You don't need a R100,000 monthly budget to make Google Ads work. You need a budget built the right way round, from your margin backwards, not from what a rep at Google tells you to spend.
Most SME owners get this backwards. They pick a number that feels affordable, split it across a few campaigns, and let Smart Bidding figure out the rest. Eighteen months later they're still asking why cost per lead keeps climbing while sales stay flat.
TL;DR
Building a Google Ads budget for SMEs in 2026 starts with your break-even cost per acquisition, not your bank balance. Work out what you can afford to pay for a customer, multiply that by the conversions you need monthly, and only then decide your daily spend. A R15,000/month account with a R450 target CPA and a 10% conversion rate needs roughly 33 clicks a day at R45 average CPC, which is a workable starting budget for most Johannesburg and Cape Town SMEs. Businesses without conversion tracking set up properly should fix that before touching budget at all: verdict is fix tracking first, size budget second.
Why this matters
Ad spend without a floor is how SMEs burn R20,000 a month and can't say what it bought them. Google's Smart Bidding algorithms need accurate conversion signals to spend your money well. Feed it broken tracking (duplicate conversions, phone calls not captured, form submissions counted twice) and it will happily scale spend towards junk.
The SME accounts reviewed by any decent Google Ads agency for small businesses tend to show the same three faults: no ROAS or CPA floor set anywhere in the account, budgets increased because a rep suggested it rather than because the data supported it, and campaign structures so broad that Smart Bidding can't tell a R50,000 lead from a tyre-kicker.
Get the budget model right in 2026 and you avoid all three. Get it wrong and no amount of creative or keyword research will save the account.
What you'll need
- Your average order value or average deal value (last 3-6 months of actual sales data, not a guess)
- Your gross margin percentage per sale
- A conversion tracking setup that fires correctly for leads, calls, or purchases (test this before anything else)
- Historical account data if you've run ads before (impressions, clicks, cost, conversions for at least 90 days)
- Your sales team's close rate from lead to paying customer, if you sell B2B or services
- 45-60 minutes with a calculator and no interruptions
If you've never run Google Ads before, you'll need an estimated cost per click for your category. Search Ads benchmarks in South Africa for 2026 vary hugely by sector, from around R8-R15 for local service keywords up to R60+ for competitive B2B and legal terms.
The steps
1. Calculate your maximum allowable cost per acquisition
This number anchors everything else. Take your average gross profit per sale and decide what percentage of that profit you're willing to spend acquiring the customer.
Most SMEs land somewhere between 15% and 30% of gross profit as their acquisition cost ceiling, depending on whether the customer is a one-off purchase or has repeat value. A gym with a R650 average monthly membership and 70% margin has roughly R455 gross profit to work with. Spending R200 to acquire that member (30% of margin) still leaves headroom before the first renewal.
Common mistake: using revenue instead of gross profit. A R2,000 product with 20% margin only gives you R400 to play with, not R2,000.
2. Work out your realistic conversion rate
Pull your website's actual conversion rate from Google Analytics or your CRM if you've run any traffic before. If not, use category benchmarks: service businesses in South Africa typically convert landing page traffic at 2-5%, ecommerce at 1-3%, and high-intent local searches (like "plumber near me") can hit 8-12%.
Don't use industry-wide averages pulled from US case studies. A Sandton law firm's conversion rate on "divorce attorney Johannesburg" looks nothing like a Takealot seller's rate on a generic product term.
Expected outcome: a number between 1% and 12% that tells you how many clicks equal one conversion.
3. Reverse-engineer your click volume
Divide your maximum allowable CPA by your expected cost per click, then multiply by your conversion rate, to get the number of clicks you need for one conversion.
Say your CPA ceiling is R450, average CPC in your category is R45, and conversion rate is 10%. You need 10 clicks to get 1 conversion at that rate, costing R450, which exactly matches your ceiling. That's your break-even point, not your target. Build in margin: aim for the model to produce conversions at 70-80% of your ceiling so there's room for underperforming keywords.
4. Set your monthly conversion target from sales capacity, not ad spend
Ask your sales or ops team a blunt question: how many new customers can we actually service well this month? An SME that can only onboard 15 new clients shouldn't budget for 40 leads. That's wasted spend and an annoyed sales team chasing leads that go cold.
Multiply your target conversions by your break-even CPA to get your monthly budget. Fifteen conversions at R450 CPA is a R6,750 monthly floor, before you add buffer for testing new keywords or ad variations.
5. Add a testing buffer of 15-20%
No campaign performs at target from day one. Google needs data (Google recommends a minimum of 15-30 conversions per month per campaign before Smart Bidding stabilises, a benchmark that's held steady through 2026). Add 15-20% on top of your core budget purely for testing new ad copy, audiences, and keyword expansion without cannibalising the budget that's already converting.
