Cash Flow Analysis for Beginners: NPV, IRR, Payback Period
Building the cash flow series is the hard part, not the discounting. A worked project from revenue down to NPV, IRR, payback, PI and MIRR in one CF entry.
The short answer
Cash flow analysis turns a project into one series of after-tax incremental cash flows and then reads several measures off it. NPV is the value added at your cost of capital, IRR is the rate that sets NPV to zero, and payback is how long the money is at risk. On the BA II Plus all of them come from a single CF worksheet entry. The work that decides whether the answer is right happens before the keystrokes: excluding financing costs, adding depreciation back, and putting working capital in at both ends.
Most cash flow analysis goes wrong before any discounting happens. The BA II Plus will compute an NPV to the penny from whatever series you type in, and it cannot tell you that the series included loan interest, omitted working capital, or counted a survey you paid for last year.
So this works in the order the job actually happens: build the series, then load it once, then read six measures off it. The project is a £165,000 café fit-out, and every figure below comes from the same twelve numbers.
Step one: from revenue to free cash flow
Cash flow analysis values cash, not profit. The bridge between them is short and always the same shape.
-
FCF = (Revenue − Cash costs − Depreciation) × (1 − tax) + Depreciation
Equivalently EBITDA − tax on EBIT. Depreciation is subtracted only to compute the tax, then added straight back, because no cash left the business.
The equipment costs £140,000, has a six-year life and an expected resale value of £20,000, so straight line depreciation is £20,000 a year. The fit-out also ties up £25,000 of working capital — stock, deposits and card settlement float — which makes the day-zero outlay £165,000. Tax is 25%.
| Year | Revenue | Cash costs | EBITDA | EBIT | Tax | Free cash flow |
|---|---|---|---|---|---|---|
| 1 | £182,000 | £136,000 | £46,000 | £26,000 | £6,500 | £39,500 |
| 2 | £196,000 | £142,000 | £54,000 | £34,000 | £8,500 | £45,500 |
| 3 | £208,000 | £146,000 | £62,000 | £42,000 | £10,500 | £51,500 |
| 4 | £214,000 | £149,000 | £65,000 | £45,000 | £11,250 | £53,750 |
| 5 | £214,000 | £149,000 | £65,000 | £45,000 | £11,250 | £53,750 |
| 6 | £205,000 | £148,000 | £57,000 | £37,000 | £9,250 | £47,750 |
Year six is not finished. Two things happen when a project ends: the working capital comes back, and the asset is sold.
| Component | Amount |
|---|---|
| Operating free cash flow | £47,750 |
| Working capital released | £25,000 |
| Equipment sale proceeds | £20,000 |
| Tax on disposal | £0 |
| Year 6 total | £92,750 |
The disposal is untaxed here because book value after six years of £20,000 depreciation is exactly £20,000, so there is no gain. Sell for more than book value and the excess is taxable; sell for less and the loss shelters other profit. That link between the depreciation method and the disposal tax is why depreciation methods belong in a cash flow discussion at all.
Step two: one entry, six answers
Consecutive identical flows go in as one amount with a frequency, which is what the F registers are for.
Years 4 and 5 are both £53,750, so this project needs five groups rather than six entries.
-
CF -
2ND -
CLR WORK -
165000 -
+|− -
ENTER -
↓ -
39500 -
ENTER -
↓ -
↓ -
45500 -
ENTER -
↓ -
↓ -
51500 -
ENTER -
↓ -
↓ -
53750 -
ENTER -
↓ -
2 -
ENTER -
↓ -
92750 -
ENTER
Double arrow-down skips an F register and leaves it at 1. The single arrow-down before 2 is deliberate — that is F04, set to 2 to cover years 4 and 5.
Now every measure is a few keystrokes away, and none of them requires re-entering the flows.
Net present value: the number that decides
-
NPV -
10 -
ENTER -
↓ -
CPT
NPV = 69,646.56. The café adds £69,646.56 of value in today’s money over and above the 10% return the capital could earn elsewhere. Positive NPV means accept, and the size means it is not a marginal call.
