5.6% 5.8% Systems Management: 4.1% 4.5% Systems Analysis: 3.8% 3.7% Data Analysis: 8.5% 10.7% GIS Mapping: 17.5% 15.2% Management & Administration: 8.6% 7.5% Project Management: 5.4% 5.3% Training & User Support: 4.3% 4.8% Other: 0.8% 1.2% Table 1 also shows the average allocation by the 12 categories of GIS activities. Raw average percentages are shown, as well as weighted percentages, which reflect the total number of FTE and the percentage allocation for each agency. Raw average totals that are higher than weighted average totals reflect activities that require proportionally more effort for smaller sized GIS programs. This data suggests that smaller GIS programs spend proportionally more time on data maintenance and mapping activities, while larger programs spend increased proportions of staff resources on programming and data analysis. Table 2 illustrates the weighted average for all agencies compared to the KCGIS Center estimated totals. Key differences between KCGIS and other enterprise GIS operations include a smaller percentage of time for data development. We believe this results from the mature nature of KCGIS with little current enterprise data development activity. The higher percentage of data management reflects the complex multi-agency data maintenance environment in King County, with data from many departments and outside sources needing to be integrated into our GIS data warehouse. While percentages for GIS mapping and data analysis seem significantly different, the totals for the two activities are similar for both KCGIS and the 45 survey respondents. Analysis of our TRS data suggests that many of our end-user clients are capable of doing their own mapping work, but require help with more sophisticated data analysis projects. The survey percentages for M&A and project management confirm our need to understand and monitor the higher KCGIS Center staff allocation to these categories.
- 7. TABLE 2: ALL FTE-WEIGHTED AVERAGE VS. KCGIS ESTIMATED TOTAL Total GIS FTE’s: 7.64 29.00 GIS Activity: Weighted Average KCGIS Estimate Application Development: 10.8% 10.0% Data Acquisition: 5.4% 2.5% Data Development: 10.4% 5.0% Data Maintenance: 14.8% 15.0% Data Management: 5.8% 7.5% Systems Management: 4.5% 5.0% Systems Analysis: 3.7% 2.5% Data Analysis: 10.7% 20.0% GIS Mapping: 15.2% 7.5% Management & Administration: 7.5% 12.5% Project Management: 5.3% 7.5% Training & User Support: 4.8% 5.0% Other: 1.2% 0.0% TABLE 3: COUNTY FTE-WEIGHTED AVERAGE VS. KCGIS ESTIMATED TOTAL Total GIS FTE’s: 10.34 29.00 GIS Activity: Weighted Average KCGIS Estimate Application Development: 8.9% 10.0% Data Acquisition: 3.5% 2.5% Data Development: 11.8% 5.0% Data Maintenance: 16.3% 15.0% Data Management: 4.8% 7.5% Systems Management: 3.7% 5.0% Systems Analysis: 2.3% 2.5% Data Analysis: 11.2% 20.0% GIS Mapping: 20.1% 7.5% Management & Administration: 8.1% 12.5% Project Management: 4.3% 7.5% Training & User Support: 5.1% 5.0% Other: 0.0% 0.0% Table 3 illustrates the weighted average for just the 14 counties compared to the KCGIS Center estimated totals. Counties in the survey have an average of 10.34 FTE’s, 35% more than the average for all agencies, but still less than half of the KCGIS Center total. Key differences between KCGIS and other county GIS programs include an even smaller percentage of time for data development. The combined percentages for GIS
- 8. mapping and data analysis is higher for just counties, but still similar to the KCGIS total for the two. The survey percentages for M&A for counties is closer to the KCGIS total, but still about one third lower. CONCLUSIONS The KCGIS Center is committed to continued use of a GIS activity based time recording and reporting system. With three county departments now using TRS, there is a significant community that has demonstrated the utility of detailed time recording for better management. The activity and project-based reports have improved KCGIS Center operations and illustrated areas where additional attention is needed. The regional GIS staff analysis survey provided useful data to validate KCGIS Center statistics. The result has been a better understanding of the actual GIS roles and the competencies we use to provide GIS services and products to our customers. Project estimating has been improved and convenient reporting utilities aid incremental project status reporting. An understanding of the actual GIS competencies we utilize to perform our work aids in developing an overall understanding of the knowledge, skills, and abilities we need to maintain and helps prioritize individual staff training plans. No two agencies’ operations are alike, but a comparison of staff activity statistics can illustrate common trends of value for enterprise GIS management. Widespread tracking and analysis of GIS staff activity statistics can improve future enterprise GIS operations. REFERENCE Reference: Gaudet, Annulis, Carr: Workforce Development Model for Geospatial Technology. University of Southern Mississippi, 2001. GB: gb:kc:urisa2004abstract(Gbstaffsurvey).doc August 18,2022