What's the Minimum Budget to Start Facebook Ads?

The short answer: two minimums, not one

There is a technical minimum and there is a decision-grade minimum, and confusing the two is the single most common budgeting mistake beginners make.

The technical minimum is what Meta will physically let you set. For a campaign billed by impressions, that floor is $1 per day. For more aggressive objectives designed to drive traffic or sales, Meta's recommended minimum is around $5 per day (Shopify, Nov 2025). On top of that, there is no minimum spend to open a Facebook ad account and no set-up cost, so the barrier to simply starting is functionally zero.

The decision-grade minimum is the budget that produces learnable data. Meta's delivery system needs roughly 50 optimization events in the 7 days after an ad set's last significant edit to exit the learning phase (Shopify, quoting Meta). Below that, delivery stays unstable and your cost per result bounces around. For most conversion objectives, the budget required to hit that threshold is far above $5 a day.

So when someone asks "what is the minimum budget to start Facebook ads," the honest answer is: about $1-5 a day to switch ads on, but that number is close to useless for planning. The rest of this guide explains why, and gives you a way to estimate the number that actually matters. If you want the broader question of how to size a campaign budget, how much should I budget for Facebook ads covers that separately; this guide is specifically about the literal floor and the mechanic behind it.

The hard technical floor, precisely

Let us be exact about what Meta enforces, because precision here is what stops you from over-planning or under-planning.

The confirmed floors are:

The zero-cost entry point is worth dwelling on, because it shapes beginner expectations. There is no minimum spend to open an account and no set-up fee, so nothing stops you from launching today with pocket change. That accessibility is genuinely useful for learning the interface and watching real delivery happen. The trap is mistaking access for readiness: being able to start for a dollar is not the same as being able to learn anything from a dollar, and the platform will happily spend your money at whatever budget you set, learnable or not.

A word of honesty on the edges. Whether $5/day is a strict technical minimum for every click, conversion, or low-frequency objective, versus a recommended figure, is Meta-internal and was not confirmable from a loaded official page. The reliable framing is the one above: $1/day for impression billing, and about $5/day recommended for traffic and sales. Meta's per-objective and per-currency minimums can vary, and the canonical reference is Meta's own Best Practices for Minimum Budgets page, which is worth checking against your specific account.

You will also see a popular rule that your daily budget should be at least five times your cost-per-result goal. It is widely attributed to Meta, but it could not be confirmed from a fetchable primary source, so treat it as a rule of thumb rather than a hard Meta policy. The learning-phase math below gets you to a similar place with sources you can verify.

What the floor actually buys

The quickest way to see why the technical floor misleads is to convert it into units. Using Shopify's November 2025 averages:

• At a $16.06 CPM, $5 a day buys roughly 300 impressions per day.
• At an 87-cent average CPC, $5 a day buys about 6 clicks per day.
• WordStream's 2025 median traffic CPC of $0.70 lands in the same neighbourhood, around 7 clicks per day on $5.

Three hundred impressions or half a dozen clicks a day is real activity, but it is a trickle. Now hold that against the roughly 50 optimization events per week that an ad set needs to exit learning. If each event is a purchase, and purchases convert from a small fraction of clicks, you can see the arithmetic simply does not close. You would need weeks to accumulate 50 purchases at that click volume, and by then the data window has already rolled. That gap between what $5 buys and what 50 events demands is the entire problem in one line.

Put in weekly terms, the picture is starker. A $5 daily budget is $35 a week. At Shopify's $18.75 cost per lead, that $35 buys fewer than two leads across the whole week, nowhere near the roughly 50 events an ad set needs to exit learning. Even if every single one of those $5-a-day clicks converted, the click volume itself would still fall short of 50 conversions inside the 7-day window. This is not a targeting problem you can fix with sharper audiences or better creative. It is a volume problem baked into the size of the budget itself, which is why raising the budget, changing the objective, or choosing a cheaper event are the only real levers. For a deeper look at delivery volume, the Facebook ad learning phase guide breaks down what happens inside that window.

The learning phase is the bridge between the two numbers

The learning phase is where the technical floor and the workable budget connect, so it is worth understanding as a mechanic rather than a status label.

