You reviewed several proposals and discussed the responses with your peers in the previous step. Now it is time to consider the financial or cost proposal elements of the RFP responses and rank them in terms of favorability.
While the cost associated with a proposal should not be the primary decision-making criterion, it is important to analyze the financials in order to understand the feasibility for the vendor to actually provide the requested services, whether the proposed costs are within your budget, and to ensure that all relevant costs are included.
Use the financial proposal ranking template to rank these financial proposals in order from most favorable to least favorable.
When evaluating RFP responses for selection, the financial or cost proposal should not be used as the primary decision-making criterion, but at a minimum you should conduct an analysis of the financial proposal to assess the following:
In evaluating the financial proposal, it may help to use criteria such as the following:
by Albana Madhi (Kacollja), Gjergji Shqau, and Aldi LekajAbstract: This article incorporates a theoretical and practical approach to cost-benefit analysis to help company managers to decide whether to adopt cloud computing. Cost-benefit analysis is divided into six distinct sections to determine what costs and benefits are economically significant for the case study analyzed.Based on the allocation sections, TEI (total economic impact) is applied. This methodology was chosen because it gives a general view of the impact and benefits in information technology to cloud computing adoption not only through the analysis of costs and benefits but also giving appropriate weight to existing information technology resources.The article deals with analysis, forecasts, and implementation of cloud computing, showing financial benefits acquired from “green IT” as a consequence from this technology. Through a detailed analysis of the cloud computing phenomenon and its connection with green IT, this article conducts a cost-benefit analysis as a case study using an Internet and communications services company. The aim is to show the potential benefits of the transition to cloud computing, benefits expressed in increased efficiency of the use of technological resources, or even increased environmental benefits.
Specifically, we followed these steps:
Since large undertakings have a sophisticated investment analysis, particularly with regard to IT, TEI methodology provides a picture of the whole economic impact.
The project examines what will happen from the cost-benefit point of view in the case of transition services for PRIMO Communications from on-site compared with use of Amazon cloud services (AWS, Amazon Web Services) for the data center.
We have built a general picture for the total economic impact (TIE) of this change from an on-site data center to Amazon cloud service. Our hypothesis is that benefits of the shift to cloud computing data center will reduce the total operational cost. Many of the benefits of this transition are not easily translated into numerical values.However, there are three main areas where operational costs will undergo reduction:
Engineering staff productivity will increase due to higher efficiency of systems and time management. The value and quality of products and services offered will generate increased consumer satisfaction.Although the results of cost-benefit analysis were positive, environmental benefits are difficult to define and to determine if they play a decisive role in the choice of cloud computing. This relates to the fact that many of the biggest providers of cloud computing build huge data centers and use energy in order to increase their effectiveness. They do not take into account that they are destroying renewable energy. Secondly, the current legislation in force, including the developed countries of the most experienced in this field, do not give priority to environmental protection.
A cost-benefit analysis in cloud computing transition was done for the company PRIMO Communications. The project analyzes the positive sides but also the risks if the company uses the services offered by Amazon.PRIMO Communications offers a package of services in Internet, telephone, and television. Since the number of users of these services is always growing, there is a constant need for capacity to offer quality and consistency.
The action plan for the transition to the cloud is organized as follows.
The main economic benefit of technology cloud comes from the type of payment. This is a kind of payment, based on use, usually known as the conversion of capital costs (CAPEX) to operating costs (OPEX). The payment model we use differs from the usual rent because the latter is a charge made for a certain period of use of computer resources in this case, even though they cannot be used at this time. The payment model we use is completely independent of the period of use but appears only in case of use of those resources.
Buying a computer or an application capacity (SaaS) is delivered as a cloud service technology known as operating expense (OPEX) that may increase or decrease depending on the request during an interval of time. The same solution in a data center in an organization is a sunk cost known as capital cost (CAPEX) and presented in the balance sheet as assets and amortized over the longer term.Transformation of capital costs in operating expenses makes financial decisions easier and less riskier. This is summarized in the table below.Comparison of CAPEX and OPEX
| Factor | Traditional Technology | Cloud Technology |
| 1-Type of expense | Capital expense (CAPEX) Operational expenses (OPEX) | Just (OPEX) |
| 2-The flow of cash | 0 | Payments for offered service |
| 3-Income | Maintenance expenses Amortization of capital expenditures | Maintenance payments |
| 4-Balance | Long-term assets 1-Hardware 2-Software | Not recorded in the balance 1-Hardware 2-Software |
A comparison between cloud computing and a data center is difficult because of the hidden and indirect costs that appear in a data center firm (in-house).There are many arguments and counterarguments regarding the total cost (TCO) between the types of cloud computing technology and in-house because the costs are specific and depend on the type of business.
