How Much Should You Spend on Facebook Ads? A Practical Guide for Small Businesses
Facebook and Instagram ads can bring in leads for a fraction of what Google charges. But set the budget too low and Meta's system never gets enough data to work, so you end up paying more for worse results.
Here's how to set a Facebook ads budget that actually gives your campaigns a chance.
What Facebook leads cost
WordStream's 2025 benchmarks looked at US lead-generation campaigns on Meta. Across all industries, the median was $1.92 per click, a 7.72% conversion rate and $27.66 per lead.
| Industry | Avg cost per click | Avg cost per lead |
|---|---|---|
| Attorneys & legal services | $4.10 | $18.17 |
| Home & home improvement | $2.23 | $41.26 |
| Physicians & surgeons | $2.23 | $47.47 |
| Beauty & personal care | $3.06 | $51.42 |
| Health & fitness | $2.64 | $52.98 |
| Dentists & dental services | $9.78 | $76.71 |
| All industries | $1.92 | $27.66 |
Compare that with Google search ads, where the all-industry average is about $70 per lead. Facebook leads are usually cheaper. But they're often colder: Google catches people who are searching for you right now, while Facebook reaches people who weren't looking. That's why the follow-up matters so much (more on that below).
The learning phase sets your minimum budget
Every new Facebook ad set starts in a "learning phase", while Meta works out who to show your ads to. To get out of it, an ad set needs roughly 50 results (leads, purchases or whatever you optimise for) in about a week.
That gives you a simple formula:
Weekly budget per ad set ≈ 50 × your cost per lead
- At the all-industry average of $27.66: about $1,380 a week.
- For a law firm at $18.17: about $910 a week.
- For home improvement at $41.26: about $2,060 a week.
That's more than most small businesses want to spend. The good news is you don't have to hit it exactly. You just need to stop spreading a small budget too thinly.
How to make a smaller budget work
- Use fewer ad sets. One ad set with $50 a day learns faster than five ad sets with $10 a day each.
- Keep targeting broad. Meta's targeting is now good at finding the right people itself. Tight interest targeting often just makes leads more expensive.
- Test creative, not audiences. Put 3 to 5 different ads (videos, images, hooks) in the same ad set and let Meta show the winners.
- Don't change things every day. Every big edit can send the ad set back into learning. Give it at least a week.
- If leads are expensive, optimise for an earlier step. For example, optimise for landing page views first, then switch to leads once you have more data.
A sensible starting budget
For a first test, I'd suggest a monthly budget that can buy at least 20 to 30 leads at your industry's average cost:
- Law firm: 25 × $18.17 ≈ $450 a month
- Home improvement: 25 × $41.26 ≈ $1,030 a month
- Med spa or beauty: 25 × $51.42 ≈ $1,290 a month
Run it for at least four weeks before judging. The first week or two are usually the most expensive while Meta learns.
Cheap leads are only good if they turn into customers
The biggest mistake I see is judging Facebook ads on cost per lead alone. A $15 lead that never answers the phone is worse than a $45 lead that books.
- Call or message new leads within minutes, not hours. They filled in your form while scrolling, and they'll forget about you quickly.
- Add a qualifying question to your lead form, such as budget, timescale or postcode. Fewer, better leads usually win.
- Track what happens next. Count booked appointments, signed cases and sales, not just form fills.
The bottom line
Start with your industry's cost per lead, budget for enough leads to learn from, keep the setup simple and judge results on customers, not clicks. If you'd like a hand, see our Facebook ads management or book a free strategy call.
