Business Plan: Template and Writing Tips for 2026

Quick answer. A convincing business plan runs to 15-25 pages: a narrative part (project, market, team) and a three-year financial forecast. For a bank, the four expected tables are the initial financing plan, the projected income statement, the monthly cash flow plan and the three-year financing plan.

A business plan is not an administrative formality you write once for the bank and then forget. It is the tool that turns an intuition into a fundable, manageable project. It serves three distinct goals: testing the economic viability of the idea, convincing a lender (bank, investor, honour-loan committee), and keeping a numbers-based course once the business is running.

In the start-up files we support, the most common sticking point is not a lack of ambition but a gap between the story and the figures. A founder describes a vast market, then projects modest revenue without explaining how to bridge the two. A financier reads that consistency before looking at the amount requested.

One structural shift matters: the roll-out of electronic invoicing between businesses, with mandatory reception for all French companies from 1 September 2026. Your forecast must now factor in the cost of a digitisation platform and the time needed to adapt processes. It is one of the topics we cover in our support for start-ups.

A complete business plan alternates a qualitative part (the project narrative) and a quantitative part (the financial forecast). The two must echo each other: every numerical assumption should be justified in the text. For the detailed narrative outline, see our dedicated article on the structure of a business plan.

Here are the sections we consider essential and the most effective reading order.

The forecast is the heart of the file. According to Bpifrance Création, it rests on four tables that interlock. Here is the method we apply, in order.

1. Build the initial financing plan. List your start-up needs (investments, set-up costs, working capital requirement) against your resources (equity, loans, grants). The two columns must balance to the cent.
2. Build the projected income statement. Start from revenue based on volume and price assumptions, deduct variable costs then fixed costs, to reach the result before and after tax.
3. Roll out the cash flow plan month by month. Shift inflows and outflows according to real payment terms. This is where the cash tensions hidden by the income statement appear. Our method is detailed in the 13-week cash flow plan.
4. Draw up the three-year financing plan. Project needs and resources over three financial years to check that the financial structure holds over time.
5. Test profitability. Compute the break-even point to know from which revenue level the project covers its costs.

One prudence rule shared by Bpifrance Création is worth remembering: in the three-year financing plan, resources should exceed needs by at least 15 to 20 % of the first year's self-financing capacity. That safety margin absorbs the start-up delays that are almost always underestimated.

Excel remains the most flexible tool for a numbers-based forecast, provided it is structured cleanly. The classic trap is entering hard-coded amounts in scattered cells: at the first revised assumption, the whole structure becomes wrong.

Our template follows a simple logic, reproducible in any spreadsheet.

• A single Assumptions sheet: prices, volumes, margin rates, payment terms, social charge rates. Everything else refers to it by formula.
• One sheet per table (income, cash flow, financing), fed only by references to the assumptions.
• Scenarios: a central case, a conservative case, an optimistic case, obtained by changing three or four cells.
• A control tab checking the balances (financing plan balanced, cash never negative without funding).

Start-ups that steer by data go further and connect this model to a visualisation tool to track the gap between planned and actual. We equip these dashboards, notably through Power BI and analytics solutions such as Finthesis.

Length depends on the audience, not on a universal standard. A bank loan file for a retail activity reads in 15 to 20 pages. A fundraising file usually comes with a presentation deck of 12 to 15 slides, the financial detail living in an annex file.

A business plan that is too long dilutes the message; too short, it looks rushed. Aim for density: each page should carry a decision-relevant piece of information.

Micro-business and service activity. You remain under the VAT exemption (franchise en base) as long as your revenue stays below 37,500 euros for services (85,000 euros for trade), under article 293 B of the French Tax Code. Your invoices then carry the wording "TVA non applicable, article 293 B du CGI". Your forecast must anticipate the switch to VAT, which changes your net prices and your cash position.

Company subject to corporate income tax. If your revenue stays at or below 10 million euros and the capital is at least 75 % held by individuals, you benefit from the reduced corporate tax rate of 15 % on the fraction of profit at or below 42,500 euros (Tax Code, article 219, I-b). The choice between salary and dividends, the latter bearing a flat tax of 31.4 % since the 2026 social security finance act, is prepared from the forecast stage. This reasoning depends on your choice of legal structure.

Career change and transition from employment. A project carried by a soon-to-leave employee must factor in the loss of income and the switch timetable. We explain this in our from employee to founder journey.

Highly seasonal sector project. A restaurant, for example, requires a very granular monthly forecast, as shown in our analysis of a restaurant forecast.

The underestimated risk: working capital requirement. Most young-company failures stem not from a lack of profitability but from a cash shortage at start-up. Inventory, customer payment terms, VAT timing: this need must appear in the initial financing plan, never as an adjustment variable.

What a financier looks at. An honour-loan committee or a banker tests three things: the consistency between market and revenue, the share of personal equity, and the repayment capacity. As an indication, the Initiative France honour loan generally ranges from 3,000 to 50,000 euros (average around 10,000 euros) and the Réseau Entreprendre one from 15,000 to 50,000 euros (average around 29,000 euros), with a significant banking leverage effect.

The rushed market study. A market claimed without a quantified source weakens the whole file. Build it seriously, as we detail in how to run your own market study.

Our reading is simple: a business plan is judged not by its graphic beauty but by its robustness under challenge. When we review a file, we systematically attack three assumptions: the pace of revenue ramp-up, the actual collection delay, and the working capital requirement. If the model withstands these three pressures, it will hold in front of a banker.

Recently, a founder of a young digital-services company asked us to help with a financing file already turned down by two banks. The narrative was excellent, but the forecast booked every sale in the month of invoicing, ignoring the payment terms of its large-account clients. By rebuilding the cash flow plan on real terms, we revealed a cash gap in the fifth month, then adjusted the funding need accordingly. The corrected file was approved.

As chartered accountants and statutory auditors registered with the French professional body (Ordre des experts-comptables), we see the forecast as a commitment, not a seduction exercise. Credibility lies in the realism of the assumptions and their traceability. That is also what makes a business plan reusable as a steering tool, rather than a dead document once funding is secured.

Hayot Expertise tip. Build your assumptions first, validate them with your market study, then run the numbers. Date every version of your forecast and keep the history: a financier appreciates seeing thinking that evolves. Once the business is live, compare actual to plan every month and correct the assumptions, not just the figures.