Cash flows that are used in financial analysis start with an initial investment column reflecting costs incurred at time 0 or at the beginning of year 1. These costs are not discounted. Calculations for the present value (PV) were performed for each total cost and benefit. Present value calculations that are shown in the table are summarized and include the amount of inflows and outflows, with a discount rate every year. The table below lists the main assumptions used for measurements.The table provides approximate estimates of rates of annual use of data center processing nodes during the year. A use of 100% means that the server is using 8736 hours each year. This was measured by 24 hours * 7 days * 52 weeks. Some types cannot be shared servers in use, which means that their use is only intended for a single user. These are mainly small computers. Shared servers can be used by several users at the same time.The Percentages of Annual Use of Processing Nodes
| Ref | Type of Servers | Typical Tasks | Typical Usage Model | Number of Hours Per Year | Annual Percentage of Use |
| A11 | Small | For the daily office tasks | 8 hrs * 5 days * 47 weeks | 1880 | 21% |
| A12 | Small | For software testing purposes | 8 hrs * 5 days * 47 weeks | 1880 | 21% |
| A13 | Medium | For different customers | 24 hrs * 7 days * 52 weeks | 8736 | 100% |
| A14 | Big | For larger software development, documentation | 10 hrs * 5 days * 47 weeks | 2350 | 27% |
| A15 | Very big | Servers for the main services in the company | 24 hrs * 7 days * 52 weeks | 8736 | 100% |
When companies own and operate servers in their offices, they need to hire full-time personnel to select, acquire, set, support, and manage them. Server administrators perform a variety of tasks, including updating the server, returns, movement, incident management and problem management, and monitoring of salespeople. The table below gives a forecast of annual salary for this staff, which will consist of an IT manager, two system administrators, and four developers.Annual Cost of Staff Salaries
| Ref | Measurement Unit | Basic Cost | Number | Decreasing rate | Salary |
| B4 | IT manager’s salary | 140,000 | 1 | 100% | 140,000 |
| B5 | System administrator’s salary | 110,000 | 2 | 100% | 220,000 |
| B6 | Developer’s salary | 70,000 | 4 | 20% | 56,000 |
| Total | 416,000 |
The data center consists of only one main equipment room. We only included cost estimates that were visible and important. Ancillary costs such as architecture costs, cabling, real estate companies, the building management, security, etc., are not very easily seen, and maintaining and operating an infrastructure includes the cost other than those listed below, but we will assume that the following proposed assessment is reliable enough.Number of Computers
| Type of Server | Hardware Specifications | Equivalent to AWS Specifications | Number of Computers |
| Small server | Platform 32-bit, 2 GHz CPU, 2 GB memory , 160 GB disc | Small instance 1.7 GB of memory, 1 EC2 compute unit (1 virtual core with 1 EC2 compute unit), 160 GB storage, 32-bit platform | 3 |
| Medium server | Platform 64-bit, 1 quad-core2 GHz CPU, memory 8 G RAM,500 GB disc | Large instance 7.5 GB of memory, 4 EC2 compute units (2 virtual cores with 2 EC2 compute units each), 850 GB storage, 64-bit platform | 5 |
| Big server | Platform 64-bit, 2 quad-core2 GHz CPUs, Memory 15 GB ,850 GB disc | Extra-large instance 15 GB of memory, 8 EC2 compute units (4 virtual cores with 2 EC2 compute units each), 1690 GB storage, 64-bit platform | 6 |
| Huge server | Platform 64-bit, 4 quad-core2 GHz CPUs, Memory 32 GB,1 TB disk | High-memory double extra-large instance 34.2 GB of memory, 13 EC2 compute units (4 virtual cores), 850 GB of local instance storage, 64-bit platform | 4 |
| Total | 18 |
