Chapter 15 of 32

Feasibility Study

Before a builder starts a flat, someone checks whether the land can carry the building, whether the money adds up, whether the approvals can be obtained, and whether it can be finished before the buyers' deadline. Starting first and checking later would be far more expensive.

Software projects need the same check. A college might want a mobile app for all its services, a hospital might want to digitize patient records, a shop might want an online store. Before spending lakhs of rupees on requirements, design and code, management needs an honest answer to one question: should we build this at all, and if so, which way?

A feasibility study is a short, early investigation that evaluates whether a proposed software system is worth developing and can realistically be developed, by examining its technical, economic, legal, operational and schedule aspects, and recommending whether (and how) the project should proceed.

In Requirements Engineering we saw that Sommerville places a feasibility study before or at the start of the requirements process, and in the Spiral Model the same idea appears as risk analysis in every loop. The output is a feasibility report with a clear go / no-go / change-scope recommendation.


When and Why It Is Done

  • When: at the very start, after an initial problem statement or business case, and before detailed requirements, design and the Software Requirements Specification (SRS). It may be revisited if the project changes a lot.

  • Why: to avoid investing in projects that cannot be built, cost more than they return, break laws, will not be used, or cannot be finished in time. Stopping a bad project at this stage costs a tiny fraction of stopping it after development.

  • How long: short, usually days or a few weeks. It is not a full analysis; it gathers just enough information to decide.


Types of Feasibility (TELOS)

A common way to remember the dimensions is TELOS: Technical, Economic, Legal, Operational, Schedule.

TypeKey questionTypical checks
TechnicalCan we build it with available technology and skills?Hardware, software, platforms, performance needs, integration with existing systems, team expertise, technical risks
EconomicIs it worth the money?Cost-benefit analysis, payback period, ROI, NPV, budget availability
LegalDoes it comply with laws, regulations and contracts?Data protection, IT and cyber laws, licences of third-party software, copyrights, sector regulations (finance, health)
OperationalWill it work in the organization and will people use it?User acceptance, fit with work processes, training needs, management support, resistance to change
ScheduleCan it be finished in the time required?Deadlines (legal, market, academic year), realistic duration, resource availability

Some textbooks list only three (technical, economic, operational) and treat legal and schedule as parts of them. Others add market feasibility or resource feasibility. In exams, TELOS is the safest complete answer.


Technical Feasibility

Questions to ask, using the hospital records example:

  • Is the required technology mature and available? (Cloud hosting, encrypted storage, mobile apps: yes. Fully automatic diagnosis from scanned handwriting: risky.)

  • Can it meet performance requirements? (Can 300 doctors search records in under 2 seconds?)

  • Can it integrate with existing systems (lab machines, billing software)?

  • Does the team have the skills, or can they be hired or trained in time?

  • What are the biggest technical risks, and can a small prototype or spike reduce them? (See Prototyping Model.)


Operational Feasibility

A technically perfect system fails if nobody uses it. Operational feasibility asks whether the system will solve the actual problem and fit how people work. A classic checklist is PIECES (James Wetherbe), used to judge whether the problem is worth solving and whether the proposed system solves it:

LetterAreaQuestion about the new system
PPerformanceWill it improve throughput or response time?
IInformationWill it give better, timelier, more accurate information?
EEconomicsWill it reduce costs or increase benefits?
CControlWill it improve security, accuracy and control (fewer errors and fraud)?
EEfficiencyWill it make better use of people, time and materials?
SServicesWill it improve service to customers and users?

Also consider: Do users and management support it? Will staff need training? Will someone lose power or jobs and resist? A canteen app that the kitchen staff refuse to check will fail no matter how good the code is.


For software in India, typical questions include:

  • Does it collect personal data? If so, it must follow the Digital Personal Data Protection Act, 2023 and its rules (consent, purpose limitation, security safeguards, breach reporting).

  • Does it fall under the Information Technology Act, 2000 (electronic records, signatures, cyber security obligations)?

  • Sector rules: payments (RBI guidelines), health records, telecom, education.

  • Are all third-party libraries and tools licensed for this use (open-source licence obligations, paid licences)?

  • Contract terms with clients and vendors: liability, intellectual property ownership, service-level agreements.

Schedule

A college admission portal that is ready in August is useless if admissions open in June. Schedule feasibility compares a realistic estimate of duration (see the Project Scheduling: Gantt Charts, PERT & CPM and COCOMO Model chapters later in this tutorial) against hard deadlines, and considers options such as phased delivery of the most important features first.