Change the rate and press CPT again — the flows stay loaded, which makes a whole profile cheap:
| Discount rate | NPV | Reading |
|---|---|---|
| 0% | £171,750.00 | Raw cash surplus, ignoring time |
| 6% | £104,124.63 | Debt-only funding |
| 8% | £86,002.79 | Optimistic WACC |
| 10% | £69,646.56 | The decision |
| 12% | £54,845.19 | Cautious WACC |
| 15% | £35,167.82 | Still comfortable |
| 18% | £18,072.26 | Getting thin |
| 21.6853% | £0.00 | The IRR |
Internal rate of return: the number that gets quoted
-
IRR -
CPT
IRR = 21.6853%. Against a 10% hurdle that is nearly twelve points of headroom, which is the useful thing a rate communicates and an amount does not. But IRR assumes every interim cash flow is reinvested at 21.6853%, and it cannot distinguish this project from a £16,500 one earning the same percentage. Rank on NPV; the full argument is in NPV vs IRR.
Payback period: how long the money is at risk
Payback is the year in which cumulative cash turns positive. Discounted payback does the same on discounted cash, so it respects the cost of capital.
| Year | Cash flow | Cumulative | Discounted at 10% | Cumulative discounted |
|---|---|---|---|---|
| 0 | −£165,000.00 | −£165,000.00 | −£165,000.00 | −£165,000.00 |
| 1 | £39,500.00 | −£125,500.00 | £35,909.09 | −£129,090.91 |
| 2 | £45,500.00 | −£80,000.00 | £37,603.31 | −£91,487.60 |
| 3 | £51,500.00 | −£28,500.00 | £38,692.71 | −£52,794.89 |
| 4 | £53,750.00 | £25,250.00 | £36,711.97 | −£16,082.92 |
| 5 | £53,750.00 | £79,000.00 | £33,374.52 | £17,291.60 |
| 6 | £92,750.00 | £171,750.00 | £52,354.96 | £69,646.56 |
The undiscounted total crosses zero during year 4. Interpolating: £28,500 still outstanding at the end of year 3 divided by £53,750 of year 4 cash gives 0.53 of a year, so payback = 3.53 years. The discounted column crosses during year 5: £16,082.92 divided by £33,374.52 gives discounted payback = 4.48 years.
Nearly a year of difference between the two, on a six-year project. Plain payback flatters every project by pretending capital is free, and the longer the project the more it flatters.
Three more measures from the same entry
Net future value restates NPV at the end of the project instead of the start. Compound it forward, or
use the Professional’s NFV register:
-
NFV = NPV × 1.10⁶ = 69,646.56 × 1.771561 = 123,383.13
Profitability index is the present value of the inflows divided by the outlay — value per pound committed, which is what matters when capital is rationed rather than merely expensive.
-
PI = 234,646.56 ÷ 165,000 = 1.4221
Get the PV of the inflows by entering CF0 as zero and taking NPV, then divide by the outlay. Anything above 1.00 corresponds to a positive NPV.
Modified IRR replaces IRR’s reinvestment assumption with one you choose. Compound every inflow forward at the reinvestment rate to get a terminal value, discount the outflows at the finance rate, then find the rate linking them.
| Reinvestment rate | Terminal value | MIRR |
|---|---|---|
| 7% | £389,932.60 | 15.4120% |
| 10% | £415,690.70 | 16.6490% |
Both are well below the 21.6853% IRR. That gap is the reinvestment assumption, and it is why a quoted IRR on a long project deserves suspicion — around five percentage points of this project’s headline rate are an assumption about money it has not earned yet.
