Meta Ads terms, explained.

Meta runs advertising on Facebook and Instagram, and its reports use a lot of its own vocabulary. This page explains each term simply, and why it matters to your business.

01. Accounts and assets

Business portfolio.

A business portfolio is the container that holds everything your business owns on Meta: ad accounts, Pages, Instagram accounts, datasets and the people who can use them. Meta's help pages now call it a business portfolio, and you may still hear people use the older name, Business Manager.

It matters because whoever owns the portfolio owns the assets inside it. If your ads were set up inside someone else's portfolio, such as a previous agency's, you may find you cannot see your own history or take it with you.

For example, a garden center has its Page and ad account inside its own business portfolio and gives us access as a partner. If the relationship ended, access would be removed and the center would keep everything. Had we set it up in our own portfolio, the center would lose its ad history the day we parted.

Ad account.

The ad account is where your campaigns live, where Meta records your spend and results, and where your payment method and currency are held. A business can have more than one.

Its currency and time zone are set when it is created, and changing either means opening a new ad account, so they affect how your figures line up with your own accounts.

For example, a guest house with an ad account set to a different currency from the one it banks in will see spend reported in the ad account's currency, and the monthly figure will differ from the amount on the card statement because of the exchange rate.

Facebook Page and Instagram account.

Your ads run in the name of your Facebook Page and, if you connect one, your Instagram account. Those are what people see as the advertiser, and they are also where comments and messages arrive.

A Page with an empty history or an unmonitored inbox makes ads less convincing. Somebody needs to answer the comments on your ads.

For example, if you are a physiotherapist and your ad runs under a Page with no recent posts and three unanswered questions in the comments, a would-be patient may well decide you have closed.

Meta Pixel.

The Meta Pixel is a small piece of code on your website that tells Meta when someone does something, such as viewing a product, starting a checkout or sending an inquiry form. It is what lets Meta count results that happen on your website.

It is now managed inside a dataset, but people still call the browser-based code the pixel. Used alone it can miss events, which is one reason it is paired with the Conversions API.

For example, if you are a florist with a pixel installed, every completed order on your website sends Meta a purchase event. If the checkout is rebuilt and the pixel is not put back on the new thank-you page, Meta sees no sales while your till keeps ringing.

Dataset.

A dataset is the single place where Meta collects the events sent from your website, app, shop or customer records. The pixel and the Conversions API both feed into it, and it is where you check that events are arriving.

It matters because Meta can only learn from, and report on, what reaches the dataset. A dataset that is missing events or receiving duplicates makes every figure in your report less reliable.

For example, a dentist's dataset shows inquiry events arriving from the pixel every day but none from the server connection. That tells us half the tracking is not working, even though the ads are still running.

02. Campaign structure and objectives

Campaign.

A campaign is the top level of your ads and carries one objective, which is the single outcome you want, such as sales or inquiries. Everything underneath it works towards that objective.

Meta reports totals at campaign level, so mixing different goals in one campaign makes the figures hard to read.

For example, a bridal boutique runs one campaign to book showroom appointments and a separate one to promote a sample sale. Keeping them apart means the appointments figure is not muddied by sale clicks.

Ad set.

An ad set sits inside a campaign and holds the choices about who sees the ads, where they appear, when they run and, depending on the setup, the budget. Each ad set has its own delivery.

If you want to compare two audiences or two budgets fairly, they go in separate ad sets, and only that one thing differs between them.

For example, a driving school tests two cities in two ad sets with the same ads and the same daily budget. Because nothing else differs, the gap in inquiries tells it something about the places.

Ad.

The ad is the thing a person actually sees: the image or video, the words, the headline and the button. One ad set can hold several ads, and Meta shifts delivery towards the ones getting more response.

Your report usually breaks results down by ad, because the creative is the part that most often decides whether the campaign works.

Campaign objective.

The objective tells Meta what to try to get for you. There are six: awareness, traffic, engagement, leads, app promotion and sales. Meta then shows your ads to the people it thinks are most likely to deliver that outcome.

Choosing the wrong one is a common and costly mistake. A traffic objective finds people who click, which is not the same as people who buy.