When you launch a new ad set, or make a significant edit to an existing one, Meta resets into learning. During this window the delivery system is actively exploring: it is testing which people, placements, and times produce your chosen result, and it has not yet settled on an efficient pattern. Performance is deliberately unstable because the system is gathering signal.

The exit condition is roughly 50 optimization events within 7 days of that last significant edit (Shopify, quoting Meta). Clear it, and delivery tends to stabilize and cost per result settles. Fail to clear it, and the ad set can slip into "learning limited," a state where Meta does not have enough data to optimize and your results stay volatile indefinitely. That is the trap a tiny budget springs: it is not that your creative is weak, it is that the system never got enough events to deliver it well.

It helps to separate the two things the delivery system is doing at once. It is deciding who to show your ad to, and it is calibrating how much each result should cost. Both jobs need examples. With only a handful of events trickling in, every new conversion swings the model's estimate, which is why a barely-funded ad set can look cheap one day and expensive the next. The roughly 50-events figure is the point where enough examples have accumulated that those swings settle into a stable read. It is a data threshold, not a spending reward, and that distinction is the whole reason budget and learning are linked but not the same thing.

This is why the answer to "what is the minimum budget" cannot be a single dollar figure. The real minimum is defined by events, and the dollar cost of 50 events depends entirely on what each event costs you.

The practitioner heuristic for a workable daily floor

Here is a way to turn the events requirement into a dollar estimate. It is a widely-used practitioner heuristic, not an official Meta-published formula, and the outputs are illustrative rather than guaranteed thresholds. Used with that caveat, it is genuinely useful.

The estimate is:

Practical daily floor ≈ (your cost per result × 50) ÷ 7

You are asking: to collect 50 events in a 7-day window, how much do I need to spend per day at my cost per result? Plugging in the cited cost-per-lead figures:

Both outputs sit far above the $5 technical floor, which is exactly the point. Even at the friendlier Shopify cost per lead, the workable daily budget is in the low hundreds of dollars, roughly 25 times the recommended minimum. That is the honest order of magnitude for a conversion objective, and it is why quoting "a dollar a day" to a beginner does them a disservice.

Two things to remember. First, these numbers are for a leads objective; your own cost per result determines your own floor, and cheaper events lower it. Second, the figures are estimates built on US benchmarks, so use them to understand the scale, then recalculate with your real cost per result once you have data.

Why the benchmarks disagree, and how to use them

You will notice the two cost-per-lead numbers above differ, $18.75 versus $27.66, and that is not an error. It is a lesson in reading benchmarks.

Shopify's figures are point-in-time averages captured in November 2025: a $16.06 CPM, an 87-cent CPC, and an $18.75 cost per lead. WordStream's 2025 benchmarks are medians, deliberately chosen to control for outliers, reported in USD, and drawn from 554 US traffic campaigns and 726 US leads campaigns over April 1, 2024 to June 30, 2025. WordStream reports a $0.70 median CPC for traffic campaigns, a $1.92 median CPC for leads campaigns, a $27.66 cost per lead, and average click-through rates of 1.71% for traffic and 2.59% for leads. LocaliQ, reading the same underlying dataset, corroborates the $0.70 and $1.92 CPCs, the $27.66 cost per lead, and a 7.72% average conversion rate for leads.

That 7.72% average conversion rate from LocaliQ is a useful bridge between the click numbers and the lead numbers. If roughly 8 in every 100 clicks become leads, then reaching 50 lead events means buying on the order of 650 clicks, and at a $0.70 to $1.92 CPC that is a spend measured in hundreds of dollars, not tens. The three sources tell a consistent story from different angles: clicks are comparatively cheap, but the conversions Meta optimizes toward are not, and the space between them is where small budgets quietly stall.

Different methods, different snapshots, different mixes of advertisers. The right move is to treat these as a range and attribute each figure, not to blend them into one average that would misrepresent both. When you build your own plan, your account's real numbers override all of them. If you want to sanity-check your costs against wider data, Facebook ads benchmarks by industry is the place to start.

One honest gap worth flagging: WordStream's 2025 benchmark page did not report a CPM figure, and a "$13.48 median CPM" that circulates in search snippets is unverified. The only CPM this guide will stand behind is Shopify's $16.06 from November 2025.

