9. Cost-Benefit Analysis of GIS  

9.1 Introduction  
  
Understanding and estimating the costs and benefits of a GIS project is an important part of the strategic planning and decision making process. Determining the economics of a GIS project is difficult. Cost valuation issues focus on fair determination of staff (training), equipment (hardware and software), and other costs (services, data), particularly when comparing alternative implementation scenarios. Benefit evaluation incorporates a clear division of direct, indirect, and external benefits. The user interface of the GIS application affects many of these key elements of the project, especially when we take the time to measure how long (or using how much effort) to get the job done. By better understanding the work flow of the users within their organizational environment, we can better place values on both costs and benefits.   
This section can only give an overview and expert advice. The subject "cost-benefit of GIS projects" could be described in complete books based on different theories (e.g. Born 1992).   

Cost-benefit analysis (CBA) is a method to reduce uncertainty during decision making and planning by replacing opinions, believes, and emotion by a framework for identification and determination of the benefits and cost, respectively of each alternative GIS.   
The results of CBA provide a basis for comparing GIS options. Public sector decisions are thought to be more complex because both policy and financial impacts must be considered. Private enterprises need to be concerned only with the accountant's 'bottom line'. In fact, though, both private and public sector decisions are better when they consider all aspects of a given alternative, whether those aspects have a line in the balance sheet or not.   

What we face is the difference between theoretical eloquence and real-world practicality. The goal is to find a practical way to accomplish the task at hand - that is, performing the CBA - while not opposing the basic theory of cost-benefit analysis.   
  
  

9.1.1 Objectives       The objective of Cost /  benefit analysis is the assessment of the advantages of a specific GIS application over competitive solutions and traditional work procedures (i.e. not utilising a GIS)   Cost / benefit analysis should  include the following:     a precise estimate of the total cost of use and  ownership of a GIS application (including overhead) during the life cycle of the GIS application.  estimates of the subjective assessment of the GIS user interfaces by real users (often dissociated from the objective cost), which are comparable across different GIS applications  quantitative estimates of the time required to carry out key tasks. This time naturally ranslates into cost.

Counting the Costs       The abacus is a calculator whose earliest known use  is circa 500 B.C. by the Chinese civilization. Addition, subtraction, division and multiplication can be performed on a standard abacus.

• estimates of convenience, workload, and positive acceptance by users
• elaboration of the added value and benefit offered by the GIS application
• consideration of all realistic alternatives and competitive solutions.

9.2 Time aspects  
  
An aspect of CBA is that the results must be looked at over time. In the first year or years the costs tend to outweigh the benefits. The up-front outlays in planning, system setup, data automation, application development, staffing, and training do not produce immediate benefits. Typically benefits begin to accrue after the second year, and then can surpass the yearly costs on a continual basis.    
Thus, it is important to look at costs and benefits over time -- but not too long. There are too many uncertainties if the planning horizon is extended too far into the future. Hardware and software advances are expected to occur, but the impacts of those advances on the costs and benefits are just too uncertain.    
Experience shows that a six-year planning horizon is reasonable.   
 

9.3 Counting the costs  
  
9.3.1 Estimating the 'total cost of ownership' of a GIS application  
  
The objective is to understand the costs incurred for the user when adopting a GIS. The resulting estimate takes all monetary costs into account, based on objective estimates. These costs can be calculated for relevant competing GIS applications:   

• Define user groups and usage scenarios (as described in section 4).
• Select the GIS alternative which offers the best value to the customer. 

        For each usage scenario:   

• Identify the components that the GIS project or GIS purchase consists off. The components are explained in the next section.   
• Understand the relevant cost and benefit factors. These are all factors which may influence the decision of the customer during the lifetime of a GIS application (e.g. purchase and customisation costs, integration of legacy systems, introduction cost, maintenance, cost of migration to new technology).   
• Make up the balance sheets for each GIS alternative. This should include relevant cost items and other relevant factors (cost estimates for purchase, customisation, user training, operating and maintenance, and also estimates of the time spent for learning how to use the GIS application and for using it, and of the cost of this time). As a result, the sum of all costs, which occur during the lifetime of the GIS application (i.e.the total cost of ownership) is calculated. There may be different results for different usage scenarios.

9.3.2 Cost calculation  
  
The cost estimation is a vital link in the success or failure of the GIS project/purchase. However, the price of a GIS (hardware and software) is not the most important cost factor. Issues such as usability, learning and training cost, support, (future) vision of the vendor as well as data compatibility all affect the decision for a particular GIS.      

Click on the preview   in order to see a larger   version of this photo

The price of a GIS (hardware and software) is not the most important cost factor.

Roughly one can divide the costs of a GIS into the following  components:    1. Hardware    2. Software (base software, base GIS and additional    GIS modules)    3. Maintenance    4. Services (resources to fulfil the GIS project objectives, e.g. customisation)    5. Training    6. Data (if obtained from elsewhere)

Quite often training, consultancy, customisation, maintenance, data, etc. are grouped into Services except Hardware and Software.

9.3.3 Variation of costs  
  
The cost of a GIS installation and customisation for whatever purpose varies very much per application area and the job or project that the GIS is needed for. Let£s take two examples:   
  

Click on the preview         in order to see a larger         version of this photo

An enterprise buys a GIS for GeoMarketing.     GeoMarketing is a GIS application for planning a new site (based on the optimal location), where to place billboard in a city, customer penetration etc.      The price of a GeoMarketing system is around 50 KECU. The software is relatively cheap (viewer seat), but the data consumes more than 70% of the total price.   A department of a local authority buys a GIS to build their utilities network digitally. The department estimates 3 man year work. The software for three doer's seats, will cost around 60 KECU. Data is not a cost, since it will be build by the department, using the GIS.