For example, an electrician who picks traffic gets plenty of cheap visits from people who tap everything, and few phone calls. Switching to a leads objective asks Meta to find people who submit the form instead.

Optimization event.

The optimization event, which Meta calls the conversion event, is the specific action an ad set is trying to get, such as a purchase or a submitted lead. Meta studies who did it before and looks for more people like them.

It needs enough events to learn from. If the chosen event is rare, Meta has too little to go on, and a sensible nearby event is sometimes used instead.

For example, a furniture shop gets about four purchases a week online but forty add-to-baskets. Optimizing for purchases may leave Meta short of signal, while optimizing for add-to-basket gets more data but attracts more browsers.

Special ad categories.

Some topics carry extra rules because of the risk of unfair treatment: housing, credit, employment and social issues or politics. In some regions financial products and services are included. Ads in these categories have fewer targeting options, and you must declare the category when setting up.

Failing to declare one can get ads rejected or the account restricted, so we check before anything is launched.

A/B test.

An A/B test is Meta's built-in way to compare two versions of something fairly. It splits the audience so that nobody sees both, and only one thing differs between them.

It is more reliable than eyeballing two ad sets next to each other, because those can be shown to different kinds of people and their results end up incomparable.

For example, a cafe tests a photo of its brunch against a short video of the kitchen, same words, same budget, same audience. After two weeks and enough results, the one with the lower cost per inquiry is the better creative for that purpose.

03. Audiences and targeting

Advantage+ audience.

Advantage+ audience lets Meta decide who sees your ads, using the ad itself and your tracking data to find likely buyers. Details you add as an audience suggestion, such as age, interests or a lookalike, are a starting point that Meta tries first before searching more widely, while a few controls, such as locations and a minimum age, stay firm.

It tends to work best with plenty of conversions and a range of creative. We switch it off only for a reason, such as a legal restriction on who may be shown the ads.

For example, a skincare brand with two hundred orders a month lets Meta find buyers, and discovers the best customers are women in their forties, an age group the owner had not targeted.

Broad targeting.

Broad targeting means giving Meta little or no restriction on who sees an ad, other than basics like country. It relies on the creative and your tracking to attract the right people.

It can feel uncomfortable, because you are no longer choosing the audience yourself. On most accounts today it is a sensible starting point, with narrower audiences added only where there is a clear reason.

Custom audience.

A custom audience is a group made from people you already know: your customer list, website visitors, or people who have engaged with your Page or videos. Meta matches them to accounts on its platforms.

It is mostly used to exclude existing customers from prospecting ads, and to remind recent visitors about something they looked at. Using a customer list requires a lawful basis for the data.

For example, a gym uploads its member list and excludes it from the joining-offer ad. Without that, a quarter of the spend could end up showing a discount to people who already pay full price.

Lookalike audience.

A lookalike audience is a group of people Meta judges to be similar to a source group, such as your best customers.

With Advantage+ audience, Meta lets you add a lookalike as an audience suggestion. Meta then shows your ads to people matching it first, and to other people where that is likely to improve results, so it guides delivery without fencing it in. Lookalikes are most helpful when the source list is large and made up of your best customers.

Audience exclusions.

An exclusion tells Meta not to show an ad to a particular group, usually existing customers, staff or recent converters. It stops you paying to reach people who will not act.

Where Advantage+ audience is on, only custom audience exclusions are normally available, which is why we keep a clean customer list.

For example, an online tutor excludes everyone who has already booked a lesson. The booking ad then reaches only people who have not yet signed up.

Retargeting.

Retargeting means showing ads to people who have already visited your website, watched your video or engaged with your Page. They have shown some interest, so they cost less to convert than strangers.

The risk is overdoing it. People who see the same reminder many times can get annoyed, which is where the frequency figure helps.

For example, a bathroom fitter shows a reminder about a free survey to people who viewed the quote page in the last 14 days but did not submit the form.

04. Creative and placements

Creative.

The creative is the image, video and words of an ad. On Meta today it does most of the work, because the system finds the audience by itself and the creative is what decides whether anyone stops and responds.