Cost of hardware is based on the total cost of acquisition, operation, and substrate that comprises the data center servers, network equipment, and hardware maintenance. Costs that include operating system software and application licenses and lease of the building costs are not calculated in the financial analysis. This is because these costs are unchanged in terms of passing the assets of the company to the Amazon cloud. When calculating the financial impact of a data center, the cost of replacing computers that is included in depreciation costs should be considered. As such, the initial cost of buying the hardware will be amortized over the useful life of the asset. In this case, it is about three years (A4). Therefore, using a fair depreciation, expenses for servers are minus the initial purchase at the end of life of dividing this value received for life. For simplicity, we will assume a value equal to zero rescue. Consequently, the annual cost of servers is calculated as follows:Annual server cost = (total number of servers * by cost per server) / A4Network Equipment CostServers allow network devices to connect to each other and to the Internet. Network devices include firewalls, routers, switches, and intrusion detection systems and other equipment. We have assumed costs for equipment equal to 20 percent (A5) of the original value of the cost of amortizing the right server for three years of useful life (A4). The annual costs of network equipment is calculated by:Annual cost of network equipment = (total cost per server * by server * A5) / A4
Servers and network devices are typically purchased with an annual contract maintenance plan for equipment repair in case of hardware defects. The annual maintenance cost is expressed as a percentage of the original purchase cost of servers and network devices, approximately 10 percent (A6). The annual cost of maintenance for servers and network equipment is calculated as follows:
Servers Operating Costs and Energy Costs for Cooling the ServersData center servers not only consume energy, but also transform energy into heat that must be removed from to avoid overheating equipment. A study carried out by the Green Grid estimated that most data center PUE ratio (power usage effectiveness) have a value between 1.3 and 3.0. In this calculation, we assumed a PUE of 2.5 (A3) as a conservative value. The reality may be even worse than this value because few measures are taken for increasing efficiency. The following equation was used to calculate the full power of estimating the energy demand of server multiplying it with PUE. Total energy use is multiplied by the average cost of electricity per kW/hour to assess the total annual cost of power. The annual costs for operation of servers and their cooling is calculated with the following formula:
The table below gives an estimate of the annual costs associated with maintenance and operation of the PRIMO Communications data center.Annual Cost of Operating and Maintaining the Data Center in Primo CommunicationsData Center Servers Cost
| Ref | Item | Cost Per Unit | Number of Units | Basic Cost | Year 1 | Year 2 | Year 3 |
| C1 | Small servers | 41,000 | 3 | 123,000 | 41,000 | 41,000 | 41,000 |
| C2 | Medium servers | 145,000 | 5 | 725,000 | 241,667 | 241,667 | 241667 |
| C3 | Big servers | 295,000 | 6 | 1,770,000 | 590,000 | 590,000 | 590,000 |
| C4 | Huge servers | 597,000 | 4 | 2,388,000 | 796,000 | 796,000 | 796,000 |
Data Center Equipment Costs
| Ref | Item | Cost Per Unit | Number of Units | Basic Cost | Year 1 | Year 2 | Year 3 |
| C5 | Network equipment | 1,001,200 | 333,733 | 333,733 | 333,733 |
Hardware Maintenance Cost of the Data Center
| Ref | Item | Cost Per Unit | Number of Units | Basic Cost | Year 1 | Year 2 | Year 3 |
| C6 | Hardware maintenance | 33,733 | 33,733 | 33,733 | |||
| C7 | Total hardware cost | 6,007,200 | 2,035,773 | 2,035,773 | 2,035,773 |
Switching On and Cooling Costs of the Data Center
| Ref | Item | Cost Per Unit | Number of Units | Basic Cost | Year 1 | Year 2 | Year 3 |
| C8 | Cooling (kW) | 12 lek/kW | 512,000 | 2,048,000 | 2,048,000 | 2,048,000 | |
| C9 | Total cost of the data center | 4,083,773 lek | 4,083,773 lek | 4,083,773 lek |
Adding annual staff cost to data center costs within PRIMO Communications:Actual Costs
| Ref | Item | Year 1 | Year 2 | Year 3 |
| D1 | Staff total costs (B7) | 416,000 | 436,800 | 458,640 |
| D2 | Data center total costs (C9) | 4,083,773 | 4,083,773 | 4,083,773 |
| D3 | Basic total costs | 4,499,773 lek | 4,520,573 lek | 4,542,413 lek |
14.1 PHASE COSTS OF PLANNING AND DESIGN
