In short: launching a paid Facebook campaign requires exactly five things, and they have to be built in this dependency order: a personal Facebook account, a business Page, a Meta Business Portfolio, an ad account with a valid payment method, and a destination for the ad to send people to (a site with the pixel installed, a lead form, or a WhatsApp conversation). Everything else is recommended but not blocking. Two steps that almost every guide still lists as mandatory are no longer required: domain verification and prioritising eight conversion events. On cost, Meta's technical minimum is one dollar a day, but the number that actually decides whether your campaign can work is different: an ad set needs roughly 50 results in a week to exit the learning phase, which means a daily budget of about 7.1 times your target cost per result. Last updated: August 2026.
This page answers one question: what has to be in place before launch, and how much budget the first campaign genuinely needs in order to teach you anything. If your campaign is already running and you are looking for ongoing optimisation, the right page is the guide to managing Facebook campaigns.
The five blocking requirements: what you need before you can spend a shekel
The difference between a to-do list and a list of requirements is that a list of requirements tells you exactly what breaks if you skip a step. These five components are the only ones Ads Manager will not let you launch without, and they build in this order because each one depends on the one before it:
| # | Component | Why it blocks | What breaks without it |
|---|---|---|---|
| 1 | Personal Facebook account | Every business asset on Meta hangs off a personal identity that administers it. There is no standalone business login | You cannot create a Business Portfolio or receive permissions on an ad account |
| 2 | Facebook business Page | The Page is the identity your ad runs from, including when the ad only appears on Instagram | There is nobody to advertise as. An ad needs an advertiser name and profile image |
| 3 | Meta Business Portfolio | The container that holds the Page, the ad account, the pixel and the permissions under one ownership | Assets stay tied to a private account. An employee who leaves takes the access with them |
| 4 | Ad account with a payment method | Meta charges after ads have run, so it requires a valid payment method up front | The campaign saves as a draft and never goes live |
| 5 | A conversion destination | Every campaign objective needs somewhere for the user to land: a landing page, a lead form or a chat | Sales and leads objectives cannot be selected without a defined destination |
The order matters. The most common mistake in a self-serve setup is opening an ad account from a personal profile first, then realising a Business Portfolio is needed, then discovering that transferring ownership of an existing ad account is a separate and awkward process. If you are building from scratch today, create the Business Portfolio first and open the ad account and the pixel inside it.
Beyond those five there is a second layer of things that are strongly recommended but not blocking: a pixel installed, Conversions API connected, Business Verification (needed to raise spending limits and unlock some features), and an Instagram professional account if you want ads to run from your Instagram identity rather than from the Page.
No longer required: domain verification and the eight-event prioritisation
This is where most content on the subject is out of date. Since iOS 14.5 we all got used to the idea that a website conversion campaign needs two preliminary steps: verify the domain inside the Business Portfolio, then prioritise up to eight conversion events in the Aggregated Event Measurement tab in Events Manager. Meta has removed both steps.
| The old requirement | Status today | What it changes in practice |
|---|---|---|
| Domain verification as a precondition for configuring events | No longer required for event configuration. Still recommended | You can run a conversion campaign even when the conversion happens on a site you do not control, such as an external booking or checkout system |
| Prioritising eight conversion events per domain | Removed. Event prioritisation is not needed for web conversion optimisation | An entire configuration step disappears, and you no longer have to redo the priority order every time an event is added |
| The Aggregated Event Measurement tab in Events Manager | Removed from the interface | If a guide sends you to that tab, the guide is out of date. You will not find it |
| Selecting a conversion domain when creating the ad | No longer required | One less step in the ad creation flow |
It is important not to over-read this. The aggregated measurement protocol itself has not gone away: it still maps up to eight conversion events per domain, still reports a single event per session, still delays reporting by up to 72 hours, and still withholds demographic breakdowns on affected traffic. What changed is only that you are no longer required to configure it by hand. Domain verification remains recommended for a completely different reason: it is what gives you control over the links and content attributed to your domain.
The practical takeaway for anyone starting today: the pixel and the Conversions API matter exactly as much as they did before, but the two setup steps that used to be the bottleneck on a first launch no longer exist.
What a Facebook campaign costs: arithmetic instead of a guess
Almost every answer to this question is an arbitrary number, usually some version of "start with 50 to 100 shekels a day". That number is not derived from anything. It can be derived, and Meta itself supplies both of the inputs.
