How to set a marketing budget when you're starting from zero
The standard advice — spend some percentage of revenue on marketing — is useless when you have barely any revenue and no idea which channel works. Here is a more practical way to arrive at a first number.
Start from what a customer is worth
Before you can decide what to spend, you need two figures. You can estimate both roughly and refine later.
- Average order value, or average first project value.
- How many times a typical customer buys before they drift away.
Multiply them, take off your delivery costs, and you have a rough lifetime profit per customer. If a customer is worth £400 in profit over two years, you now know that spending £80 to win one is a good trade and spending £600 is not.
Work backwards to a monthly number
Decide how many new customers a month you actually want, and how much of their lifetime value you are willing to pay to acquire them. A common starting point is 20–30% — enough to be competitive, leaving margin for the work.
Ten customers a month, at £80 each, is an £800 monthly budget. That is a real number derived from your business, not a percentage borrowed from someone else's.
Budget for the test, not the outcome
Your first three months are not a campaign, they are an experiment. You are buying information about which channel and which message work, and you should expect the cost per customer to be worst at the start and improve as you learn.
Set aside a test budget you can genuinely afford to lose, and give it long enough to produce a signal. Turning a channel off after ten days tells you nothing except that ten days isn't long enough.
Pick one channel first
Splitting a small budget across search, social, and email guarantees that none of them gets enough data to be conclusive. Choose the one that best matches how your customers buy:
- People actively searching for what you sell? Paid search, and SEO in the background.
- People who don't know they need you yet? Social, where you can interrupt with something interesting.
- Selling something considered and expensive? Content and email, because the decision takes months.
Keep some powder dry for what works
Don't commit every pound up front. When a channel starts producing customers below your target cost, you want budget available to feed it. Holding back a quarter of the budget for this is more useful than spreading it thin at the outset.
Track the number that matters
Impressions and clicks are inputs. The figure to watch is cost per customer, not cost per click, and you cannot see it unless enquiries are tracked through to sales. Even a spreadsheet where you note how each customer found you beats guessing.
When to increase it
When a channel reliably produces customers below your target acquisition cost, and you have capacity to serve more. Not before. Scaling a channel that loses money just loses money faster.