The bank wants a business plan before it will look at the £45,000 facility. You spend three evenings producing thirty-eight pages with a market overview, a competitor matrix, a section on your values and a five-year projection that ends at £2.1m of turnover.
The lending manager opens it, goes to the cash flow, looks at the monthly figures for the next twelve months, looks at what security is available, and reads two paragraphs. The rest of it was written for you, not for her, which is fine as long as you know that is what you were doing.
The lender reads four things
Whoever is deciding whether to advance money is answering four questions, in this order. Can this business service the debt out of trading cash. What happens if it cannot. Does this person know their own numbers. Is there anything here that contradicts what I can see elsewhere.
That means the parts of the plan that carry weight are the monthly cash flow for the next twelve to eighteen months, the assumptions behind it, the historic figures if you have any, and a short honest statement of what could go wrong and what you would do about it. A market size taken from a press release adds nothing. A competitor matrix where you win on every row actively costs you credibility.
It also means consistency is scored. If the plan says thirty-day payment terms and your last set of filed accounts at Companies House shows debtor days of seventy-one, you have a problem you will be asked about. Lenders will look at your filed accounts, your personal credit file, and, for a Start Up Loan through the British Business Bank, a personal survival budget showing you can live on what you say you will draw.
Build the forecast from the bottom
There are two ways to produce a revenue line. One is to take a market size, apply a share, and write down the answer. The other is to count the things you actually sell.
The second one is the only one worth doing. A café forecasts covers per day by day of week, average spend, and the seasonality of its street. A plumber forecasts chargeable days a month after holiday, sickness, quoting time and the days spent going back to finish jobs, times a day rate, times a realistic conversion rate on quotes issued. A shop forecasts footfall, conversion and basket. In every case you end up with a small number of physical quantities you can argue about, rather than a percentage of a market that nobody can check.
Do the same on costs. Rent, rates, insurance and your own drawings are known. Staff are a headcount times a wage plus employer National Insurance, the pension contribution under automatic enrolment, and holiday cover. Stock is a percentage of sales, which you should be able to state for your trade and defend.
Then lay it out monthly, on a cash basis, with VAT in it. This is where most forecasts fall apart, because profit and cash are not the same thing and the plan that shows a profitable first year often shows an empty bank account in month seven. Put the VAT quarters in as payments on the dates they fall. Put Corporation Tax nine months and a day after the year end. Put the annual insurance renewal in the month it renews rather than spread evenly. Put the customers on the payment terms they will actually take, not the terms on your invoice.
A thirteen-week version of the same thing, updated weekly, is the single most useful management document a small British business can have, and it comes free once you have built the annual one properly.
The assumptions sheet is the plan
Every forecast rests on maybe fifteen numbers. Average spend. Days worked. Conversion rate. Gross margin. Debtor days. Wage inflation. Rent. Put all of them on one sheet, with a source for each and a note of how confident you are, and drive the whole model off that sheet so that changing a cell changes the forecast.
Doing that gives you three things. You can answer the lender's questions instantly, because every question is really about one of those cells. You can run sensitivities: what happens at 80% of forecast revenue, what happens if the rent goes up 15% at review, what happens if your biggest customer takes ninety days. And you can go back in six months and see which assumptions were wrong, which is how you get better at forecasting.
A lender who sees an assumptions sheet with a downside case already on it is being told that you have thought about failure. That is worth more than a confident projection, because confident projections are the only kind they ever receive.
A plan you update is worth more than one you file
The plan written for the bank is a snapshot for an audience. The plan that runs the business is a page you keep.
One page is enough for most owner-managed businesses: what the business does and for whom, the three things that have to happen this year, the numbers you are steering by, and the cash position. Everything else is supporting material. Review it quarterly against what actually happened, and re-forecast the remaining months rather than clinging to a budget that was superseded in March.
Some of the plan does need to be written down properly and kept regardless of any lender. If there is more than one shareholder, a shareholders' agreement matters far more than the plan does, and covers what happens when one of you wants out, dies, or stops turning up. If the business depends on one person, work out what happens if that person is ill for three months, and price key person cover. Neither of those is a growth question. Both of them are the questions that actually end businesses.
Write the long version once if a lender demands it. Then keep the page.
Where this shows up in the game
Business Baron does not ask you to write a plan, it asks you to live with the consequences of not having one. The bank manager is a named individual who forms a view of whether you know your own figures, and that view sets your margin at the annual review and decides whether the facility is renewed. The thirteen-week cash flow is on screen because it is the number that decides whether you survive February. The free game is the real game rather than a slice of it, so the first plan you get wrong costs you nothing.