This is why we plan creative tests carefully and refresh ads before they wear out.

Creative testing.

Creative testing means running several versions of an ad against each other to learn what works. A fair test changes one thing at a time: the idea, the opening few seconds, the format or the offer.

It needs enough budget for each version to collect a meaningful number of results before anyone picks a winner. Declaring a winner after two days and a handful of clicks is guessing.

For example, a roofing firm tests three ads that differ only in the opening line, one asking about leaks, one showing a before and after, and one quoting a local review. After three weeks and a few dozen inquiries each, one clearly has the lowest cost per inquiry.

Placements.

Placements are the places your ad can appear: the Facebook and Instagram feeds, Stories, Reels, search results and others across Meta's apps. An ad usually needs different sizes to look right in each.

Meta lets you choose them yourself or let the system choose, which is the default and is called Advantage+ placements.

Advantage+ placements.

With Advantage+ placements Meta shows the ad wherever it expects the best result for the money, across all its apps. It generally gives the system more room to find cheaper results.

We might switch it off for a reason such as brand rules about where an ad can be seen, but otherwise leave it on.

Advantage+ creative.

Advantage+ creative lets Meta make small automatic changes to your ads, such as adjusting brightness, adding music or trying different text. It can be switched on or off for each campaign.

If your brand has fixed fonts, colors or wording rules, or the ad carries a legal statement, automatic edits can be unwelcome, so we check what is turned on.

Creative fatigue.

Creative fatigue is when an ad stops working because the same people have seen it too often. It shows up as frequency rising while click-through rate falls and cost per result climbs.

No single number proves it. We watch the three moving together over several days and plan a fresh version before the results sag.

For example, a wedding photographer's best ad ran for eight weeks. Frequency went from 1.8 to 4.2, click-through fell by a third and cost per inquiry rose from $18 to $34. The ad was not bad, it was tired, and a new version with the same idea was needed.

05. Delivery, budgets and bidding

Daily and lifetime budget.

A daily budget is the average you are willing to spend each day, and Meta may spend up to a bit more on a good day and less on a quiet one. A lifetime budget is a total for a set period, which Meta spreads as it sees fit.

Over a month the daily budget should average out to what you agreed, but a single day can look higher.

For example, a daily budget of $20 can see $25 spent on a strong day, then $16 on a quiet one, and still come to about $600 over a 30 day month.

Advantage+ campaign budget.

With an Advantage+ campaign budget, you set one budget for the whole campaign and Meta moves it between ad sets towards where it sees the best results. Meta's pages describe it as formerly called campaign budget optimization.

It helps when the ad sets are similar. It can starve an ad set you wanted to keep running, so if a specific area or product needs guaranteed spend, we set budgets for each ad set instead.

Learning phase.

When a campaign starts, or is changed significantly, Meta needs time to work out where results will come from. During this period delivery is less stable and the cost of each result can swing around.

Ads Manager shows the delivery status, so we check it there instead of working to a fixed number. Making many edits at once keeps restarting the process, which is why we change one thing at a time.

For example, a bakery doubles the budget, swaps the audience and replaces the photos all on the same afternoon. Results wobble for a week, and nobody can tell which change caused it.

Bid strategy.

The bid strategy tells Meta how to spend the budget. The default aims to get the most results for the money. Other options aim for a target cost per result, a maximum bid, or a target return on spend.

Tighter limits can protect your costs but also reduce how much Meta can spend, so they suit accounts with steady volume.

06. Measuring results

Results.

Results is the number of times the thing the campaign was set up to achieve happened, as counted by Meta. In one campaign it might be purchases, and in another it is leads or messages.

It means different things in different campaigns, so you cannot add results up across campaigns that have different goals.

For example, a report might say a campaign got 40 results and another got 900. If the first counts form submissions and the second counts video views, the second is not better, it is counting something far easier to get.

Cost per result.

Cost per result is the amount spent divided by the results. It is the number most often quoted, and it is only as good as the thing being counted.

A cheap result is not always a good one. We look at it alongside how many of those results turned into real customers.