The first input: an ad set needs roughly 50 optimisation events inside a seven-day window to exit the learning phase. Until it does, performance is volatile and you cannot draw conclusions from it. The second input: each result costs money, and you either know or can estimate the cost per result you are aiming for.
From there the calculation is simple and reproducible:
Minimum daily budget per ad set = (50 results × cost per result) ÷ 7 days ≈ cost per result × 7.1
This is the number that decides whether your first campaign can learn at all. Here it is in shekels:
| Target cost per result | Results needed per week | Minimum daily budget | Minimum monthly budget |
|---|---|---|---|
| 30 ILS | 50 | 214 ILS | approx. 6,400 ILS |
| 50 ILS | 50 | 357 ILS | approx. 10,700 ILS |
| 80 ILS | 50 | 571 ILS | approx. 17,100 ILS |
| 120 ILS | 50 | 857 ILS | approx. 25,700 ILS |
| 200 ILS | 50 | 1,429 ILS | approx. 42,900 ILS |
| 350 ILS | 50 | 2,500 ILS | approx. 75,000 ILS |
These numbers surprise a lot of business owners, so it is worth being precise about what they do and do not say. They do not say you may not start with less. They say that below this level the ad set stays inside the learning phase, so you must not draw conclusions from it and must not compare audiences on the basis of it. A business starting at 100 shekels a day against an 80 shekel cost per lead will still get leads, but at a pace that never lets the system stabilise, and the volatility will look like campaign failure.
There are three legitimate ways to handle the gap when budget is limited: concentrate the whole budget in one ad set instead of splitting it across three, choose an objective with a more frequent and cheaper event (optimising for leads rather than purchases, for example), or start on a traffic objective and build audiences before moving to conversions. Working out the cost per result you can actually afford starts from your gross margin, and that is broken down in the ROI versus ROAS guide.
And the audience itself? Since Meta removed interest exclusions, getting Facebook targeting right relies mainly on custom audiences and creative that filters the audience.
Meta's official minimums, and why they mislead
Alongside the formula above, Meta publishes technical minimums. They are very low, and that is exactly why they mislead: they are the number that lets a campaign run, not the number that lets it work.
| Optimisation type | Official daily minimum | What it actually guarantees |
|---|---|---|
| Charged by impressions | 1 USD per day | That the ad will be shown. Nothing beyond that |
| Clicks and low-frequency events | 5 USD per day | That the system can buy that event type. Not that it can learn from it |
| Cost-per-result goal bid strategy | 5× the cost-per-result goal | That the system can hold the target price. Still below the learning threshold |
Look at the third row, because it explains the confusion. The 5× rule is a genuine Meta rule, and anyone who only knows that rule concludes that a 50 shekel cost per lead needs 250 shekels a day. But that rule exists to make the target price achievable, not to exit the learning phase. The learning threshold is higher: 7.1× rather than 5×. When the two conflict, the binding constraint is the higher of the pair.
The minimums are quoted in dollars because that is how Meta states them. Actual billing happens in the currency configured on the ad account, and if that currency is not your local one you take on exchange-rate exposure that is not part of the budget and does not appear in Ads Manager reports. It is worth fixing the account currency at setup, because changing it later requires opening a new ad account.
Choosing the objective: what you want versus what the system optimises for
The objective is the first decision in the setup flow and it dictates who Meta shows the ad to. It also determines which prerequisites you have to supply:
| Business goal | Campaign objective | What has to be ready |
|---|---|---|
| Phone calls or fast enquiries | Leads | An instant lead form, or a landing page with a form and the pixel |
| Online store sales | Sales | Pixel with a purchase event, ideally Conversions API server-side as well |
| Sign-ups, downloads or content views | Traffic or engagement | A working destination and reasonable load time. A conversion event is not mandatory |
| First exposure for a new brand | Awareness | Creative only. This is the objective with the lightest prerequisites |
| A direct conversation with a customer | Leads with a messaging destination | WhatsApp Business or Messenger connected to the Page |
For sales, leads and app install objectives, Meta merged the manual and Advantage+ creation paths into a single flow. For anyone starting today that means there is no longer a choice between "manual" and "automatic" at the top of the process, only a set of toggles inside one path. The full explanation of what stays under your control is in the Advantage+ section of the campaign management guide.
If the objective you picked is conversions, read the Facebook conversion campaign guide before you launch, and if you already have site traffic, Facebook remarketing is almost always the first campaign that pays for itself fastest.