9.4 Counting Benefits  

There are no fixed categories of benefits. This makes counting benefits much more difficult than counting costs.    
In a large, government-wide project benefits were grouped into four categories: direct, agency, government, and external. A valid equivalent for the private sector would be direct, departmental, company-wide, and external.   

• Higher productivity (of the end-user's final product). For instance compared to using old methods (hand drawn maps) productivity increases with a factor 4 to 6.    
• Resource reduction. By using a GIS, the number of people working in the department can be reduced. 
• 
  Although care should be taken with the statement. Many people have thought the GIS and automated mapping would costs jobs. The opposite proves to be true. Mostly the same number of people were employed, only the production went up.   
• Quicker response. In some applications, such as traffic management using a GIS, the time factor is important.    
• Easier (cheaper) maintenance. Often the real payoff for a GIS lies in the maintenance of data. The initial data capturing is the big hurdle to overcome (technically and price-wise). For instance the initial capturing may cost 400 KECU and the yearly maintenance cost 50 KECU. 
• 
   

9.5 GUI customisation 
Ideally, optimal GUI customisation should be finished before using the GIS application instead of adapting and changing layout and functionality when the GIS is in use. Users tend to get used to the workflows supported by the system and do not want to adapt to changes. Unless optimal customisation is a very expensive operation, it will normally pay off as an increase in productivity.   

GUI customisation is a part of the services which form a substantial part of the overall GIS sale.    
The task (and thus involved cost) of GUI customisation should be split in three categories:   
   

1. Simple customisation that the GIS offers. For instance rearranging menus so that only the valid options for the particular applications are presented to the end-user. The customisation effort mostly varies from a few days to two weeks. The costs are low, simple customisation should always be done to increase the productivity of the GIS.   
2. Intermediate customisation  could involve sequencing GIS commands to reflect the user' s workflow, automatically passing data from one command to the other, (standard) connection to other systems, databases, etc. These customisations vary from weeks to months and require a good business plan: Is it worth 40,000 ECU to optimise the workflow? Thinking 6 years into the future, the answer may very well be yes.   
3. Complex customisation requires an intensive study and analysis of non-standard tasks and problems that should be solved with the GIS. New commands, combination of GIS commands, automatic uploading and controlling of real time databases are examples. When complex GUI customisation is performed it is actually no longer part of the GUI customisation, but forms part of the "Services".   
  

9.6 Recommendations  
 
Above many issues were presented that should be considered as best practice when making a cost-benefit analysis. Below some main recommendations for the users: 

• Definition of the components that build up the costs.
• Modularise the components.
• Maturity.
• Plan customisation. 

 
9.6.1 Budget constraints  
 
The end-users of the GIS are generally not the people who determine or control the budget for the GIS purchase. The classical case is that the IT department demands a system that would require twice the budget than is made available by management. Simply increasing the budget is not an acceptable solution in most organisations. Dropping functionality could be an option. Ideally, each functional option should be described in cost saving.  The old adage "you get what you pay for" is relevant here. The costs can be quite high, but also the benefits if planning and implementation are properly executed.  
 

9.6.2 Time constraints  
 
If a project has to be finished in one year, with 12 man years of work, the cost for software and hardware will be quite high, because 12 GIS seats will be necessary. If the project could be done in three years, the cost of services will remain the same, but the hardware/software will drop substantially, only 4 GIS seats are needed. Time constraints should not be confused with time aspects. Time constraints are the factors that cause a different division of the (standard GIS) components, time aspects are the considerations that a GIS should not be seen as a one-off purchase. 
 

9.6.3 Quality  

Obviously high quality operations require more cost in services than medium or lower quality. For instance if a street network has to be captured with an accuracy of 1 meter, it requires much more resources (including quality control) than the case where an accuracy of 5 meters is sufficient. It could even be such that the method of capturing high quality data is different than low quality, which may again, double the costs.  
 

9.6.4 What is the user's goal ?   

GIS is not a miracle machine that solves all the customer's problems. It should be considered as a tool that forms part of a project of which the service cost is bigger than the system itself. A few objectives could be: 

• Cost savings. Install a GIS in order to save cost that would be spent without a GIS.
• Better response times. Install a GIS because the traditional workflow is too slow. 
• Higher quality. Install a GIS because the traditional work procedures produce inaccurate results 
• Other 

The key question always to keep in mind is "what is the user's goal". It has been demonstrated in many failed IT projects that a seemingly wonderful system does not solve the desired problem because it was poorly focused and aimed at the outset. User-centred design and development concepts can help to get at the user's true work flow, and then can help design software which augments that workflow or even redesigns (reengineers) and radically improves the workflow.   
   

9.7 Conclusion  
Cost-benefit analysis is not an easy undertaking. The guidelines suggested here offer a structured approach to counting the costs and benefits associated with each potential output of the GIS. The resulting benefit/cost ratios are used along with other key information to complete a strategic plan for GIS development. A good strategic plan will indicate when purchases should be made and when each justifiable output can be undertaken.   
If desired, a number of strategies could be produced, for instance, to compare the long-term results of a high front-end investment versus a more steady rate of implementation at a longer-term. When a clear picture of the costs and benefits of a GIS is presented it can be compared to other competing alternatives to identify the best use of resources.    
 

9.8 Examples  

Three ficticious examples of GIS purchases are given. Case 1 was rated as a bad expenditure, case 2 as medium, and case 3 as a good estimate. In order to keep the information confidential, no names of organizations that were involved are mentioned.   
    
        Case 1: Municipal GIS(1)   
        Case 2: Municipal GIS(2)   
        Case 3: GIS Laboratory at University