How to reach 50 events without a huge budget

If the workable floor is genuinely in the low hundreds per day for a purchase objective, what do you do when you cannot spend that? You change the shape of the problem rather than brute-forcing it with money.

The most effective lever is to optimize for a higher-frequency event. A purchase is a rare, deep-funnel action, so it accumulates slowly. Add to Cart, Initiate Checkout, or Lead events happen more often, so an ad set optimized for one of them reaches the roughly 50-per-week threshold at a much lower spend. You are trading precision (Meta finds people who take the cheaper action) for the ability to exit learning at all, which is usually the right early-stage bargain. As volume and budget grow, move optimization back down the funnel.

The second lever is consolidation. If you split a small budget across five ad sets, each one is trying to reach 50 events on a fifth of the money, and none of them get there. Fewer ad sets with more budget each is how small accounts collect enough signal. The CBO vs ABO guide covers how budget distribution works across a campaign.

The third lever is restraint. Every significant edit resets the learning phase and restarts the 7-day clock. What counts as significant is broader than beginners expect: changing the budget substantially, editing targeting, swapping creative, or altering the optimization event can each restart the window. That does not mean you should never touch a live ad set, only that every change has a cost paid in lost learning, so batch your edits and give each configuration a genuine run before you judge it. Beginners often tinker daily, which means their ad sets never finish learning. Once a test is live, leave it alone long enough to accumulate events. Knowing which changes count as significant, and when to stop fiddling, is half the battle.

A realistic starting plan

Putting the pieces together, here is how to think about your first budget without pretending the $5 floor is a plan.

Start by deciding your objective and your realistic cost per result. If you genuinely do not know, use a cheaper upper-funnel event so your cost per result is low and your event volume is high. Then run the heuristic: cost per result times 50, divided by 7, gives your rough daily floor to clear learning. If that number is affordable, fund it and leave it alone. If it is not, either move to a cheaper optimization event or accept that you are running a smaller test that will not exit learning, and interpret the results with that limitation front of mind.

This is also where an honest platform view helps: keeping research, creative generation, and launch in one place, like a platform such as AdPlay.ai does, makes it easier to test cheaper creative angles without burning budget on production, so more of your spend goes to accumulating events. But no tool changes the underlying math. The learning phase needs events, and events cost what they cost.

What is confirmed and what is not

Because budgeting advice online is full of confident numbers, here is a clean line between what this guide stands behind and what it does not.

Confirmed: the $1/day impression floor and the ~$5/day traffic-and-sales recommendation; no account minimum and no set-up cost; the roughly 50 optimization events in 7 days to exit the learning phase; and the unit costs, Shopify's $16.06 CPM, 87-cent CPC, and $18.75 cost per lead, alongside WordStream's $0.70 and $1.92 CPCs, $27.66 cost per lead, and 1.71% / 2.59% CTRs from its stated US median methodology.

Not confirmed, and therefore not asserted as fact here: the exact per-objective and per-currency technical minimums beyond the $1 and $5 figures; the "daily budget must be at least 5x your cost-per-result goal" rule, which is widely attributed to Meta but not verifiable from a primary source; the specific dollar outputs of the (cost x 50) / 7 heuristic, which are illustrative; any CPM other than Shopify's $16.06; and how Meta applies its USD minimums to accounts billed in other currencies, which is why every figure here stays in USD.

The safest way to resolve the uncertain items is to check them against your own account and Meta's live help pages, rather than trusting any secondhand number, this guide's included. Your delivery data is the only benchmark that is truly yours.

The takeaway

The minimum budget to start Facebook ads is a genuinely small number: about $1-5 a day, with no account minimum and no set-up cost. But that is the answer to the wrong question. The number that decides whether your ads work is the budget that lets Meta collect roughly 50 optimization events a week and exit the learning phase, and for most conversion objectives that is tens to low-hundreds of dollars a day, not five.

Start by picking an objective and event you can actually feed, use the heuristic to estimate your real floor, consolidate your budget, and stop editing once a test is live. Do that, and you will spend the same money far more effectively than the person who proudly runs a dollar a day and wonders why nothing converts. For the next step after budgeting, daily vs lifetime budget explains how to structure the spend you have just sized, and how to run a Facebook ad walks the full launch from account to live campaign.