How Much Should a Small Business Spend on Digital Marketing?

Should you budget as a percentage of revenue or a fixed sum?

Percentage of revenue is the better default, because it scales automatically and forces an honest conversation about growth ambition. It has one weakness: it shrinks exactly when you might need to spend more, which is during a slowdown.

A fixed sum works better for businesses with lumpy or seasonal revenue, or for a specific campaign with a defined goal. The risk is that a fixed number becomes a ceiling nobody revisits, so a business that tripled in size is still spending last year's figure. Whichever you choose, set the number deliberately each year rather than rolling forward whatever you spent last time.

How should the budget be split?

Split between brand and demand, and expect the ratio to shift as you grow. Early on, most spend should be demand capture — activity that produces enquiries this quarter. As the business matures, a larger share goes to brand, because branded search converts dramatically better and is cheaper to defend than to buy.

A practical starting split is 70% demand, 30% brand, moving toward 50/50 as you scale.

Split by function before you split by channel. A workable breakdown for a growth-stage business: roughly half on demand capture, a fifth on content and SEO that compounds, a fifth on brand and awareness, and a tenth on tools, testing and measurement. Measurement is the line most often cut, and the one that makes every other line accountable.

How much should you spend on brand versus demand?

Early on, most spend should be demand capture — activity that produces enquiries this quarter. That is not a failure of ambition; it is how a young business funds itself. Brand spending pays back over years, and a business that cannot survive the next two quarters cannot collect on a five-year investment.

As you grow, the ratio shifts. Branded search converts far better than cold search and is cheaper to defend than to buy, so a larger brand share lowers your blended acquisition cost over time. A practical path is 80/20 in year one, 70/30 by year two, and closer to 60/40 once you have real market presence.

Which numbers tell you if you are spending right?

Three metrics, tracked together. Customer acquisition cost tells you what you pay to win a customer. Customer lifetime value tells you what that customer is worth. The ratio between them tells you whether you can afford to spend more.

A healthy benchmark is lifetime value at least three times acquisition cost, with payback inside twelve months. Below three times, growth spending is fragile. Above five times, you are probably underinvesting in growth.

Add two supporting numbers. Payback period — how many months of gross profit it takes to recover the acquisition cost — should sit inside twelve months for most businesses and inside six where cash flow is tight. And marketing-attributed revenue share tells you whether your reporting is credible: if every sale is claimed by three channels, your attribution is broken.

Where do small businesses waste the most?

The largest waste is not in the channel — it is in the destination. Sending paid traffic to a page that doesn't clearly state the offer, price band and next step wastes budget at a rate no bid optimisation can recover.

The second largest is measurement: no conversion tracking, or tracking that counts every form fill as equal. Both make it impossible to tell a good month from a lucky one.

The third is inconsistency. A business spending $4,000 in March, nothing in April and $1,500 in May gets worse results than one spending $1,800 every month, because most channels need continuity to compound and every pause resets the learning. Budget for twelve months, not for a quarter you feel optimistic about.

What does the first $2,000 a month actually buy?

Not much, and that is worth saying plainly. At $2,000 a month you can run one channel competently or two badly. The sensible allocation is almost entirely demand capture on the platform where your customers already search, plus a small reserve for measurement and one landing page to send traffic to.

What you should not do at this level is buy brand awareness, sponsor a podcast, or split the budget across four platforms to 'test' them. Pick the one channel with the clearest intent signal and get good at it.

• One paid channel, tightly themed, with conversion tracking working
• A dedicated landing page per offer, not your homepage
• Roughly $300–$500 a month held back for testing and measurement
• A monthly review of cost per qualified lead, not impressions

How do you know when to spend more?

Increase when your lifetime value is at least three times your acquisition cost and payback sits inside twelve months. Those two conditions mean the unit economics work and the constraint is capacity rather than profitability — which is exactly when more budget produces more profit rather than more loss.

Cut when a channel's cost per qualified lead has risen for two consecutive months without a change in what you are selling. That pattern usually means audience saturation, creative fatigue or a competitor outbidding you, and none of those are fixed by spending more. Fix the input, then restore the budget.

How do you build a marketing budget from scratch?

Start from the customer, not the channel. Work out what a customer is worth over their lifetime, what gross margin you keep, and therefore what you can afford to pay to acquire one. That number is the budget's real constraint, and it is more useful than any percentage of revenue.

Then work backwards. To acquire fifty customers at a $400 acquisition cost you need $20,000 and a credible plan for which channel produces them. If the plan cannot produce that volume at that cost, you have a strategy problem rather than a budgeting problem.

• Calculate lifetime value and gross margin per customer
• Set a maximum acquisition cost at roughly one-third of lifetime value
• Estimate the volume each channel can realistically produce
• Multiply volume by cost to get the budget, then sanity-check it against revenue
• Reserve around 10% for measurement and testing