Less time will be needed for project planning, understanding of AWS technology, necessary design changes and in particular the design of modifications to be performed in a virtual portal application. This portal is an application that was designed and built within PRIMO Communications to facilitate the provisioning of virtual machines. This application needs to be modified in order to enable the creation and deployment of instances and volumes (EBS) in the AWS cloud. There is a need to develop additional functionality to support the rights of access control and automatic shutdown of computer resources that are not used. No external professional services will be required to assist in the design and implementation of the project. Engineers working in PRIMO Communications are virtualization experts, so the manager in charge of this project should not have trouble finding appropriate resources to properly design the project. However, the project will require time from other departments from networking and software engineers to give a more accurate solution. It is anticipated that efforts to become familiar with AWS technology and design of the solution will require about 20 days. The average full salary workers likely to be involved in the planning and designing of the project is around 46400 daily.Costs of Planning and Design Phase, Without Risk Adjustment
| Ref | Unit | Calculation | Beginner |
| E1 | Work day | 20 | |
| E2 | Cost of daily charge | ((B1+B2+B3)/3)/A7 | 46,400 |
| E3 | Planning and design costs | E1*E2 | 928,000 |
14.2 COSTS OF IMPLEMENTING THE PROJECT
The implementation phase of the project will be determined in about five months, which includes the transfer of servers. The initial stages and those that will follow it will require sufficient time for engineers to make the transition (migrate) the development of their existing environment to environment application using AWS—two days per server for a total of 36 days.Cost of the Project Implementation, Without Risk Adjustment
| Ref | Unit | Calculation | Beginner |
| F1 | Work day | 30 | |
| F2 | Cost of daily charge | ((B1+B2+B3)/3)/A7 | 46,400 |
| F3 | Planning and design costs | F1*F2 | 1,392,000 |
14.3 CLOUD INFRASTRUCTURE COSTS
In order to have an accurate financial analysis, we determined the exact number of servers that will be transferred to AWS and the number of those who stay in the existing data center. Price for each hour varies by type of instance and by geographical location. Amazon offers several possibilities for the manner of payment, as follows:
Therefore, determining the number of instances you need and how long they are needed is the first step toward assessing the annual costs of AWS instances. For this purpose, we defined two categories of instances to be provisioned to meet the demands for resources during the year:
Annual use of instances per hour base load is calculated as:
The table below provides a summary of all costs for a cloud solution.Cloud Infrastructure Without Risk Adjustment
Baseline instance based on AWS hourly use/rate
| Ref | Units | Cost per Unit | Number of Units | Basic Cost | Year 1 | Year 2 | Year 3 |
| J1 | Small server | 3.9 lek | 5242 | 11,925 | 20,448 | 20.448 | |
| J2 | Medium server | 16.1 lek | 8387 | 78,764 | 135,030 | 135,030 | |
| J3 | Big server | 30.1 lek | 34,944 | 613,523 | 1,051,814 | 1,051,814 | |
| J4 | Huge server | 50.2 lek | 17,472 | 511,609 | 877,094 | 877,094 |
Peak instances loaded in AWS planned for their use
| Ref | Units | Cost per Unit | Number of Units | Basic Cost | Year 1 | Year 2 | Year 3 |
| J5 | Small server | 8.1 lek | 0 | ||||
| J6 | Medium server | 30.5 lek | 350 | 6227 | 10,675 | 20,675 | |
| J7 | Big server | 70.9 lek | 2097 | 86,734 | 148,677 | 148,677 | |
| J8 | Huge server | 130.2 lek | 874 | 66,376 | 113,795 | 113,795 | |
| J9 | Volume EBS (GB) | 8.9 | 4533 | 23,532 | 40,344 | 40,344 | |
| J10 | EBS IOPS in millions | 8.9 | 8286 | 43,016 | 73,745 | 73,745 | |
| J11 | Total cost of cloud | 1,441,706 | 2,471,637 | 2,471,637 |
14.4 STAFF COSTS
Since there will be fewer physical severs to maintain the existing data center and since each engineer will be able to provide authority previously configured automatically via the virtual application, thought loads of staff data center management will be reduced significantly, especially for both systems administrators. The table below provides a recalculation of their annual salary based on the assumption that system administrators will spend less time on operations and will have free time to focus on other activities.Recalculation of Annual Salary