A R6,750 core budget becomes roughly R7,900-R8,100 with testing buffer included. That's your real starting monthly figure.
6. Split budget across campaign types with intent in mind
Don't spread one budget evenly across Search, Performance Max, and Display. Search campaigns targeting high-intent keywords should get 60-70% of budget for most SMEs early on. Performance Max can take 20-30% once you have at least 30 conversions of historical data to feed it. Display and pure awareness plays should get whatever's left, usually under 10%, unless brand awareness is an explicit, separately measured goal.
Accounts managed without this split often see budget quietly drift into Performance Max because it's the default recommendation inside the platform, not because it's earning its share.
7. Set a daily budget cap and a ROAS or CPA floor inside the account
Divide your monthly figure by 30.4 to get a daily budget, then set that as a hard cap in the campaign settings, not a suggestion. Set a target CPA or target ROAS in the bidding strategy that matches the number from Step 1, not a number Smart Bidding suggests based on "maximising conversions."
Common mistake: leaving bid strategy on "Maximise Conversions" with no target. Google will spend your full daily budget every day regardless of whether those conversions are profitable.
Troubleshooting
Cost per click is double what you budgeted for. Check Quality Score first. A Quality Score under 5 on your core keywords usually means landing page relevance is off, and you're paying a penalty on every click.
Conversions look healthy but sales haven't moved. This is almost always a tracking problem, not a Google Ads problem. Test every form and call tracking number manually before assuming the campaign is broken.
Budget runs out by midday. Your daily cap is fine but your bid strategy is too aggressive for the traffic volume. Lower the target CPA slightly or narrow the keyword match types.
Performance Max is spending 40% of budget with no visibility into where. This is common and it's a structural issue, not a bug. Split search-intent budget into a separate standard Search campaign so you retain visibility over your highest-value queries.
Smart Bidding won't stabilise after 60 days. You likely don't have enough conversion volume for the algorithm to learn from. Consider switching to manual or enhanced CPC bidding until you cross 30 conversions per month.
Tools and resources
- Google Ads' own Keyword Planner for realistic CPC ranges by South African region
- Google Analytics 4 for verifying your actual site conversion rate before you build the budget model
- A simple spreadsheet with the five inputs above (AOV, margin, CPA ceiling, conversion rate, sales capacity) beats any budget calculator tool
- If your account structure is already messy, a digital marketing agency for SMEs can audit conversion tracking before you commit new spend to a broken foundation
What to do next
Once your budget model is set, the next leak to check is account structure. Bad campaign segmentation feeds mixed signals to Smart Bidding regardless of how well you've sized the budget. If you're based around Gauteng and want a second set of eyes on structure specifically, Google Ads management for Johannesburg businesses is worth a look before you scale spend further.
FAQ
What's a reasonable Google Ads budget for an SME in South Africa in 2026?
Most SMEs start viably between R6,000 and R20,000 per month, depending on category CPC and sales capacity. There's no universal number, your budget should come from your CPA ceiling and conversion rate, not a rule of thumb.
How much should I spend on Google Ads as a percentage of revenue?
A common benchmark is 5-12% of revenue for growth-stage SMEs, though this varies with margin. Businesses with thin margins should anchor to CPA ceiling instead of a revenue percentage.
Is R5,000 a month enough to run Google Ads?
It can work for low-competition local service keywords with CPCs under R15, but it's tight for competitive B2B or legal terms where CPC exceeds R50. Check your category's average CPC before committing to a figure this low.
How long before a Google Ads budget starts producing results?
Smart Bidding typically needs 30-60 days and at least 15-30 conversions to stabilise. Expect the first two to four weeks to be a learning phase, not a performance phase.
Should I increase budget if cost per lead looks high in month one?
No. Increasing budget before the algorithm has enough data usually makes cost per lead worse, not better. Fix targeting and tracking first, then scale once CPA is stable near your ceiling.
What's the difference between daily budget and monthly budget in Google Ads?
Daily budget is what you set in the platform and Google can spend up to roughly 2x that on any single day, evening out over the month. Your monthly budget is the daily figure multiplied by roughly 30.4, the average days per month Google uses for billing.
Does Performance Max need a different budget approach than Search?
Yes. Performance Max needs more historical conversion data to bid well and tends to consume budget faster with less visibility, so most SMEs should feed it a smaller share until Search campaigns prove the CPA model works.
Can I run Google Ads and Meta Ads on the same budget model?
The CPA ceiling and margin math carries across both platforms, but conversion rates and CPCs differ enough that you should size each channel's budget separately rather than splitting one combined figure.
One last thing
The single biggest budget mistake SME owners make isn't spending too much, it's spending consistently on a campaign that's never had its tracking verified. Run a test conversion through your form, your call tracking number, and your checkout flow this week, before you touch the budget model above. A perfectly sized budget feeding false conversion data to Smart Bidding still burns cash, just more efficiently.
