FP&A for Startups Comes Before the First Finance Hire

FP&A for startups means turning a company’s books into three documents: a cash projection, an operating budget and a monthly comparison of actuals against both. Young firms formalise these about two years after founding, before any other management system, and a finance hire does not by itself bring them forward. The founder should write the first versions.

The best evidence comes from a field study of 78 startups by Antonio Davila and George Foster of IESE and Stanford. Financial planning was the first category of management system the companies formalised, and the timing of the operating budget tracked the CEO’s experience, the presence of venture capital and the CEO’s beliefs about planning (Davila & Foster, 2005). Companies with a financial manager adopted budgets sooner. Yet once the hire was modelled as a decision driven by those same factors, “time-to-hiring a financial manager is unrelated to operating budget adoption”.

FP&A for startups is the first system young firms formalise

Davila and Foster asked each company when it had formalised 46 systems in eight categories, where formalised “means having documented a process and/or periodically and purposefully executing the process” (Davila & Foster, IESE working paper, 2005). Financial planning, meaning cash flow projections, sales projections and the operating budget, came first on average, and it was the category most companies reached.

When 78 young firms formalised eight kinds of management system

Category	Examples in the study	Years from founding, mean	Median	Share of companies that got there
Financial planning	Cash flow projections, sales projections, operating budget	2.36	2	87.2%
Strategic planning	Investment budget, headcount plan, milestones	2.45	2	76.9%
Human resource planning	Organisational chart, job descriptions	2.58	2	80.8%
Financial evaluation	Performance against target, profitability, spending approvals	2.75	2	85.9%
Human resource evaluation	Written objectives, performance pay	2.80	2	78.2%
Product development	Project milestones and budgets	3.06	3	78.0%
Sales management	Sales targets, compensation, customer records	3.20	3	69.2%
Partnership management	Partnership plans and milestones	3.38	3	47.4%

Source: Davila & Foster (2005), IESE Working Paper 603, Tables 3 and 4. Years are the average time from founding to adopting half of the median number of systems in each category; the share is the percentage of the 78 companies that reached that point by the end of the observation period.

The mean for financial planning, 2.36 years, is about 28 months. The sample has limits that matter here. Only 19% of eligible firms agreed to take part. Of the 78, 62% were in information technology and 77% were venture-backed, and the median company was 5 years old with a peak of 113 employees. Managers recalled the adoption dates, which the authors flag as a source of bias, and the companies were sampled in 2002 and 2003. The study describes venture-backed technology firms of that era; for other startups it is a weaker guide.

The need for the documents arrives with outside money. Gavin Cassar studied a representative sample of people starting businesses and found that “the use of outside funding, level of competition, and venture scale are positively associated with the intended frequency of financial statement preparation” (Cassar, 2009). Some investors ask within the first year. Paul Graham, who co-founded Y Combinator, asks a startup “operating for more than 8 or 9 months” whether it reaches profitability on the money it has left. He wrote that “Half the founders I talk to don’t know whether they’re default alive or default dead” (Graham, 2015). That question is the cash projection, asked by someone else.

Earlier adoption also travels with better outcomes. In the published study, “faster adoption of operating budgets is associated with faster growing companies” (Davila & Foster, 2005), and CEOs who had adopted fewer management control systems had shorter tenures (Davila & Foster, 2007). The working paper concludes that “growth and the adoption of management systems reinforce each other”. These are associations in a small field sample, and growth itself forces systems on a company, so they show that early planners did well, not that planning made them do well.

A finance hire does not bring the budget forward

One prescription for FP&A for startups is to hire for it: Paul Mondollot of Aircall, quoted by Spendesk from a 2022 panel, recommends dedicated FP&A professionals “after series A or B fundraising rounds” (Spendesk). Davila and Foster tested whether the hire is what starts the budget. The raw association was positive, and in interviews managers often described a system arriving with “the hiring of a particular manager”. But the factors that predicted the hire also predicted the budget: the CEO’s experience, venture capital, the CEO’s beliefs and headcount. Once the hire was treated as a consequence of those factors, its timing no longer explained when the budget arrived. On this evidence, a CEO who believes in planning both hires the finance manager and adopts the budget, and the hire alone changes little.

Boxed, the online bulk retailer, shows what a finance hire builds once the company has one. Its 2021 prospectus says the company was founded in 2013. Its chief financial officer “joined Boxed in October 2016 as one of its initial finance hires”, worked up from senior analyst, and “established the company’s financial planning and analysis function”. He became CFO in April 2021, 54 months later, when Boxed had 238 full-time employees (Boxed S-1, 2021). The filing does not say who kept the plan in the three years before him, and one company proves no rule.