For example, a campaign spends $800 and gets 40 leads, so cost per lead is $20. If only four of those leads were genuine buyers, the real cost per customer is $200.

Reach.

Reach is the number of different people who saw your ad at least once. It is not the number of times it appeared, which is the impressions figure.

It tells you how wide the ads have spread. Reach stuck while spend rises usually means the same people are seeing the ads again and again.

For example, an ad shown 30,000 times to 10,000 people has an impressions figure of 30,000 and a reach of 10,000.

Frequency.

Frequency is the average number of times each person reached has seen your ad, worked out as impressions divided by reach.

A higher figure is not automatically bad, since some repetition helps. Many practitioners start watching closely once frequency passes about three over a short window, but that is a rule of thumb rather than a Meta rule, and we read it alongside click-through rate and cost per result.

For example, an ad with 24,000 impressions and 8,000 people reached has a frequency of 3.

Cost per thousand impressions (CPM).

CPM is what it costs to show your ad a thousand times. It reflects how competitive the audience and the season are, rather than how good the ad is.

When CPM rises sharply in busy periods, such as before Christmas or Black Friday, everything gets pricier. That is not necessarily a fault in the campaign.

For example, a CPM of $10 in March might become $18 in late November, so the same $1,300 buys fewer views even though nothing about the ads changed.

Link click-through rate.

The link click-through rate is link clicks divided by impressions. It shows how often your ad made someone want to know more.

A rising figure with a falling number of inquiries can mean the ad is attractive but promises something your website does not deliver.

For example, 380 link clicks from 40,000 impressions is a rate of 0.95 per cent.

Return on ad spend (ROAS).

Return on ad spend is the sales value Meta attributes to your ads divided by what you spent. A ROAS of 4 means $4 of sales for each $1 spent.

It is revenue, not profit, so a good-looking ROAS can still lose money once your costs are taken off. It also depends on Meta having the right sales values passed through.

For example, a candle shop spends $2,000 in a month and Meta reports $8,000 of purchases, so ROAS is 4. If the products carry a 40 per cent gross margin, the gross profit is $3,200 and the ads cost $2,000, leaving a gross profit of $1,200 after advertising.

Conversions API.

The Conversions API sends results from your server to Meta directly, instead of relying only on the visitor's browser. It catches events the pixel misses, for instance when someone blocks tracking.

It is usually run alongside the pixel. The same event is sent both ways with a shared identifier so Meta can count it once.

For example, a car dealer with only the pixel loses inquiries from people using privacy browsers. Adding the Conversions API can recover some of those, and Meta's optimization has more to learn from.

Deduplication.

Deduplication is how Meta recognises that the pixel and the Conversions API have both reported the same action, and counts it once. It depends on the event name and a shared event ID matching on both.

If it is set up wrongly, every inquiry is counted twice and the cost per result looks twice as good as it is.

For example, a website reports 80 inquiries in Meta while the inbox has 41. The gap points to duplicate events.

Event Match Quality.

Event Match Quality is a score out of ten in Events Manager showing how well Meta can link an event to a real account, based on details like email, phone number, browser and device information. It applies to website events.

Higher scores usually mean more events can be used for reporting and optimization. It is a diagnostic to improve, not a guarantee of results.

For example, a score of 4 for a lead event, where emails and phone numbers are collected but never passed to Meta, can often be lifted by sending them correctly, with personal data hashed as Meta requires.

Attribution settings.

The attribution setting is the rule for how long after someone sees or clicks an ad Meta still gives it credit. Meta's Help Center lists click-through windows of one or seven days, a view-through window of one day and an engage-through window of one day, and says its default is 7-day click, 1-day view and 1-day engagement.

Meta's windows differ from those in other tools such as Google Analytics, so a report compared with Google Analytics or with your own records can show gaps even though nothing changed in the ads.

For example, a customer clicks an ad on Monday and buys the following Sunday. Within a seven-day click window Meta takes the credit, while your analytics may have credited it to a later Google search.

Meta's figures versus your own records.

