How to increase a Facebook ads budget without breaking results

What Meta actually says about raising a budget

Meta lists the edits that send an ad set back into learning. Changes to targeting, creative, the optimisation event or bid strategy always count; a budget change counts depending on its size:

If you increase your budget from $100 to $101, that isn't likely to cause one or more ad sets to reenter the learning phase

The same page says a change from $100 to $1,000 may. Between the two, Meta gives no number. The closest it came was a prompt Jon Loomer reported from Ads Manager in 2023: You can increase your daily budget up to $179 without reentering the learning phase. — a 79% raise, shown for one campaign (source). Not every account sees such prompts.

The only “20%” we found on Meta’s sites is in a developer guide for game ads, which suggests raising budgets by no more than 20% per day (source). The same page says daily spend can vary by up to 25%, while Meta’s help pages put it at 75% — so it is advice for one ad type, not a rule for lead ads.

Loomer, who used to recommend tiny raises, wrote in 2025: Meta’s more forgiving about budget increases now, so we don’t need to be as cautious as we once were. (source)

What happened when one business raised its budgets

We looked at every budget change on the ad sets of one online education business we work with, from March to October 2026. For each change we compared cost per lead in the 7 days before and the 7 days after, on ad sets with at least five leads in each week:

Three things stand out:

• Small steps weren’t safer. Raises of up to 20% hurt as often as raises of 21–50%. There were too few raises above 50% to say anything about them.
• Raises cost something; cuts didn’t. A raise made the following week’s cost per lead worse about two times in three, against roughly half the time for ad sets left alone or cut.
• Raising on top of a fresh edit was worst — see Learning limited for why.

This is one advertiser, and a week after a change is not a verdict on it. Treat it as a reason to stop trusting the 20% rule, not as a new rule.

Don’t raise because of one cheap week

The most common reason to raise a budget is a week that went unusually well. That week rarely repeats. In the same data, after a week at least 20% cheaper than the four weeks before it, cost per lead rose the following week about four times in five — with a raise (77–83%), and in the smaller set of ad sets nobody touched as well (7 of 9 weeks).

Meta’s own “raise your budget” recommendation works the same way. It appears for ad sets with stable delivery and a lower cost per result than similar ones — including you or your peers, meaning other advertisers (source). The percentage of extra results it shows is, in Ads Manager’s words, based on data from real advertisers who took this recommendation over 28 days — an average, not a forecast for your ad set (source).

The extra leads cost more than the average

Meta says it plainly in its bid strategy documentation, about the default Highest volume strategy: Costs can rise as you exhaust least expensive opportunities or as you increase budget. (source) The help pages add that cost per result may increase because your budget is significantly higher than usual (source).

So look at the cost of the additional leads, not the new average:

1. Before: $30 a day, $210 a week, 14 leads — $15 a lead.
2. After: $45 a day, $315 a week, 18 leads — $17.50 a lead on average.
3. The extra: $105 more bought 4 more leads — about $26 each.

If a lead is worth $40 to the business, the raise pays. If it is worth $20, the average still looks fine while every extra lead loses money. For a school or clinic in one city, the audience runs out faster, and the extra leads get expensive sooner. In our data, the first week after a raise usually brought more spend but hardly more leads; check the CRM for which ones turn into customers (lead quality).

How to raise it, step by step

1. Check the ad set is ready. It has left learning, cost per lead has been steady for a few weeks, and it spends its current budget. If it doesn’t spend what it has, more budget changes nothing — look for the reason first (not spending).
2. Decide what an extra lead may cost. From what a lead is worth to the business, not from the current average.
3. Make one change. Raise the budget and nothing else — no new ads, audience or bid strategy in the same week, or you won’t know what worked. A new budget takes effect in about 15 minutes, and Meta recommends waiting that long before touching the bid (source).
4. Expect uneven days. Meta may spend up to 75% over the daily budget on a good day and less on others; a week never goes above 7 times the daily budget (source). One day after a raise tells you nothing.
5. Wait a full week. Meta: Wait at least 7 days after any adjustment before re-evaluating. (source) It also suggests looking at at least a full week of results (source).
6. Judge the extra leads, as above, against the four weeks before. Then raise again, hold, or go back.

Small steps cost time. Getting from $30 to $100 a day in 20% steps takes seven raises: seven weeks if each one gets its week, or seven days if you raise daily and never see a full week after any of them. In the business we looked at, almost half of budget changes were followed by another significant edit of the same ad set within a week, so their effect could never be read.

Other ways to spend more

Budget scheduling for busy days

If you only need more spend for a launch, an enrolment deadline or a seasonal peak, schedule it:

Increasing your budget by scheduling a high demand period through budget scheduling won't prompt the delivery system to enter a learning phase.

It works with daily budgets only; a period lasts at least 3 hours; the scheduled budget can be up to 8 times the daily one; and when the period ends, the budget goes back by itself. Each period is set separately — Meta doesn’t support repeating ones yet (source).

Raise the ad set that works, don’t duplicate it

A copy of an ad set is a new ad set: it starts learning from zero and goes after the same people as the original. Meta enters only one of your ads in each auction, so they don’t bid against each other, but too much overlap makes results less predictable, especially when scaling your budget (source). Loomer’s practice: I don’t duplicate ad sets to scale, choosing instead to increase the budget of the ad set that’s working. (source) A new ad set makes sense for a genuinely new audience, such as another city.

The best-known forum thread on this — a $2,000-a-day campaign duplicated at $5,000 that then did much worse, viewed about 147,000 times on Shopify Community — was a duplicate, not a raise (source).

Campaign budget

With Advantage+ campaign budget, Meta moves money between the ad sets of a campaign, and that doesn’t restart learning: ad sets within the campaign won't reenter the learning phase as budget is distributed. (source) Sales and Leads campaigns get it by default, and it can be switched on or off at most once every 2 hours (source). Meta doesn’t say whether switching an existing campaign from ad set budgets to a campaign budget counts as a significant edit, so do it on its own, not in the same week as a raise.

A cost per result goal

If the extra leads get too expensive, a cost per result goal tells Meta where to stop. The trade-off, in Meta’s words: it May not spend full budget once you hit the cap. (source) Changing the bid strategy is itself a significant edit, so don’t combine it with a raise.

Limits that cap a bigger budget

• Account spending limit. When the amount is reached, all delivery stops. (source) Raise or remove it before raising budgets.
• Daily spending limit of new accounts. You can set a budget above it, but Meta won’t spend more than the limit. Our guide to spending limits and failed payments explains how it grows.
• Payment threshold. A higher budget reaches the billing threshold sooner; make sure the card can take the larger charges.

Budget advice you’ll read, and what Meta says

How Variafy helps

Ask your assistant “Raise the budget of the ad set with the cheapest leads by 30%” and Variafy:

• Shows the change before it runs: the new budget, the most Meta may spend on a single day and over 30 days. Nothing changes until you approve it.
• Tells you about learning: what Meta reports for the ad set, when it was last significantly edited, and a warning if that was less than 72 hours ago.
• Measures the result: 7 days later it compares the ad set’s spend, results and cost per result before and after the change, and says so when there are too few results to judge.
• Lets you take it back: for 14 days the change can be reversed, as a new change you approve.

Add https://www.variafy.com/mcp to your assistant. Variafy is in private beta: join the waitlist.

How we checked

Questions

Related guides

• Learning limited on Facebook ads: what it means on a small budget
• Why is my Facebook cost per lead going up — and what’s a good CPL?
• Facebook ads not spending: every cause and a 2-minute check