The whole dashboard
| Measure | Value | What it answers |
|---|---|---|
| NPV | £69,646.56 | How much value does this add? |
| IRR | 21.6853% | What rate does it earn? |
| MIRR | 15.4120% | What rate, with realistic reinvestment? |
| Payback | 3.53 years | How long until I have my money back? |
| Discounted payback | 4.48 years | How long, allowing for the cost of capital? |
| NFV | £123,383.13 | How much better off am I at the end? |
| Profitability index | 1.4221 | How much value per pound committed? |
Only the first of those is a decision rule. The rest are descriptions of a decision NPV has already made, and their job is to tell you how robust it is.
How wrong can the forecast be?
A single NPV is a point estimate built on numbers somebody guessed. The useful follow-up question is how much of the guess has to be wrong before the answer changes — and because the flows are still loaded, each scenario costs one re-entry rather than a rebuild.
| Revenue against plan | Year 1 cash flow | NPV at 10% | IRR |
|---|---|---|---|
| On plan | £39,500.00 | £69,646.56 | 21.6853% |
| 5% short | £32,675.00 | £36,703.91 | 16.2886% |
| 10% short | £25,850.00 | £3,761.26 | 10.6600% |
| 15% short | £19,025.00 | −£29,181.39 | 4.7359% |
A 10% revenue miss removes 94.6% of the NPV. That is operating leverage: cash costs barely move when sales fall, so almost the entire shortfall drops through to cash flow. Solving for the point where NPV reaches zero, the project tolerates a 10.57% revenue shortfall or a 14.78% cost overrun, and those two percentages say far more about the risk than any of the seven headline measures.
Why payback alone is dangerous
Two projects, both costing £100,000, cost of capital 10%.
| Quick | Patient | |
|---|---|---|
| Year 1 | £60,000 | £30,000 |
| Year 2 | £60,000 | £30,000 |
| Year 3 | £0 | £40,000 |
| Year 4 | £0 | £60,000 |
| Payback | 1.67 years | 3.00 years |
| Discounted payback | 1.92 years | 3.44 years |
| NPV at 10% | £4,132.23 | £23,099.52 |
| IRR | 13.0662% | 19.0239% |
Quick pays back in twenty months and creates £4,132.23. Patient takes three years and creates £23,099.52 — five and a half times as much. Payback prefers Quick because it stops counting at the moment of recovery, and everything Patient does in years three and four is invisible to it.
That is not an argument for ignoring payback. A business that cannot survive four years of negative cumulative cash has a real constraint, and payback measures it. It is an argument for never using payback to rank.
Four ways the series gets built wrong
- Financing costs deducted from the flows. Subtracting £9,900 of annual loan interest cuts NPV from £69,646.56 to £26,529.48. The project still passes, but a smaller one would not, and the error is invisible in the output.
- Working capital ignored. Leaving the £25,000 out of both the outlay and the release lifts NPV to £80,534.71 — a £10,888 overstatement, because £25,000 is committed today and returned in six years, and those two are not the same amount of money.
- The depreciation tax shield forgotten. Taxing EBITDA instead of EBIT costs the project its £5,000-a-year shield and drops NPV to £47,870.26. Non-cash charges have very cash consequences.
- A zero year skipped. If a project earns nothing in year 3,
C03must be entered as 0. Omitting it pulls every later flow one year forward, which raises NPV and IRR while looking entirely normal.
Which measure to lead with
| The situation | Lead with | Why |
|---|---|---|
| Accept or reject one project | NPV | The sign is the answer |
| Ranking projects with a fixed budget | NPV, then PI | PI ranks by value per pound of scarce capital |
| Cost of capital genuinely disputed | IRR and the NPV profile | Lets the reader apply their own hurdle |
| Cash-constrained business | Discounted payback | Measures exposure, which is the binding constraint |
| Long project with big interim flows | MIRR | Strips out IRR’s reinvestment assumption |
| Non-conventional flows, several sign changes | NPV only | IRR may have several values or none |
The series is the analysis; the worksheet is arithmetic. Once the flows are loaded, NPV on the cash flow worksheet covers the register-by-register detail and the IRR walkthrough covers what to do when the solver returns an error instead of a rate. If the project is financed rather than funded from cash, the debt service comes from loan payment calculations — and stays out of the cash flow series.