This is one of the questions owners ask most, and the answers they usually hear are rules of thumb: some percentage of revenue, whatever competitors seem to spend, whatever's left at the end of the month. Each of those gives you a number. None of them tells you whether the number is right for your business.
So instead of a rule, I want to give you a way of reasoning about the budget from the inside: what marketing actually does, what a customer is worth to you, and how to find out what your spending returns before you commit more of it.
What marketing spending buys
Let's start with what you're paying for. Marketing does two jobs, and they work on different timescales.
The first job is building memory. Every time someone who buys in your category sees your business, in a way they recognise as you, it becomes a little more likely you'll come to mind next time they need what you sell. This is slow and cumulative, and it fades if you stop. It's what research on buying behaviour, most notably the Ehrenberg-Bass Institute's, identifies as the main engine of growth: being thought of by more of the people who buy in your category.
The second job is prompting action now. An ad with an offer, a post about tomorrow's delivery slot, a reminder before a festival. This works within days, and it works best on people who already remember you.
A budget that funds only the second job tends to work well at first and then flatten out, because it keeps reaching the same people who already know you. A budget that funds only the first job can feel like money disappearing, because the results arrive slowly. A small business generally needs both, with the balance depending on how well known they already are.
Does that make sense? An everyday picture: memory works like a footpath across a field. Each person who walks it wears it a little deeper. Stop walking it for a season and the grass grows back. Marketing that builds memory is the regular walking. Marketing that prompts action is a sign at the start of the path, and the sign helps most when the path is already there.
Start from what a customer is worth
Before deciding how much to spend, work out how much you could afford to spend winning one customer.
Take a typical new customer and add up the gross profit (sales minus the direct cost of what was sold) they bring in their first year. If an average customer spends LKR 25,000 a time, buys four times a year, and your gross margin is 40%, that's LKR 40,000 of gross profit.
Now look at what it costs you to win one. If you spent LKR 60,000 on marketing last month and won 12 new customers, your customer acquisition cost (what it costs to win one new customer) was LKR 5,000.
A customer who brings LKR 40,000 of gross profit and costs LKR 5,000 to win is very affordable. You could spend considerably more per customer and still be well ahead. That's the first useful thing the arithmetic can tell you: whether your marketing is underfunded relative to what a customer is worth to you.
Two cautions. First, be honest about which customers the marketing actually brought. Some of those 12 would have come anyway, through word of mouth or habit. Asking every new customer "how did you hear about us?" is crude, and it's far better than guessing. Second, the cost of winning extra customers tends to rise as you spend more, because the easiest-to-reach people come first. So test increases in steps.
Steady reach over occasional bursts
Because memory fades, how you spend matters as much as how much. A business that spends heavily for two weeks before a season and then disappears for three months has to rebuild much of what it built. A business that stays visible at a modest level all year keeps its memory topped up, and its seasonal pushes land on people who already know it.
This matters especially in Sri Lanka, where many businesses have strong seasons: Avurudu, December, school terms, tourist seasons. Plan the year as a steady base of presence, with heavier spending timed just before the moments people buy.
Imagine two businesses that each spend LKR 600,000 on marketing in a year. One spends it in two big pushes of LKR 300,000 before its busy seasons, and nothing in between. The other spends about LKR 50,000 a month, shifting a little more into the weeks before its busy seasons. Over the year they've spent the same. The second has stayed in people's minds the whole time, and its seasonal pushes land on people who already remember it.
Spending relative to your competitors
There's one more useful lens. Research for the UK's Institute of Practitioners in Advertising, by Les Binet and Peter Field, found that brands tend to grow when their share of the advertising in their category, their "share of voice," is larger than their share of the market, and to shrink when it's smaller.
You won't have precise figures for your competitors' spending. But you can get a rough sense by watching how often their ads and posts appear, and checking Meta's Ad Library, which shows the ads any Facebook or Instagram page is currently running. If your competitors are far more visible than you and you want to grow, being less visible than them is unlikely to get you there.
Where to spend
For many Sri Lankan small businesses, the main channels are Facebook and Instagram, Google (search ads and your free Google Business Profile), WhatsApp for follow-up, marketplaces, signage and print in your area, and word of mouth.
A few principles for choosing between them:
Go where the people who buy in your category already are, and aim to reach as many of them as possible. Narrow targeting that shows ads only to people who look like your current customers misses the much larger group of occasional buyers who could become customers.
Be recognisable everywhere. The same colours, logo, name and style across every channel, so each piece of spending builds on the others. (See our guide to distinctive brand assets.)
Fix the basics before paying for reach. A complete Google Business Profile, fast replies on WhatsApp, clear prices. Paying to send people to a business that's hard to buy from wastes the spend.
Test before you scale
A sensible way to set the budget, then:
1. Work out what a customer is worth to you, and what you're paying to win one now.
2. Choose one channel and one clear message, and run it at a modest level for four to six weeks.
3. Track enquiries, sales and, as best you can, where new customers came from. Compare with the weeks before.
4. If the cost per new customer is comfortably below what a customer is worth, increase the spend in steps, watching whether the cost per customer holds.
5. Keep a steady base of presence running throughout, and add seasonal pushes on top.
6. Review every quarter, and move money towards what's working.
The budget you end up with won't be a percentage someone else picked. It will be the amount your own numbers say is worth spending, which is a much better place to stand.
The bigger point
There's a chain here worth seeing. Next year's sales are largely a result of how many people remember you when they need what you sell. That memory is a result of how many of them you've reached, recognisably, over the past months. And that reach is a result of this year's budget. So cutting marketing to protect this month's profit often shows up as lower sales much later, long after anyone connects the two.
Owners tend to start by spending when sales are slow, move on to treating marketing as the first cost to cut, and arrive, eventually, at treating it as a steady investment in being remembered, sized by what a customer is worth.
How Gridwerk helps
Setting the marketing budget, choosing channels, planning reach across the year and measuring what each rupee returns is part of our Sales & Marketing service, including media buying. If you're not sure your marketing spend is working, start with the free diagnostic. It's eighteen questions, about seven minutes, and then we call you to go through the rest.
Gridwerk diagnostic