Instagram, WhatsApp and Messenger: what each placement requires
A question that comes up constantly: can you advertise on Instagram without Facebook? The precise answer is that you can advertise on Instagram without advertising on Facebook, but you cannot advertise on Instagram without Meta's infrastructure.
| Placement | What is required | Note |
|---|---|---|
| A business Page | The Page is the advertiser identity in every case | |
| An Instagram professional account, or the Facebook Page used as the identity | You can run Instagram ads without a dedicated Instagram profile, with the Page acting as the identity | |
| Messenger | An active business Page | A messaging destination requires the Page to be able to receive messages |
| A linked WhatsApp Business number | WhatsApp is a full conversion destination, not just a service channel |
If you are still deciding which network suits your business before you even open Ads Manager, the platform comparison is in the paid social media management.
Special ad categories: when declaration is mandatory
If your campaign concerns credit, employment, housing, social issues or politics, you are required to declare it in the setup flow. The declaration is not a formality: it narrows the targeting options available to you, and in regulated categories an explicit declaration is also required for the automatic audience expansion mechanisms. An advertiser who skips the declaration risks having the ad, or the entire ad account, disabled, in some cases retroactively.
Five mistakes that specifically kill a first campaign
A beginner's mistakes are different from an experienced advertiser's. These five recur the most:
| The mistake | Why it feels sensible | What to do instead |
|---|---|---|
| Splitting the budget across three ad sets on day one | It sounds like an orderly test of three audiences | Concentrate everything in one ad set until it exits the learning phase, then split |
| Changing budget or audience after two days | Performance looked bad and we wanted to fix it | A material edit resets the learning phase. Give the ad set a full week |
| Running a single ad in the ad set | Focusing on the one best creative | Meta's retrieval engine needs several variants to find a match. Three to six ads |
| Optimising for purchase in a business with few sales | That is the result you actually want | Pick a more frequent event further up the funnel, such as add-to-cart or lead, until volume exists |
| An audience that is too narrow, out of fear of waste | It looks economical and precise | A narrow audience raises delivery cost. Start broad and let the algorithm narrow it |
Building the audiences themselves, including custom and lookalike audiences, is covered in the Facebook audience building guide.
Frequently asked questions
What do you need to start a paid Facebook campaign?
Five things in a fixed dependency order: a personal Facebook account, a business Page, a Meta Business Portfolio, an ad account with a payment method, and a destination for the ad to send people to. A pixel, Conversions API and Business Verification are strongly recommended but do not block the launch itself.
How much money do you need to start advertising on Facebook?
Meta's technical minimum is one dollar a day when charged by impressions and five dollars a day for click and low-frequency-event optimisation. The practical minimum is far higher: an ad set needs roughly 50 results per week to exit the learning phase, which means a daily budget of about 7.1 times your target cost per result. At a 50 shekel cost per lead that is roughly 357 shekels a day.
Do you still have to verify your domain before a conversion campaign?
No. Domain verification is no longer required for event configuration, and prioritising the eight conversion events per domain has been removed along with the Aggregated Event Measurement tab in Events Manager. Domain verification is still recommended, but for a different reason: control over the links and content attributed to your domain.
Can you advertise on Instagram without a Facebook Page?
You can run ads that appear only on Instagram, but the underlying infrastructure is Meta's: you need either an Instagram professional account or a Facebook Page acting as the advertiser identity. There is no Instagram advertising path that bypasses Meta Ads Manager.
How long before a first campaign shows results?
The learning phase is measured as 50 results within seven days, so a week is the realistic floor for a first conclusion, and two to three weeks on a smaller budget. Any material change to budget, audience or objective resets the count and restarts the learning phase.
Is it better to start with a leads campaign or a sales campaign?
It depends on event frequency, not preference. Optimisation works better on a frequent event, so a business generating few monthly sales will stabilise faster on a leads objective or an intermediate funnel event, then move to a sales objective once volume is enough to clear the 50-results-per-week threshold.
What is the difference between setting up a campaign and managing it?
Setup is a one-off: the prerequisites, the objective, and the audience, budget and creative definitions. Ongoing management is what comes after: reading the metrics, A/B testing, rotating creative and optimising weekly. That side is covered in full in the Facebook campaign management guide.
Next step
If the list above looks long, that is a good sign: it is long precisely because it separates what blocks a launch from what does not, instead of leaving you to discover the difference mid-setup. A business that puts the five requirements in place properly, picks an objective matched to its own event frequency, and allocates a budget that lets the ad set learn, goes live with a campaign it can actually draw conclusions from in week one.
Want us to build and run it for you? Here is SFB's Facebook advertising service, or simply talk to us.




