A discounted cash flow puts a price in today's money on cash you haven't received yet. You forecast what your business will generate over the next few years, then reduce each future year by a rate that stands in for the return you'd need to justify waiting for it. Add the reduced figures together and you have a present value.

The maths is simple enough to do on paper, and this page works through every step of it on a company that makes £120,000 a year. What it can't tell you is whether to believe the answer. That turns out to be the more useful thing to know, because moving two assumptions by a couple of points each takes the same company from £787,000 to £1.42m.

‍

The discounted cash flow formula

For a single future amount, the formula is:

PV = CF ÷ (1 + r)ⁿ

• PV is the present value: what that money is worth to you today.
• CF is the cash flow you expect in that year.
• r is the discount rate, written as a decimal, so 15% is 0.15.
• n is how many years away it is.

Say someone owes you £1,000 in three years and you use a discount rate of 15%. Divide £1,000 by 1.15 cubed, which is 1.520875, and you get £657.52. That is what the promise is worth now.

For a run of years you do the same sum on each one and add the results. Nothing more complicated happens, however large the spreadsheet gets.

‍

Forecasting the cash flows

The top of the formula is cash, not profit. What you want is the cash the business is left with after it has paid its costs and its tax and spent what it needs to spend to keep running, which is the cash the business actually generates rather than the figure at the bottom of your profit and loss account.

Most forecasts run three to five years. Beyond five, you are not forecasting so much as extending a line, and the method has a separate way of handling everything after the forecast ends.

Build the forecast from what your accounts already record rather than from an ambition. Three years of history tell you what your business does in a normal year, a good one, and a bad one, and the range between them is worth more than a single confident line.

‍

Choosing a discount rate when you're a small company

Every guide to this subject tells you to use your weighted average cost of capital. For a private limited company that advice is unusable. A weighted average cost of capital needs a cost of equity, a cost of equity needs a share price and a beta, and your company has neither. If you follow the instruction literally you end up inventing a number and then treating it as though you had looked it up.

Here is something you can actually source. The Bank of England publishes what new borrowing costs each month, split by the size of the borrower. In August 2026 the effective interest rate on new loans to small and medium-sized businesses was 6.63%, against 5.57% for UK private companies of all sizes, with Bank Rate at 3.75% since 18 December 2025. Being small costs you about a point, every month, in public data.

That 6.63% is a floor rather than an answer. Borrowed money is the cheapest money in your business, because a lender gets paid before you do and can ask a director for a personal guarantee. The money you put in yourself carries the risk the lender handed back, so it has to earn more. How much more is a judgement, and nobody can look it up for you.

If you want a risk-free anchor to build from, use a real one and date it: the 10-year UK gilt yielded 5.43% on 13 October 2026. It is a long-dated government rate rather than a business one, so it won't carry the whole argument, but it beats borrowing a bond yield from an American textbook.

The example below uses 15%. That is an illustration and not a benchmark, and the next section shows how much rests on the choice.

‍

A worked example, every step shown

Take a limited company that generated £120,000 of free cash flow last year and expects that to grow 5% a year. These figures are an illustration rather than a typical business, but the shape of the calculation holds on any size of company. The discount rate is 15% and the forecast runs five years.

Each discount factor below is one divided by 1.15 to the power of the year, printed to four places so you can check the multiplication yourself.

So five years of growing cash flow, worth £663,076 if you simply added the five years up, comes to £438,555 once you have accounted for the wait. The discount has cost the fifth year half its face value.

Then there is everything after year five, because the company does not stop trading. That part is the terminal value, and on these assumptions it comes to £1,144,446 at the end of year five, which is £568,992 in today's money once discounted back. The section below sets out where that figure comes from.

Add the two halves together and the business is worth £1,007,547.

‍

How much the answer moves when you change the assumptions

That £1,007,547 looks precise and it is nothing of the kind. Two of the numbers behind it are assumptions: the 15% discount rate, and the 2% long-term growth rate used in the terminal value. Neither can be proved, and both move the answer a long way.

Here is the same company, with the same forecast, valued nine times. Nothing changes except those two assumptions, and only within a range any competent adviser could defend.

The top corner is 1.8 times the bottom corner. The spread is £633,000 on a business that generates £120,000 a year, so the uncertainty in the method is worth more than five years of the company's entire cash flow.

Not one cell in that grid is a mistake. Each one is correctly calculated from assumptions a reasonable person could argue for, which is the whole difficulty. A discounted cash flow does not produce a number, it produces whichever number you were already inclined to believe. If you are handed one, the first thing to ask is what rate and what growth assumption were used, and the second is what the answer looks like two points either side.

This is also why a single figure from a discounted cash flow should never be quoted on its own. The grid is the result.

‍

Terminal value, and why most of the answer is in it

Your forecast stops at year five. The company doesn't. Terminal value is what you put in for everything after that, and the usual method treats the final year's cash flow as continuing forever at a steady growth rate:

Terminal value = final year cash flow × (1 + g) ÷ (r − g)

In the example, that is £145,861 grown by 2% to £148,778, divided by 0.15 minus 0.02. The answer is £1,144,446 at the end of year five, worth £568,992 today.

Two constraints matter here, and both are absolute.

Growth must be smaller than the discount rate. If g is larger the division turns negative and the formula returns nonsense, and if the two are close it returns a very large number for the same reason. The arithmetic breaks down long before the economics do.