| Ref | Measurement Unit | Basic Costs | Number | Decrease Rate | Salary |
| K1 | IT manager salary | 140,000 | 1 | 100% | 140,000 |
| K2 | System administrator salaries | 110,000 | 2 | 80% | 176,000 |
| K3 | Developers salaries | 70,000 | 4 | 10% | 28,000 |
| K4 | Total | 344,000 |
14.5 TOTAL PROJECT COSTS OF HYBRID CLOUD
The table below summarizes the overall project costs associated with the design, implementation, construction of infrastructure, and staff salaries by 5 percent annual growth. The present value (PV) is calculated as follows:PV = initial cost + cost of year 1 * .9091 + .8264 cost of year 2 * + * .7513 cost of year 3Summary of Total Costs of the Project
| Ref | Units | Starting Cost | Year 1 | Year 2 | Year 3 | Total | Actual Cost |
| L1 | Planning and project cost (E3) | 928,000 | 928,000 | 928,000 | |||
| L2 | Implementation costs (F3) | 1,392,000 | 1,392,000 | 1,392,000 | |||
| L3 | Staff costs (K4) | 344,000 | 361,200 | 379,260 | 1,084,460 | 896,164 | |
| L4 | Cloud infrastructure costs (J11) | 1,441,706 | 2,471,637 | 2,471,637 | 6,384,980 | 5,210,157 | |
| L5 | Total costs of the project | 2,320,000 | 1,785,706 | 2,832,837 | 2,850,897 | 9,789,440 | 8,426,321 |
Most of the benefits that can be thought of in this project are not easily measurable in terms of return on investment (ROI). The hypothesis is that the benefits of cloud computing should obviously reduce the overall costs of operations performed in PRIMO Communications. The purpose of this section is to express these benefits in money so we can put a monetary value to their effects. Some calculations made in this section shall be conducted under uncertainty. Thus, we have two kinds of benefits:
However, qualitative benefits are just as important as quantitative and should be considered in the final evaluation of the project. The following assumptions concerning the assessment of benefits are accomplished by following the categorization of benefits according to TEI.
15.1 BENEFIT FROM INCOME
PRIMO Communications earns income from the sale of products and services. The objective of this project is to cut costs where possible while staying within the budget that is allocated. What is interesting in the category of benefits by income reductions are operational costs that can be achieved by the use of outside resources at PRIMO Communications. As shown in the table below, there are two areas in which reductions in major costs can be achieved:
Operational Benefits Without Risk Adjustments
| Ref | Units | Calculations | Year 1 | Year 2 | Year 3 | Total |
| M1 | Benefits from electricity power | C8-0 | 2,048,000 | 2,048,000 | 2,048,000 | 6,144,000 |
| M2 | Benefits from paying staff salary | D1-L3 | 72,000 | 75,600 | 79,380 | 226,980 |
| M3 | Total operational benefits | 2,120,000 | 2,123,600 | 2,127,380 | 6,370,980 |
Key Financial Indicators: Financial Indicators with Risk Adjustment
| Start | Year 1 | Year 2 | Year 3 | IRR | Payback | POI | NPV | |
| Costs | -2,320,000 | -1,785,706 | -2,832,837 | -2,950,897 | ||||
| Benefits | 3,294,667 | 3,298,267 | 3,302,047 | |||||
| Savings | -2,320,000 | 1,508,961 | 465,430 | 451,150 | 3.5% | 3 | 4.6% | 503,508.05 |
The internal rate of return (IRR) of an indicator is used in finance to compare odds from the investment. It is calculated as below:
Payback is the period that the company should take the return from the initial investment, i.e., to refund the amount of the initial investment. This indicator is expressed in years. Calculate returns based on savings rather than benefits.Savings Year 1 + Year 2 + Savings Year 3 > = initial investmentNPV is the best indicator for decision making if a project will be resolved or not. It is calculated:
NPV = 1508961 / (1 +0.1) 1 + 465 430 / (1 +0.1) 2 + 451 150 (1 +0.1) 3-2320000 = 503,508.05Since NPV of the project has been positive and as NPV is the best method of assessing whether a project should be elected or not, the company should choose the project.
Recommendations
part 2
In the previous step, you analyzed and ranked the financial proposals. Now, you’ll analyze and rank the proposals by their technical specifications.The general criteria that you should use includes the following:
You may wish to include additional evaluation criteria, based on the materials in the course, or your own research.