← AIAP Tool GuidesAll programsHomeSearch
AIAP Tool Guides·Foundations of Meta Ads·19:32

Meta Ads Audience Targeting: Saved, Custom and Lookalike Audiences, Detailed Targeting and Audience Size

Harshit Vadiparti AI consultant and ex-founder ("member of Forbes Technical Council"); narrates every module as a Meta Ads Manager screen-share
▶LMS lesson
← PreviousChoosing a Meta Campaign Objective: the Six Objectives, Objective versus Metric, and Three
Lesson 4 of 14 in Foundations of Meta Ads
Next →Meta Ad Placements: Advantage+ (Automatic) versus Manual, Platforms, Feeds, Stories and Re

The short version

  1. Three families: core/saved (cold, described by you), custom (warm, already connected), lookalike (new but similar) (0:00-0:02).
  2. Saved audience build: location with radius and exclusions, age, gender, languages, then detailed targeting by demographics, interests and behaviours - e.g. job title CEO, first-person-shooter fans, iPhone users (0:02-0:11).
  3. Custom audiences come from pixel visitors, customer lists or Meta engagement; lookalikes run 1%-10%; start near 1% and build warm audiences first (0:11-0:19).

At a glance, three clicks deep

Skim here first: the closed row is the glance, open is the study card with the key points and timestamps, and the ↓ link drops to that concept's full write-up below.

01Cold, warm, new-but-similar: the three audience familiesCore (saved) audiences are described from scratch;›

Core (saved) audiences are described from scratch; custom audiences already know you; lookalikes are new people resembling a source.

Families (0:00-0:02)

Custom sources (0:12-0:14)

↓ Full write-up of this concept

02Building a saved audience: location, age, gender, detailed targetingNarrow by place, age and gender only when the offer is genuinely specific, then add a few strong interests…›

Narrow by place, age and gender only when the offer is genuinely specific, then add a few strong interests or behaviours.

Location (0:02-0:03)

Age and gender (0:04-0:06)

Detailed targeting demos (0:06-0:11)

↓ Full write-up of this concept

03Lookalikes from 1% to 10%, and sizing the audienceA lookalike clones a source audience;›

A lookalike clones a source audience; 1% is the tightest match. Build warm custom audiences first, then expand.

Slider 1-10% (0:16)

Warm first (0:17)

Size (0:18)

↓ Full write-up of this concept

The concepts in full

01

Cold, warm, new-but-similar: the three audience families

"Core is cold, custom is warm, and lookalike is new but similar."

Saved audiences are built from location, age and interests. Custom audiences come from "your sources" (website visitors via the pixel - "the heart of retargeting" - or an uploaded customer list) or "Meta sources" (people who watched your video, opened your lead form, visited your Instagram profile). Engagement audiences work even without a website.

Why it matters

Choosing the family decides how warm the message can be.

02

Building a saved audience: location, age, gender, detailed targeting

how-to

Every restriction you add makes the audience smaller.

Location takes a country, region or radius with include and exclude (a cafe only wants walk-in distance). Age and gender narrow further (25-45 for a PC or server; women for a skincare line). Detailed targeting browses demographics (education, income, life events, parents, job title), interests and behaviours. The trainer's favourite is mobile-device users - an iPhone case for "iPhone 17 Pro Max users", or Apple-device owners for a MacBook accessory. Browse first, then pick a couple of strong options rather than piling on dozens.

Do it in this order
Why it matters

Precise targeting without starving delivery.

03

Lookalikes from 1% to 10%, and sizing the audience

"The clone is only as good as the original."

Pick a source (customer list or site visitors) and Meta finds similar new people; the slider runs 1% (smaller, usually higher quality) to 10% (broader). Start near 1% and widen only for scale; on a brand-new account with no data, a lookalike is not the first move. Size rule: too narrow makes delivery expensive and unstable, too broad wastes budget - "focused enough to be relevant, large enough for Meta to optimize."

Why it matters

The standard path from first customers to scale.

This idea elsewhereappliedModule 12: broad versus over-narrow audiences

Tools referenced

ToolCoverageMomentContext
Meta Ads Managerdemonstrated
Meta Pixelexplained
Meta AImentioned
Instagrammentioned
Facebookmentioned

Action items

    Resources mentioned

    Resources
    • docLesson page on the platform

    Extraction notes

    This page was built from an auto-generated transcript, which garbles product and people's names. Those were corrected silently in everything above and logged here for transparency. The warnings flag claims that were true on the recording day but change fast.

    Transcript corrections applied

    The transcript saysThe trainer actually means
    location and adlocation and age
    macbook attachmentMacBook accessory

    True on recording day — verify before relying

    ← PreviousChoosing a Meta Campaign Objective: the Six Objectives, Objective versus Metric, and ThreeNext →Meta Ad Placements: Advantage+ (Automatic) versus Manual, Platforms, Feeds, Stories and Re