Growth must also stay below long-run inflation. The Bank of England's inflation target is 2%, which is why 2% is the conventional ceiling. A perpetual rate above it assumes your company outgrows the whole economy for ever, and no buyer and no tax inspector will accept that.

Now the part worth sitting with. In the example, £568,992 of the £1,007,547 total, or 56% of the answer, is the terminal value: a single figure standing in for a period nobody forecast at all. Across the nine cells in the grid above its share ranges from 48% to 67%, rising as the discount rate falls, because a lower rate preserves more of the value of distant years.

So most of what a discounted cash flow tells you comes from the part of the calculation you did least work on. The careful five-year forecast, where all the effort goes, accounts for less than half.

‍

When a discounted cash flow is the right tool

None of this makes the method useless. It is the right tool in three situations, and they have something in common: the forecast is short, or it isn't really a forecast.

Income that is contracted rather than hoped for. A signed multi-year contract, a long commercial lease, a maintenance agreement with a fixed term. You are discounting money somebody already owes you on terms already agreed, so the top of the formula is close to a fact. This is the case the method was built for.

A single asset with a known life and a known price. Whether to buy a £50,000 machine that should last 10 years is exactly the question a discounted cash flow answers well. The forecast is short, you are estimating cost savings rather than growth, there is no terminal value to swamp the result, and you can set the answer against what the money to buy it would cost you. If the present value of the savings doesn't clear the price plus the cost of the finance, the project doesn't pay.

A sale with no open market. In a management buyout there is no competitive bidding and often no comparable transaction to look at, so a forecast-based valuation may be the only tool available. It still carries every weakness above. It is simply better than nothing, and both sides should see the sensitivity grid rather than the single figure.

In all three cases the method values what the business will earn, which is a different question from what the business has tied up in it. For an asset-heavy company the second answer can be the larger of the two.

‍

When it will mislead you, and what to use instead

If you run a profitable trading company with a few years of uneven history and you want to know what it is worth, a discounted cash flow is the wrong instrument. The reason is in the grid: the method asks you for a long-run growth rate and a risk-adjusted discount rate, and a small company with a short, lumpy record cannot supply either with any confidence. Three good years and one bad one give you no defensible view of what happens in perpetuity, and that is where most of the answer comes from.

Practitioners know this, and so does HMRC. Its Shares and Assets Valuation manual states the order of preference plainly: 'there is often no better indication of value than actual transactions involving the same or similar asset', and a valuation technique is for use 'in the absence of an active market and any recent transactions'. The manual's own simplified examples for small companies bear that out. Not one of them uses a discounted cash flow. The example for an established trading company takes a quoted company's price-earnings ratio of 12.03, discounts it heavily to reflect the gap between a small unquoted holding and a listed share, and applies the resulting multiple of 4.5 to maintainable post-tax profits.

Which is what a buyer will do to you as well. If you want to know what your company would fetch, the method is a multiple of its adjusted earnings, and the work that moves the price is the adjustments to the profit figure rather than anything in this formula.

The example above makes the point against itself. That £1,007,547 is 8.4 times the £120,000 the company generated in year one. The highest multiple Dealsuite publishes for any size of UK company is 8.4 times, and the UK200Group's median for small businesses was 4.5 times in November 2025. Free cash flow and adjusted earnings are not the same measure, so treat that as a sense check rather than a like-for-like comparison. Even so, a reasonable-looking discounted cash flow has landed at the top of what the market pays, and roughly double the middle of it. The formula did nothing wrong. It answered a different question from the one a buyer asks.

So use a discounted cash flow to understand what your own forecast implies, and to test a price somebody else has put to you. Don't use it to set one.

‍

Discounted cash flow, net present value and undiscounted totals

Three terms get used as though they were interchangeable, and they aren't.

A discounted cash flow is the present value of the money coming in. In the worked example, £1,007,547.

Net present value is that figure after you take off what you paid to get it. If buying the business cost £900,000, the net present value is £107,547, and the sign is the point: positive means the deal adds value at your chosen discount rate, negative means it destroys it. This is why an investment appraisal usually ends in a net present value and a valuation ends in a discounted cash flow.

An undiscounted cash flow is the raw total, with no adjustment for time. The five years in the example come to £663,076 undiscounted and £438,555 discounted, and the £224,521 difference is the cost of waiting. Counting up future cash until it covers what you paid gives you a payback period, which is worth knowing and is not a discounted cash flow. If a calculation has no discount rate in it anywhere, nothing is being discounted.

‍

Sources

• Bank of England, Money and Credit, August 2026 (published 29 September 2026): effective interest rates on new loans to small and medium-sized businesses (6.63%) and to private non-financial corporations (5.57%).
• Bank of England, official Bank Rate history: 3.75% from 18 December 2025.
• Bank of England interactive statistical database, series IUDMNZC: 10-year nominal gilt spot yield, 5.43% on 13 October 2026.
• Bank of England monetary policy remit: the 2% inflation target.
• HMRC Shares and Assets Valuation Manual, SVM110050.
• UK200Group SME Valuation Index (November 2025) and the Dealsuite M&A Monitor (February 2026), for the earnings multiples used as a cross-check.

All figures verified on 22 October 2026.