Meta's report counts what Meta believes it caused. Your CRM, till or website shows what actually arrived. They will disagree, because of different attribution rules, blocked tracking, and inquiries that came in by phone.

We always say which source a number comes from, and we judge success on your own records of leads and sales.

For example, Meta reports 62 leads for the month while the sales team logged 48 genuine inquiries from social. Both can be right, because Meta also counted duplicates and people who never answered the phone.

07. Lead forms

Instant form.

An instant form is a form that opens inside Facebook or Instagram, pre-filled with the person's profile details, so they can ask for contact in a couple of taps without leaving the app.

The default type is designed to bring more leads, and because it takes so little effort the leads can have lower intent. We pair it with checks on quality.

For example, a tutoring center gets 90 leads at $8 each from an instant form, and 25 leads at $24 each from its website form. Fewer of the instant leads pick up the phone, so the two are compared on bookings, not on lead count.

Higher intent form type.

Instant forms come in types. More volume, the default, makes it easy to submit quickly on a mobile. Higher intent adds a review screen where the person checks and confirms their details before submitting, and a line saying your business may follow up. Meta also offers a rich creative type that lets you add more branding.

Meta says the extra steps in a higher intent form help you receive more intentional submissions and fewer from people who are only marginally interested. So you may get fewer leads but more serious ones, which for a business with limited capacity to follow up is often the better trade.

Website leads.

Website leads are inquiries made on your own website after someone clicks through from an ad. They take more effort for the person, so they are usually fewer and more serious.

They need your tracking to be working, because Meta must be told when the form is submitted.

For example, a solicitor sends ad traffic to a short inquiry page. Cost per lead is higher than on an instant form, but most leads are people who clearly want advice on the matter described.

Lead quality.

Lead quality is whether a lead turns out to be a real, reachable, interested person. It cannot be seen in Meta's figures, only in what happens after the lead arrives.

The way to improve it is to feed back to Meta which leads were good, and to add questions that filter out people who are not a fit.

For example, a roofing company asks one extra question about the type of property. Lead numbers fall by a fifth, but the share of leads that lead to a survey rises.

Lead status feedback.

Where your CRM allows, you can tell Meta which leads became qualified or became customers. Meta can then optimize towards people who looked like the good ones, not just towards anyone who fills in a form.

It is the single most useful thing to add to lead campaigns, and it needs someone on your side to keep the lead statuses up to date.

08. Catalog and Advantage+

Catalog.

A catalog is the list of your products in Meta, with titles, images, prices and availability. It is what product-based ads draw from.

With a catalog, the quality of the data is the main lever. Weak titles, poor images or out-of-date prices all show up in the ads.

For example, a shoe shop whose catalog lists everything as "Shoe" gives Meta nothing to match against shoppers' interests, while "Women's navy leather ankle boot, size 5 to 8" does.

Advantage+ catalog ads.

Meta now describes this as your product catalog working across all sales campaigns, not just one ad format, and says it was formerly known as Advantage+ catalog ads. You select the catalog when setting up the campaign, then choose the creative format.

Meta says the richer your product data, the better its AI can match products to people likely to buy, so what we control is catalog quality and the tracking signal feeding it.

For example, someone who looked at a blue sofa on a furniture website sees that sofa, and two similar ones, on Instagram the next day.

Advantage+ sales campaign.

An Advantage+ sales campaign is Meta's automated setup for online shops, formerly called Advantage+ shopping. It picks the audience, placements and the mix of ads with little input from us.

It does best with real conversion volume and a good range of creative. It still has settings that matter, such as whether existing customers are excluded or limited, which we check.

For example, an online pet shop with around 100 orders a month moves its main prospecting into an Advantage+ sales campaign, and keeps a small separate campaign to hold back spend on existing customers.

Automation controls.

Advantage+ settings such as audience, placements, creative and budget can each be used or switched off for a particular campaign. They are not all or nothing.

We keep them on when there is enough data and a variety of creative, and switch one off only for a proven reason such as a legal exclusion or fixed brand assets. We test one change at a time against a control.

For example, a regional builder needs spend kept inside two counties, so we set the location firmly, and leave the other Advantage+ settings on.