Before starting, Davila and Foster consulted five chief financial officers with start-up experience. They said that “smaller companies rely on informal mechanisms” and that control systems “only become relevant when companies reach a size of 40 to 50 employees”. That is the opinion of five people, and the founder’s documents should come well before it: a finance hire works best rebuilding documents the CEO already produces every month, with better data and a shorter close. A hire made to start FP&A pays a salary from day one and still waits on the CEO for the targets, the hiring plan and the decisions the model is meant to test.

The monthly comparison does more work than the budget

Gavin Cassar and Brian Gibson studied the accuracy of managers’ revenue forecasts. They found that “internal accounting report preparation significantly improves forecast accuracy”, with the benefit “only observed for firms with high uncertainty”. Their results gave “limited support for linkages between budget preparation and forecast accuracy” (Cassar & Gibson, 2006, published in Contemporary Accounting Research in 2008). The study covers firms in general rather than startups, but a young company plausibly sits in its high-uncertainty group, where the gains appeared.

The corporate calendar is a poor template. AFP’s 2026 survey of 332 practitioners found that the average budget takes 8.7 weeks to produce, unchanged in three years, and that only 43% of organisations use rolling forecasts (AFP, 2026). Those are corporate finance teams. FP&A for startups that copied them would spend about 61 days on a budget, about a tenth of the median wait between a seed round and a Series A (our arithmetic). Carta put that wait at 616 days in the second quarter of 2025 (Carta, 2025).

The cost of a late comparison is easy to compute. An invented 20-person company has $4,000k in the bank and plans to burn $200k a month, which lasts 20 months, but actually burns $250k a month from month 1. To still reach month 20, it needs a cut of about 21% if it is caught after one month and 50% if it waits a year (our arithmetic), assuming the cut takes effect at once. The overrun is the same $50k a month either way. Cutting half the burn of a 20-person company usually means layoffs; a fifth can often come from delaying planned hires.

Do investors read a startup’s plan at all?

The case against all this is that investors barely read the plan. William Sahlman of Harvard Business School wrote in 1997 that “Every seasoned investor knows that detailed financial projections for a new company are an act of imagination” (Sahlman, 1997). In 722 funding requests to one American venture firm, planning documents were only “weakly associated with VC funding decisions” (Kirsch, Goldfarb & Gera, 2009). Sahlman and Kirsch are describing the pitch model, which is imagination and barely moves the decision. The three documents here are written for the CEO, and the evidence that they pay comes from management outcomes. In a random sample of 223 Swedish ventures started in 1998, planning reduced the likelihood of disbanding and accelerated product development (Delmar & Shane, 2003).

Before the first finance hire, then, FP&A for startups is the CEO’s job: a cash projection, an operating budget by team and by hire, and a monthly comparison of actuals against both, kept from the month outside money arrives. A company that waits for the hire makes its first large decisions from the bank balance, which reports only after the money has gone. The order itself comes from 78 venture-backed technology firms sampled in 2002 and 2003, and a bootstrapped company with no investor asking for the cash projection may reach it later.

Notes and sources

• Association for Financial Professionals (2026). AFP Survey Reveals Structured Scenario Planning Separates Top-Performing Corporate Finance Teams (2026 AFP FP&A Benchmarking Survey, 332 respondents), press release, 20 January.
• Boxed, Inc. (2021). Form S-1 registration statement, filed 22 December 2021.
• Carta (2025). Series A Funding Slides in Q2 2025, by Kevin Dowd, 19 September.
• Cassar, G. (2009). Financial statement and projection preparation in start-up ventures. The Accounting Review 84(1), 27–51.
• Cassar, G. and Gibson, B. (2008). Budgets, internal reports, and manager forecast accuracy. Contemporary Accounting Research 25(3), 707–738; SSRN working paper 939332 (2006).
• Davila, A. and Foster, G. (2005). Management accounting systems adoption decisions: evidence and performance implications from early-stage/startup companies. The Accounting Review 80(4), 1039–1068.
• Davila, A. and Foster, G. (2005). Startup firms’ growth, management control systems adoption and performance. IESE Business School Working Paper No. 603, July.
• Davila, A. and Foster, G. (2007). Management control systems in early-stage startup companies. The Accounting Review 82(4), 907–937.
• Delmar, F. and Shane, S. (2003). Does business planning facilitate the development of new ventures? Strategic Management Journal 24(12), 1165–1185.
• Graham, P. (2015). Default Alive or Default Dead? paulgraham.com, October.
• Kirsch, D., Goldfarb, B. and Gera, A. (2009). Form or substance: the role of business plans in venture capital decision making. Strategic Management Journal 30(5), 487–515.
• Sahlman, W. A. (1997). How to write a great business plan. Harvard Business Review, July–August.
• Spendesk (undated). FP&A best practices for startups, reporting a 2022 CFO Connect Summit panel.

The figures come from the studies, surveys, filings and essays listed above. Davila and Foster’s tables are from their IESE working paper, and several other journal findings are cited from the articles’ published abstracts. Any figure without a cited source is worked out from the studies and surveys above, and the 20-person company is hypothetical.