You've asked four agencies the same question and gotten four different numbers.
One quoted $500. Another quoted $3,000. A freelancer quoted $150. None of them explained why. The quotes differ because nobody explained what each one covers.
Here's the breakdown.
Ad spend and management fee are two different bills.
Ad spend is what you pay Meta for impressions and clicks. Management fee is what you pay a person or agency to build and run the campaign. When an agency says "$500 a month," they usually mean the fee. Your ad spend sits on top of that. Mixing the two up is why every quote you've gotten sounds like a different price for the same service.
See a real quote broken into both numbers →
What you should actually expect to pay, by business size:
- Small local business: $300–1,000/month ad spend, $150–500/month fee
- Growing ecommerce or lead-gen business: $1,000–5,000/month ad spend, $500–1,500/month fee
- Scaling brand: $5,000+/month ad spend, flat fee or percentage of spend
Start with what a customer is worth, not with a budget.
Most people pick an ad budget the way they pick a tip. A round number that feels fine. Then the results do not cover it and they blame Meta.
Work backward instead. Take what one new customer is worth to you in their first year. Decide how much of that you can afford to spend to win them. That is your ceiling for cost per customer. Divide it by your close rate and you get the most you can pay for a single lead.
Here is an example with made-up numbers, not a quote. A customer is worth $600. You are happy to spend 30% of that, so $180. You close 1 in 5 leads, so each lead can cost up to $36. If your leads cost $15, you have room to grow. If they cost $60, no budget size fixes it. Spending more just loses money faster.
Then the budget is arithmetic.
Ad spend divided by cost per lead gives you leads. $1,500 at $15 a lead is about 100 leads. At 1 in 5, that is about 20 customers. Run this on your own numbers before you pick a figure.
The management fee should buy specific work.
Ask for the list. A fee worth paying covers account setup and tracking, new ad creative on a schedule, ongoing testing, and a report you can read in five minutes. If one of those is missing, you are paying for less than you think, or paying someone else to fill the gap.
Costs the quotes leave out:
- Creative production: video, photos, and copy do not make themselves. Someone has to pay for them.
- A landing page that converts: sending paid traffic to a slow or confusing page wastes the click you just bought.
- Tracking setup: the Meta Pixel and Conversions API need to be installed correctly, or the system optimizes toward the wrong thing.
- Taxes on ad spend: in some countries Meta adds VAT or GST on top of your budget.
Cost per lead is not cost per customer.
A $10 lead who never picks up the phone costs more than a $40 lead who buys. Track both numbers. Ask whoever runs your ads to flag any week where cost per lead improves and cost per customer gets worse. Cheap leads from a loose audience and a weak offer are the most common way to look good on a report and lose money.
The single mistake that inflates cost more than anything else: boosting a post.
A boosted post has no retargeting, no lookalike audience, no funnel. Every dollar buys one cold, one-time shot at a stranger who's never heard of you. That's the real reason people say "Facebook ads don't work." They never ran ads. They ran a boost and called it a campaign.
Give a new campaign room to learn.
Meta needs data before a campaign settles. As a rule of thumb, an ad set wants about 50 results in a week to leave the learning phase. A $10 a day budget on a $40 lead produces under two results a week, so the numbers stay noisy and people quit too early.
Judge the first month on direction, not perfection. Are leads getting cheaper week over week? Are the people coming in the right people? Those two answers tell you more than any single day.
If your budget is small, narrow the test.
Splitting $300 across five campaigns gives each one too little to learn from. Pick one offer, one audience, and one clear action. Run two or three ads against each other for two weeks. Keep the winner and test the next idea against it.
Five questions to ask before you commit to any budget:
- What does one customer have to be worth for this to pay off?
- Who owns the ad account and the pixel, you or them?
- What exactly does the management fee include?
- How many new ads will be tested in the first 30 days?
- What will be reported, and how often?
Check three numbers every Friday.
Cost per lead, cost per customer, and the share of leads that are actually a good fit. Write them down every week. After four weeks you have a trend instead of a hunch, and a trend is what you need to decide the next move.
When to raise the budget.
Raise it when cost per customer is under your ceiling and has stayed there for a couple of weeks. Do it in small steps, around 20% at a time, which is a common rule of thumb. Doubling overnight can throw a stable campaign back into learning.
When to cut it.
If cost per customer sits above your ceiling for several weeks after you have tested new ads, stop adding money. Fix the offer, the creative, or the landing page first. More spend on a leaky funnel only makes the leak bigger.
"But every agency quotes something different for my industry."
That part's actually true. Here's why. Cost per lead depends on your industry more than your platform. A real estate seller lead usually runs $6–15. We got one client down to $6.80 with a structured, retargeted campaign, not a boost. A SaaS demo signup runs $30–80. A law firm inquiry runs $50–200+. Anyone who quotes you one flat number without asking your industry first is guessing, not calculating.
What actually controls your cost, no matter your budget:
Targeting precision, creative quality, and whether you're retargeting warm traffic instead of only buying cold clicks. Improve those three and your cost per result usually drops without extra spend.
CPMs rise in November and December.
As retailers bid for the same audiences, CPMs (cost per 1,000 impressions) climb every November and December. A test you start before then costs less than one started in the middle of it.
Numbers from a current client account:
One of our current clients has spent PKR 51,454.54 total and generated 630 messaging conversations: PKR 81.67 per conversation. That's what a properly structured campaign costs to run, not a guess.
Tell us your industry and target city. Get a real number, not a range →