Economic Feasibility: Cost-Benefit Analysis

This is the part with numbers, and the part most often asked in exams.

Costs and benefits

CategoryExamples
Development (one-time) costsAnalysts' and developers' salaries, hardware purchase, software licences, training, data migration
Operating (recurring) costsCloud hosting, maintenance staff, licence renewals, support, SMS and payment gateway charges
Tangible benefits (measurable in money)Fewer staff hours, reduced paper and printing, fewer errors, more sales
Intangible benefits (hard to measure)Better customer satisfaction, better decisions, improved reputation, staff morale

Intangible benefits are real but cannot be added into formulas; the report mentions them separately.

The four standard measures

MeasureFormulaDecision rule
Payback periodTime until cumulative net benefits equal the initial investment. With equal yearly net benefit: initial cost ÷ annual net benefitShorter is better; compare with a maximum acceptable period
Return on Investment (ROI)(Total net benefits − total cost) ÷ total cost × 100%Higher is better; must exceed a target rate
Present value (PV)PV = amount ÷ (1 + r)t, for an amount received after t years at discount rate rMoney later is worth less than money now
Net Present Value (NPV)Sum of PVs of all yearly net benefits − initial costNPV > 0: project earns more than the discount rate; choose the option with the higher NPV

Worked example: Option A

A college plans an online canteen ordering system.

  • Development cost (year 0): ₹12,00,000

  • Annual benefits for 5 years (less queue staff, less food waste, more sales): ₹5,50,000

  • Annual operating cost (hosting, maintenance, SMS): ₹1,50,000

  • Discount rate: 10% per year; evaluation period: 5 years

Step 1: annual net benefit.

Net benefit per year = 5,50,000 − 1,50,000 = ₹4,00,000

Step 2: simple payback period.

Payback = 12,00,000 ÷ 4,00,000 = 3 years

Step 3: simple ROI over 5 years.

Total net benefit = 5 × 4,00,000 = ₹20,00,000
ROI = (20,00,000 − 12,00,000) ÷ 12,00,000 × 100 = 66.7%

Step 4: NPV at 10%. Discount factor for year t = 1 ÷ 1.1t.

Year  Net benefit   Factor   Present value   Cumulative PV
 1     4,00,000     0.9091      3,63,636        3,63,636
 2     4,00,000     0.8264      3,30,579        6,94,215
 3     4,00,000     0.7513      3,00,526        9,94,741
 4     4,00,000     0.6830      2,73,205       12,67,946
 5     4,00,000     0.6209      2,48,369       15,16,315

NPV = 15,16,315 − 12,00,000 = ₹3,16,315   (positive → worthwhile at 10%)

(The factors add up to 3.7908, so a quicker way is 4,00,000 × 3.7908 − 12,00,000.)

Step 5: discounted payback. Cumulative PV crosses ₹12,00,000 during year 4:

After year 3: 9,94,741  → still need 12,00,000 − 9,94,741 = 2,05,259
Year 4 PV   : 2,73,205
Discounted payback = 3 + 2,05,259 ÷ 2,73,205 ≈ 3 + 0.75 = 3.75 years

Notice the discounted payback (3.75 years) is longer than the simple payback (3 years), because future rupees are worth less today.

Comparing two options

Option B is a cheaper off-the-shelf product customized for the college: cost ₹8,00,000, with net benefits growing as adoption increases: ₹2,00,000, ₹2,50,000, then ₹3,00,000 in each of years 3, 4 and 5.

Year  Net benefit   Factor   Present value   Cumulative PV
 1     2,00,000     0.9091      1,81,818        1,81,818
 2     2,50,000     0.8264      2,06,612        3,88,430
 3     3,00,000     0.7513      2,25,394        6,13,824
 4     3,00,000     0.6830      2,04,904        8,18,728
 5     3,00,000     0.6209      1,86,276       10,05,005

NPV = 10,05,005 − 8,00,000 = ₹2,05,005

Simple payback for B: after 3 years cumulative net benefit is 7,50,000; the remaining 50,000 comes from year 4's 3,00,000, so payback = 3 + 50,000 ÷ 3,00,000 ≈ 3.17 years. Total net benefit = 2,00,000 + 2,50,000 + 3 × 3,00,000 = 13,50,000, so simple ROI = (13,50,000 − 8,00,000) ÷ 8,00,000 = 68.75%.

