The most common question we get about Meta Ads is "how much should I spend?" The honest answer is that it depends less on what you can afford and more on what the platform needs to work at all.
That distinction matters, because a budget below the working threshold does not produce proportionally smaller results. It usually produces nothing, and then people conclude that Facebook ads do not work in Pakistan.
Why small budgets fail disproportionately
Meta's delivery system learns which people to show your ads to by observing conversions. Until it has gathered enough of them, it is guessing, and performance during that period is erratic — this is the learning phase.
The threshold is roughly 50 conversion events per ad set per week. That is not a rule you can negotiate with.
Work backwards from it. If your cost per conversion is 400 rupees, exiting the learning phase needs about 20,000 rupees a week on that ad set. If you spend 5,000 a week instead, you never accumulate enough data, the system never optimises, and your cost per result stays high and unpredictable. You are paying learning-phase prices indefinitely.
This is why halving a working budget often loses far more than half the results.
Pick a conversion event you can actually afford
The way out of that arithmetic is usually to optimise for a cheaper event, higher up the funnel.
If purchases cost 3,000 rupees each and your budget cannot produce 50 a week, optimising for purchases will not work. Optimising for "add to cart" or a lead form submission — events that happen far more often — can reach the volume threshold on a realistic budget.
You lose some precision. The system optimises for people likely to fill in a form rather than people likely to buy. But a slightly imprecise signal with enough volume beats a perfect signal with no volume, every time.
For service businesses, click-to-WhatsApp and lead form campaigns work well in Pakistan for exactly this reason. The conversion event is cheap and frequent, and qualification happens afterwards in the conversation.
Budget for a testing phase separately
New accounts need a genuine testing period before anything is optimised. Expect two to four weeks of spending that buys information rather than customers.
During testing you are learning which audiences respond, which creative angles work, which offer converts, and what your realistic cost per result is. Treat that money as research, because that is what it is.
The mistake is judging week one by return on ad spend and switching everything off. You bought data and then threw it away.
A workable structure: dedicate a defined testing budget, run it for the full period, then decide with actual numbers whether to scale, adjust or stop.
Ad spend and management fees are different things
If you are working with an agency, keep these separate in your head and in your accounting.
Ad spend goes to Meta. Management fees pay for strategy, creative, campaign build and ongoing optimisation.
Two things to insist on. First, ad spend should be billed to your own payment method on your own ad account — never routed through an agency that marks it up. Second, you should own the ad account, the pixel and the audiences, so that if the relationship ends you keep the asset you paid to build.
What to do when your budget is genuinely too small
Sometimes the honest answer is that Meta Ads is not where your money should go this quarter. Options that produce more per rupee at small scale:
Fix conversion first. If your landing page converts at 1% and could convert at 2.5%, that is a bigger gain than any budget increase — and it improves every future campaign too.
Do the local SEO work. A Google Business Profile, citation cleanup and a review habit cost time rather than media spend, and the traffic they produce does not stop when you stop paying.
Retarget only. With a small budget, showing ads exclusively to people who already visited your site or engaged with your profile is far more efficient than trying to reach new audiences. The audience is small, so the budget goes further.
Concentrate rather than spread. One campaign, one audience, one offer, with enough budget to exit the learning phase — beats four campaigns each starved of data. Almost every underperforming small account we audit is spread too thin.
A rough framework
Not prices — a way to think about the number.
- Estimate what a customer is worth to you, including repeat business.
- Decide what you can afford to pay to acquire one and still profit.
- Estimate a realistic conversion rate from click to customer. If you have no data, 1–2% is a conservative starting assumption for a service business.
- That gives you a target cost per click and per conversion.
- Check whether your monthly budget can produce roughly 50 conversion events a week at that cost.
- If it cannot, either optimise for a cheaper event higher in the funnel, or spend the money on something else this quarter.
That final step is where most of the value in this exercise sits, and it is the step most people skip.
Things worth ignoring
Cost per click on its own. A 5-rupee click that never converts is more expensive than a 40-rupee click that does.
Reach and impressions. Neither pays for anything. If a report leads with reach, ask what it produced.
Engagement on boosted posts. Boosting optimises for engagement, which is why boosted posts get likes and rarely get customers.
Anyone promising a specific return on ad spend. ROAS depends on your margins, pricing, offer and landing page. Anyone guaranteeing a number is either misleading you or planning to report on a metric that flatters them.
The part that decides most of it
Budget gets the attention, but creative is where most of the performance difference actually lives. The same budget with strong creative and weak creative produces results that differ by multiples — and creative fatigue, not bidding strategy, is what kills most campaigns that stopped working.
Plan for enough variety to test properly and enough refresh to stay ahead of fatigue. A larger budget behind tired creative buys you a faster decline.
If you want a second opinion on whether your current spend is structured to work, we audit ad accounts for free and will tell you plainly if the budget is not viable for your objective.