A consulting business plan is a description of who will pay you, for what, at what rate, for how many days, and what is left once the people doing the work are paid. Written as most templates suggest, with a market-size slide and an acquisition funnel, it describes a product company that does not exist. Written as a working-backwards brief, a dated announcement of where the firm will be in three years, the questions a sceptical reader would ask of it, and a review structured so that it can say no, it becomes a plan a founder keeps using after the bank has said yes. The revenue line underneath it is three words long: people, rate and utilisation (utilization). This guide sets out that version and maps it onto the sections a lender, grant body or partner expects.
I have written two consulting business plans that mattered. In 2013, my co-founder and I started the business that became DevOpsGroup, a services company helping other businesses build and run software. In 2014 a funding application forced us to finish the first plan; in 2020 a board request for a three year model produced the second, in a form I had borrowed from a partner. I draw on my notes, diaries and recollections for both, checking the dates and sequences as I write.
Why most templates are written for product companies
Open a business plan template and count the sections that assume you will build something once and sell it many times: market size and share, unit economics, acquisition cost against lifetime value, a funnel with conversion rates, a product roadmap. They exist because a product company’s central question is whether enough people will buy the thing at a price above what it costs to make.
A consultancy’s central question is different. Its inventory is people’s time, which cannot be stockpiled and expires at the end of every working day. Its revenue is capped by how many people it has, how many of their days it can sell, and at what rate. Its cost of sales arrives on the first of the month whether or not the work has. So the plan has to answer questions the product template never asks. When does each person become billable? What happens to the month in which a client signs late and a hire started on time? Which services are priced per day and which per outcome, and what does an overrun do to the margin? A template that skips those questions produces a plan that reads convincingly and cannot be financed.
The working-backwards brief as the plan
In 2020 our board asked for a thirty six month growth model. What I sent back was a press release, dated three years out, announcing the acquisition of our own company by a buyer I had invented. It described the company being bought: the customers it served, named by type, what it was known for, and the one number a stranger would need. Beneath it sat a set of questions and answers, and beside it a three year model.
The method is Amazon’s; Colin Bryar and Bill Carr, who spent years inside the company, have written up how it was used there. We took it from AWS, our cloud platform partner, who shared the template during a piece of joint work, and it became part of how we planned. The buyer did not exist and every figure in the release was a scenario; nothing in it should be read as history. The company was sold in 2021 for over $30m, and the imagined announcement was not how that happened.
The brief has three parts, and the plan is all three together.
The press release. One page, dated, past tense, no internal language: who benefits, what has changed and why anyone outside the company should care. For a consulting firm it names the clients you serve, the problem you are known for solving, the size and shape of the team, and the share of revenue that is contracted rather than won afresh each quarter. Every number in it is labelled an assumption, and it is never presented so often that it acquires the status of a forecast.
The FAQ. The questions a sceptical reader would ask, answered in writing. What must clients choose for this to happen? What must the firm learn to deliver that it cannot deliver today? What has to stop? Where does the money come from, and when? Which dependency do we understand least? What would make this future unattractive even if we reached it? The FAQ is where the ambition starts surviving contact, and in a consultancy the hardest questions are always about people: who is hired, when, and how long before they bill.
The review that can say no. This is the part I got wrong. At Amazon the document is read cold, in silence, by people who did not write it and do not report to the person who did, and it can be refused. In our version the author, the sponsor and the decider were the same person, and in two years of using the format it never once produced a no. The output was useful, but it was an agenda, not a filter. Two additions would have fixed that: one reader who does not report to you, who returns written questions in writing and speaks last; and a pre-mortem beside the release, imagining the plan failed and putting each plausible cause next to the assumption it attacks.
The full method, with the gate added, is in How to write a three year plan for a consulting business, a playbook in development.
The sections a lender or partner expects, and how each maps
A bank, a grant body or a prospective partner will not ask for a press release. They will ask for a business plan with familiar headings. The brief supplies every one of them; the work is mostly rearrangement.
| Section they expect | Where the brief supplies it |
|---|---|
| Executive summary | The press release, unchanged. One page, dated, saying where the firm will be and who will care. |
| The business and its services | The FAQ answers on what the firm does, for whom, and what it has chosen not to do. |
| Market and customers | The FAQ answers on what clients must choose and why; ten named conversations count for more than a market-size estimate. |
| Pricing | The rate card, the basis for each service (per day, fixed, per unit) and the FAQ answer on what happens when a job overruns. |
| Team and hiring | The FAQ answers on capability: who joins, when, and when they become productive. For a grant body, the costed jobs. |
| Financial forecasts | The model beside the brief: profit and loss, cash flow and balance sheet, linked so that a change in one moves the others. |
| Risks | The pre-mortem, with each cause set against the assumption it attacks and what you would do about it. |
| Location and case for support | The comparison you ran when choosing where to build, with the cost difference beside the reasons. Grant bodies expect this section; banks rarely do. |
| Milestones | The stages between today and the dated release, each with its own revenue, margin and headcount. |
The 2014 plan taught me why the financial section says linked. A grant adviser was introduced that year, and within a few months we had a plan detailed enough to put in front of a regional government body: profit and loss, cash flow and balance sheet forecasts that related to each other, a costed account of the proposed jobs, a comparison of Cardiff with an alternative location, and assumptions visible enough for a stranger to question. The application succeeded. The lasting value was the connected forecasts. A growing consultancy can look convincing in a revenue forecast and remain difficult to finance, because people are hired before they bill and customers pay after salaries fall due.
A realistic first-year revenue model, in words
The revenue line of a consulting business plan is a product of three things, and writing it out in words before opening a spreadsheet keeps it honest.
Start with people: how many billable people you have in each month, counting yourself, and the month each new person joins. Then billable days: how many days a year each can sell after holiday, illness, training and the internal work of running the firm, a far smaller number than the working days in a calendar. Then utilisation: what share of those days you expect to sell each month, low while a person is new, lower again in the months you are selling rather than delivering. Then rate: the day rate by grade from your rate card, at the blended rate you will actually achieve once discounts and fixed prices are counted, not the headline. Multiply people by billable days by utilisation by rate, month by month, and you have revenue.
Now test a slower version. Move the first client’s signature back a quarter. Delay the second hire. Push payment terms out to what your largest client will actually impose. Ask how much of the spending moves with the revenue, and how much is already committed. The gap between the two versions is the cash you need, and the month in which the plan is most exposed. The pipeline coverage ratio playbook, in development, turns that test into the two questions to ask of the pipeline every month afterwards.
Cost of sales is the mirror: salaries and on-costs for the billable people, contractors and their tools, arriving on the first of each month regardless. Overheads sit beneath. What is left is the plan’s answer to whether the model works, and a lender reads that line first.
What the plan is for after the money
The useful output of a business plan is a document the founders keep using, with assumptions updated as events unfold. A beautifully written application left in a folder is a poor return on the effort. Reread the press release when the plan drifts; add a question to the FAQ each time reality asks one you had not; re-run the model when a hire slips or a client signs. And the next piece of evidence that could change the decision, a client test, a delivery experiment or a cash constraint, is worth more than another round of polishing the announcement.
An imagined ending earns its place when it changes what you do before lunch tomorrow. The same is true of a business plan.