Economic Feasibility

Intangible benefits are real but cannot honestly be priced: time returned to people, less stress, better service, fewer mistakes, a better reputation. Measure them in their own units, such as hours or complaints, and do not force a rupee value on them.

Revenue is not benefit

This is the mistake to avoid. When a system helps a business sell more, the business does not gain the full price of each extra sale. It gains the profit on it: the price minus what the sale cost to make. A canteen that sells one more Rs 68 lunch has spent something on the ingredients, the gas and the cook's time, and gains only what is left.

Counting revenue as benefit makes every payback look several times faster than it is. The worked example below shows the mistake and then the corrected figure.

The measures

Cost-benefit analysis

Put the costs and the benefits side by side, over the same period, and compare. If benefits exceed costs over a period the owner cares about, the project is economically feasible.

Payback period

The payback period is the time the benefits take to repay the one-time cost:

It is the most useful single measure for a small system, because it answers the owner's real question: "how long before this has paid for itself?"

Return on investment, and money over time

For large projects running over years, analysts also compute the return on investment, the net gain as a percentage of the cost, and the net present value, which discounts money received in future years because a rupee next year is worth less than a rupee today. Both matter when payback takes years. When payback takes weeks, as it usually does for a well-chosen student project, they add little, and it is enough to say so.

Sensitivity: what if you are wrong?

Every benefit figure is an estimate. A careful analysis asks how the conclusion changes if the estimate is too optimistic, for example if only half the hoped-for benefit arrives. If the project is still worth doing under the pessimistic figure, the conclusion is robust. This is called sensitivity analysis, and one extra line of it makes a feasibility study far more convincing.

The worked project: economic feasibility

What the owner told them

In her interview (Chapter 4), the owner, Lata Pawar, gave the team four figures, all recorded as her estimates:

• the average lunch bill is Rs 68;
• her margin is about 25 per cent of the bill;
• about Rs 600 of food is thrown away on an ordinary day, measured at what it cost her to make;
• the canteen is open about 180 college days a year.

One-time costs

The development effort is recorded separately, in hours: 240 person-hours, 60 for each of the four members, as MU allots.

Running costs

Benefits: the first attempt, and what was wrong with it

Aditi's first draft of the benefits looked like this:

It gave a payback of 9,500 divided by 1,260.80, which is about 7.5 days. Farhan questioned the first line: the canteen does not keep Rs 68 from a sale, it keeps its margin. The first line was revenue, and the benefit is the profit.

Benefits: corrected

Half of the 31.2 students is 15.6 students a day. Each sale of Rs 68 at a 25 per cent margin leaves Rs 17 of profit.

The waste line was already right, because the Rs 600 was measured at cost: food not thrown away is money not spent.

Payback, and a year

So the tablet pays for itself within the 21st college day, a little over four weeks of college. Over a year of 180 college days, the benefit is 180 multiplied by 465.20, which is Rs 83,736, against a one-time cost of Rs 9,500.

The corrected payback is nearly three times longer than the first draft's, and it is still short. That is the honest conclusion, and it is the one that survives an examiner's question about margins.

Sensitivity

The team asked what happens if they are wrong about the recovered sales, and only a quarter of the students who leave today come back instead of half.

A quarter of 31.2 is 7.8 students a day, at Rs 17 of profit each, which is Rs 132.60. With the Rs 200 saved on waste, the benefit is Rs 332.60 a day, and the payback is 9,500 divided by 332.60, about 28.6 days. Still within six weeks of college. The conclusion does not depend on the optimistic figure.

Intangible benefits

The team left all four unpriced. Thirty-six hours of students' time a day is a striking number without a rupee sign beside it.

Verdict: economically feasible. For a one-time cost of Rs 9,500 and no running cost, the canteen gains an estimated Rs 465.20 a day, paying back within about four weeks of college days, and within six weeks even if only a quarter of the lost sales return.

Do this for your project

1. Ask the owner of the problem for the figures you need, and record each as their estimate: prices, margins, volumes, days of operation, what the problem costs them now.
2. List one-time costs and running costs, each with a note of where the figure came from.
3. Record the development effort in hours; do not invent a rate for it.
4. List tangible benefits in rupees, as profit or saving, never as revenue, and intangible benefits in their own units.
5. Compute the payback period, and a year's benefit.
6. Redo the payback with a pessimistic benefit, and say whether the verdict survives.
7. End with a one-sentence verdict that states its conditions.

Mistakes that cost marks

"Economically feasible because the software is free." It answers whether the students can afford to build it, not whether the owner should adopt it.

Revenue counted as benefit. The most common arithmetic error in feasibility studies, and an examiner who spots it will doubt every other figure.

Invented figures. A cost or rate with no source, dropped in to make a table look complete, weakens the real figures around it.

Intangibles given a price. "Student satisfaction worth Rs 50,000" is a guess dressed as a number.

No pessimistic case. A conclusion that depends on everything going well is a hope.

Quick revision

• Economic feasibility compares costs with benefits.
• Costs: one-time (development, hardware, software, setup, training) and running (hosting, network, maintenance, paid services).
• Record student effort in hours; do not invent a rate.
• Benefits: tangible in rupees, intangible in their own units.
• Revenue is not benefit: count the profit on extra sales, and savings at cost.
• Payback period = one-time cost divided by benefit per period.
• Sensitivity analysis: redo it with a pessimistic benefit; a robust verdict survives.

Questions you must be able to answer

1. Distinguish one-time costs from running costs, with an example of each for a canteen system. One-time costs are paid once to bring the system into use, such as a tablet bought for the counter; running costs are paid for as long as it is used, such as rented hosting or paid SMS messages.

2. Why is the price of a recovered sale not the benefit of recovering it? Because the business spent money to make what it sold. Its gain is the profit, the price less the cost of making the sale. Counting the full price overstates the benefit and makes the payback look far quicker than it is.

3. Compute the worked project's payback period, and explain each figure. The one-time cost is Rs 9,500 for the counter tablet. The daily benefit is Rs 265.20 of profit on 15.6 recovered sales, at Rs 17 each, plus Rs 200 of food no longer wasted, a total of Rs 465.20. Payback is 9,500 divided by 465.20, about 20.4 days.

4. What is sensitivity analysis, and what did it show for the worked project? Recomputing the result with a less favourable estimate to see whether the conclusion still holds. With only a quarter of the lost sales recovered, the benefit falls to Rs 332.60 a day and the payback rises to about 28.6 days, so the project remains worthwhile.

5. Why did the worked team not put a rupee value on the time returned to students? Because there is no honest price for a student's time, and a guessed figure would be a false precision that weakens the real figures. They measured it in its own unit instead: about 36 hours of students' time a day.