MeasureOption A (custom)Option B (off-the-shelf)Better
Initial cost₹12,00,000₹8,00,000B
Simple payback3.00 years3.17 yearsA
Simple ROI (5 years)66.7%68.75%B
NPV at 10%₹3,16,315₹2,05,005A
Discounted payback3.75 years3.91 yearsA

The measures disagree, which is common. ROI is a ratio, so it favours the smaller investment; NPV measures the absolute value created, taking the time value of money into account. When options are mutually exclusive and money is available for either, financial practice prefers the option with the higher NPV, here Option A. The final recommendation would also weigh non-economic factors: B is less risky technically and faster to deploy, while A fits the college's processes better.


The Feasibility Study Process

  1. Understand the problem and objectives: what is wrong now, and what must the new system achieve?

  2. Study the current system (manual or automated): its workflow, costs and weaknesses.

  3. Identify alternative solutions: do nothing, improve the manual process, buy off-the-shelf, customize an existing product, build custom software, outsource.

  4. Evaluate each alternative against TELOS: technical, economic, legal, operational, schedule.

  5. Choose and justify the best alternative.

  6. Write the feasibility report and present it to decision makers.

Contents of a feasibility report

1. Introduction: problem statement, objectives, scope of the study
2. Current system: description and problems
3. Alternatives considered
4. Evaluation of each alternative
     technical | economic (cost-benefit) | legal | operational | schedule
5. Recommended alternative and justification
6. Risks and assumptions
7. Recommendation: proceed / proceed with changes / do not proceed
8. Appendices: cost tables, calculations, survey results

Common Mistakes

  • Considering only one alternative. "Do nothing" and "buy instead of build" should always be compared.

  • Ignoring recurring costs. Hosting, maintenance and licences over five years can exceed the development cost.

  • Ignoring the time value of money. Adding rupees from different years without discounting overstates benefits.

  • Choosing by ROI alone between options of very different sizes; check NPV too.

  • Confusing feasibility with requirements analysis. Feasibility is short and decides whether; detailed requirements come after a "go".

  • Treating technical feasibility as the only feasibility. Many projects fail operationally (nobody uses them) or legally.


Interview and Exam Questions

Q1. What are the types of feasibility?

Technical, economic, legal, operational and schedule (TELOS). Some texts also include market or resource feasibility.

Q2. A system costs ₹6,00,000 and gives a net benefit of ₹1,50,000 per year. Find the simple payback period.

6,00,000 ÷ 1,50,000 = 4 years.

Q3. What is the present value of ₹1,21,000 received after 2 years at 10%?

PV = 1,21,000 ÷ 1.1² = 1,21,000 ÷ 1.21 = ₹1,00,000.

Q4. Cost ₹10,00,000; net benefits ₹3,00,000 per year for 4 years. Find simple ROI over 4 years.

Total net benefit = 12,00,000. ROI = (12,00,000 − 10,00,000) ÷ 10,00,000 = 20%.

Q5. What is the difference between tangible and intangible benefits? Give examples.

Tangible benefits can be measured in money (reduced staff cost, less paper, more sales). Intangible ones cannot be measured directly (customer satisfaction, better decisions, reputation). Both are reported, but only tangible ones enter the cost-benefit calculations.

Q6. What does operational feasibility evaluate?

Whether the proposed system will solve the problem and be accepted and used in the organization: fit with workflows, user and management support, training needs and resistance to change (PIECES is one checklist).

Q7. Why can the NPV and ROI rankings of two options differ?

ROI is a ratio and ignores scale (and, in simple form, timing), so a small cheap project can have a high ROI but create less total value. NPV measures total value in today's money, discounting later benefits. For mutually exclusive options, prefer the higher NPV.


Summary

  • A feasibility study is a short early investigation deciding whether and how a project should proceed.

  • Dimensions: Technical, Economic, Legal, Operational, Schedule.

  • Economic feasibility uses cost-benefit analysis: payback period, ROI, present value and NPV.

  • NPV = Σ (net benefitt ÷ (1 + r)t) − initial cost; positive NPV means the project beats the discount rate.

  • Always compare alternatives, including "do nothing" and "buy instead of build", and include recurring costs and intangible benefits.

  • The result is a feasibility report with a go / no-go / change-scope recommendation.

The cheapest project to cancel is the one you cancel before it starts.